Direct activation of murine T cells by staphylococcal enterotoxins

CELLULAR

IMMUNOLOGY

140,267-28

1 (1992)

Direct Activation of Murine T Cells by Staphylococcal DENNIS

D. TAUB

Department of Microbiology

AND

THOMAS

Enterotoxins’

J. ROGERS~

and Immunology, Temple University School of Medicine, Philadelphia, Pennsylvania 19140

Received May &-M%?nccepted

D-;-I491

The capacity of staphylococcalenterotoxins to stimulate all T cells bearing certain TCR variable region alleles has generated a great deal of interest. This stimulation appears to involve specific binding of the toxin to class II molecules and subsequent stimulation of the T cell via the TCR V, elements. Recent studies from our laboratory have focused on the ability of staphylococcal enterotoxins to directly activate purified lymph node T cells and a panel of T cell clones and hybridomas. A T cell costimulation assaywas performed to assesscellular activation requirements and cytokine receptor expression.Activation of highly purified lymph node T cellsby staphylococcal enterotoxin B (SEB) required costimulatory signals which could be provided by IL-l, IL-2, IL4, or IL-6, whereas SEB alone demonstrated no significant proliferative response.Using a panel of TH, and TH2 cell clonesand T cell hybridomas possessingvarious responsiveand nonresponsive V, alleles, it was possible to demonstrate that SEA and SEB costimulate T cells via the TCR complex. Additionally, enterotoxin-pretreated T cells demonstrated a significant proliferative responseupon exposure to classII-bearing accessorycells, suggestingthat these toxins bind directly to T cells. Highly purified T cells cultured with both SEB and IL-1 exhibit significantly increased levels of IL2 receptor, whereas cells cultured with SEB or IL-1 alone demonstrated low levels of this receptor. These results do not exclude-an-associationofthe staphyhcoccal enterotoxins with class II molecules in a manner which results in a high avidity binding to the TCR required for transduction of the appropriate activation signals. In the absenceof class II molecules, however, these superantigens can still bind to T cells, and the activation signal is delivered in the presence of cytokines that trigger T cell growth and lymphokine production. o 1992 Academic PWS, 1~.

INTRODUCTION The stimulation of T cells from a resting to an activated state is the result of signals received by the cells from their environment. The activation of antigen-specific T cells is highly dependent on accessory cells which present processed antigenic fragments associated with MHC antigens and provide costimulatory signals for T cells (l-4). Antigen-specific activation requires the crosslinking of the TCR complex and the presence of one or more antigen-nonspecific signals. This activation sequencecan be mimicked by lectins (5-7), bacterial enterotoxins (8,9), and anti-TCR antibodies (lo12) in conjunction with APC or by a combination of PMA and calcium ionophores ( 13). It has become clear that many cytokines such as IL- 1, IL-2, IL-4, IL-6, and IL7 can, in certain circumstances, provide the secondary signals required for T cell ’ This work was supported by Grants AI23828 and 5T32 AI07 101 from the National Institutes of Health. ’ To whom correspondenceand reprint requests should be addressedat the Department of Microbiology and Immunology, Temple University School of Medicine, 3400 N. Broad St., Philadelphia, PA I9 140. 267 0008-8749192 $3.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

268

TAUB AND ROGERS

activation ( 14-20). APC-independent T cell activation has been extensively described (21-27) using lectins and anti-TCR antibodies at high concentrations to extensively crosslink surface elements or suboptimal mitogen dosesin combination with a costimulatory signal to induce T cell proliferation. The APC-independent costimulation of T cells provides a mechanism to propagate T cell activation when weak signals are generated at the level of the TCR. The staphylococcal enterotoxins (28-32), along with a Mycoplusma arthriditis-secreted protein (33), toxic shock syndrome toxin-l (TSST-1) (34), the streptococcal pyogenic toxins (35), the streptococcal M protein (36,37), and the minor lymphocytestimulating (Mls) antigens (28, 30, 38-40) possess“superantigen” properties. Superantigens share a number of intriguing characteristics, including, for example, the fact that recognition appears to be largely a function of TCR V, gene usage(28, 32). For example, recent work indicates that V,8. I-, V,6-, and VB3-expressingT cells dominate the responseto the product of Mls-la and Mls-2a, respectively (40-42). Staphylococcal enterotoxins and the AI. arthriditis-secreted superantigen have also been shown to stimulate T cells in a VP-selectivefashion (28-33, 43-52). Recent reports have shown that both Mls- la (53) and SEC2 (54) interact with V, amino acid residues within the P-pleated sheet of the TCR, away from the complementarity-determining regions of the receptor. Superantigens have also been shown to bind selectively and with moderate affinity to MHC class II molecules (55-66). Several laboratories have described specific binding of staphylococcal toxins (55, 57, 58,60,6 1) and other superantigens (62,64-66) to HLA-DR and murine Ia molecules. This binding appears to be inhibitable with anti-class II antibodies and is non-MHCrestricted. The responseof T cells to a variety of superantigens has been shown to be dependent on class II-bearing APC. Despite the requirement for class II molecules, most superantigens can be presented by allogeneic accessorycells or cells transfected with class II molecules regardlessof the isotype or allotype (67-69; unpublished observations). The lack of a strict MHC restriction suggeststhat the mode of T cell activation does not mimic typical antigen recognition and that MHC classII molecules may be required to assist in the binding of these superantigens to the TCR complex. Studies with streptococcal M protein and other superantigens have demonstrated that the requirement for APC in the responseof human T cells can be bypassedusing the costimulatory molecules IL-1 and IL-6 provided together with PMA (45, 56, 70). In the present study, we investigated the requirements for the response of purified murine T cells, and a variety of T cell clones and hybridomas, to the superantigens staphylococcal enterotoxins A and B (SEA and SEB), in the absenceof classII-bearing accessorycells. Our results show that staphylococcal enterotoxins interact directly with purified murine T cells via the TCR complex and that the requirement for APC can be bypassed by addition of certain costimulatory signals. Under theseconditions, SEA and SEB induce cytokine receptor expression, lymphokine production, and proliferation of T cells. MATERIALS AND METHODS Mice. BALB/c, C3H/HeJ, and C57BL/6 mice were obtained from the National Cancer Institute (Frederick, MD). All mice used in these studies were male between the agesof 6 and 8 weeks. Staphylococcul enterotoxins. Purified SEB was obtained from Sigma Chemical Co. (St. Louis, MO). Purified SEA was obtained from Toxin Technology, Inc., (Madison,

COSTIMULATORY

ACTIVITY

OF ENTEROTOXINS

269

WI). The staphylococcal enterotoxins (SE@were diluted in RPM1 1640 (GIBCO, Grand Island, NY) and frozen at -20°C in aliquots for use. Cytokines. Recombinant hIL-1, was obtained from Dr. John Lee (Smith, Kline & French, King of Prussia, PA), recombinant murine IL-6 was generously donated by Dr. David Hilbert (National Institutes of Health, Bethesda, MD), and recombinant murine IL-4 and IL-5 were donated by Dr. Yung-Wu Chen (Temple University School of Medicine). Recombinant human IL-2, TNF,, TNF,, and IFN-7 were obtained from Genzyme Corp. (Boston, MA). T Cell clonesand hybridomas. The THr clone A.E7 was obtained from Dr. Helen Quill (University of Pennsylvania, Philadelphia, PA), is specific for pigeon cytochrome C (PCC) in the context of I-Ek, and bears the V,3 T cell receptor allele (7 1). A subclone of the TH2 clone, D 10.G4.1, was obtained from Dr. John Cebra (University of Pennsylvania), is specific for conalbumin in the context of I-Ek, and is alloreactive to I-Ab molecules (72). This clone also bears the V&2 T cell receptor allele. The TH2 cell clone 1O-5 17 was obtained from Dr. Richard Hodes (National Cancer Institute, Bethesda, MD), is specific for keyhole limpet hemocyanin (KLH) in the context of I-Ab (73) and bears a V,8 T cell receptor allele. The T cell clone GLL8 was also obtained from Dr. Richard Hodes and is alloreactive to BlO.A(SR) cells. This clone also bears a V$ T cell receptor allele. All cloned T cells were maintained on a 2-week stimulation schedule and rested for 7-10 days prior to use. Subclones of the T cell hybridoma BDK 23.1, originally obtained from Dr. John Kappler (University of Colorado Health Science Center, Denver, CO), is specific for KLH in the context of I-Ad and bears a V,8 T cell receptor allele. The T cell hybridoma 22.Dll was obtained from Dr. Yvonne Patterson (University of Pennsylvania), is specific for PCC in the context of I-Ek, and also bears a V,8 T cell receptor allele. Assay for T cell costimulation. The axillary, inguinal, cervical, and mesenteric lymph nodes or the spleens from BALB/c mice were dissected and single-cell suspensions were prepared. These cells were either cultured directly or further purification of the T cells were prepared. Highly purified T cells were prepared by passage of single-cell suspensions over a nylon-wool column followed by serial treatments with monoclonal antibodies directed against B cell and macrophage determinants. Lymphocytes were depleted of B cells, granulocytes, and RBC by treatment with a l/20 dilution of anti-mu antiserum (Cappel, Malvern, PA) in supernatant containing the mAb J 11D (obtained from Dr. Jonathan Sprent, Scripps Clinic, La Jolla, CA). Cells were then treated using a mixture of supernatants from the MKD6 (anti-I-Ad) and 14.4.4s (anti-I-E) cell lines. After each incubation with these antibody mixtures, the cells were treated with a l/ 15 dilution of rabbit complement (Low Tax, Cedarlane, PA) for 30 min at 37°C. Cellular debris and dead cells were removed by Ficoll-Hypaque centrifugation followed by two passagesof plastic adherence. These cells were found to be >99% Thy l-positive and >98% CD3-positive, with no contaminating macrophages or B cells detectable by fluorescent flow cytometric analysis. These T cells were then cultured in 96-well flat-bottomed microtiter plates at a density of 4 X lo5 purified T cells/well in RPM1 1640 plus 10%FCS, sodium pyruvate, nonessential amino acids, 2-mercaptoethanol, L-glutamine, and gentamicin. In certain cultures, SEB or Con A were added to the cultures in the presence or absence of lymphokine or accessory cells. After 72 hr the cells were pulsed with [3H]TdR (0.5 pCi/well) for the last 4-6 hr of the culture. In T cell hybridoma assays,stimulated

TAUB AND ROGERS

270

supematants were taken at 24 hr of culture and were added to lo4 IL-2-dependent CT-20 to a final concentration of 25%. CT-20 cells were pulsed with [3H]TdR after 18 hr in culture and harvested 6 hr later. The pulsed cells were then harvested and the incorporated [3H]TdR was determined by liquid scintillation spectrometry. RESULTS

StaphylococcalEnterotoxins Induce the Proliferation of Highly Purified T Cells in the Absenceof AccessoryCells Recent data from several laboratories have demonstrated that staphylococcal enterotoxins require MHC class II elements to stimulate T cells bearing specific V, sequences(28-32). In an attempt to assessthe nature of T cell activation by staphylococcal enterotoxins, we initially wished to analyze the accessorycell and cytokine requirements for enterotoxin-induced proliferation. Highly purified lymph node T cells were cultured in the presence of IL-l, IL-2, IL-4, IL-5, IL-6, IFN-7, or TNF,, together with SEB at a final concentration of 10 pg/ml. The results show (Fig. 1) that neither SEB nor Con A alone was mitogenic for resting murine T cells, but a marked proliferative responsewas observed when T cells were cultured in the presenceof both SEB or Con A and splenic accessorycells. In addition, resting T cells cultured in the presence of SEB together with IL- 1, IL-2, IL-4, or IL-6 partially reconstructed the proliferative response after 4X hr. Lymphokine alone at these same concentrations

IL- 1 SEB + IL-2 IL-2 SEE + IL-4 IL-4 SEB + IL-5 IL-5 SEB + IL-6 IL-6 SEB + IFN-y IFN-y SEB + TNFa TNFa 0

10

30

20

CPM x lo3

40

(&SE)

FIG. 1. Costimulation of purified T cells with staphylococcal enterotoxin B in combination with various cytokines. Purified lymph node T cells (4 X lo5 cells/well) were cultured with SEB or Con A at 10 &ml in the presence or absence of IL-l (100 U/ml), IL-2 (1000 U/ml), IL-4 (1000 U/ml), IL6 (1000 U/ml), IFIVy (1000 U/ml), IL-5 (1000 U/ml), TNF, (1000 U/ml), or syngeneic accessorycells (8 X lo5 cells/well). T cell proliferation was quantitated after 48 hr by [3H]TdR incorporation as described under Materials and Methods. Each bar representsthe mean (+SEM).

COSTIMULATORY

ACTIVITY

271

OF ENTEROTOXINS

exhibited no mitogenic activity on resting T cells. Con A induced a strong proliferative response when cultured with purified T cells in combination with IL-l or with IL-6, and a much weaker response was observed with the Con A together with IL-2. This costimulatory activity was only weakly demonstrated with SEB at 1 pg/ml in combination with any lymphokine (data not shown). Further studies showed that incubation of highly purified T cells with SEB together with IGl, IL-2, IL-4, and IL-6 resulted in a substantial increase in the level of IL-2/ IL-4 production (Fig. 2). These data demonstrate that soluble SEB can directly interact with purified murine T cells to induce both proliferation and lymphokine secretion.

Murine T Cell Clonesand Hybridomas Are Activated by StaphylococcalEnterotoxins Plus Cytokine Recent studies from several laboratories have demonstrated that staphylococcal enterotoxins require MHC class II elements to stimulate T cells to proliferate and produce lymphokines. The class 1I:SE ligand formed has specificity for the V, region of the T cell receptor. For example, SEA stimulates V,l, 3, 10, 11, 17, and SEB stimulates V,3, 7, 8.1-3, 11, 17-bearing T cells in the mouse (28). Using these V, restrictions, it is possible to assessthe manner in which accessorycell-independent T cell activation occurs via enterotoxin. A panel of T cell clones and hybridomas possessingresponsive and nonresponsive V, alleles was used to determine whether the costimulatory effectsobserved with purified T cells represent a direct interaction with

T CELLS + MEDIUM LYMPH NODES + MEDIUM T CELLS + CON A LYMPH NODES + CON A T CELLS + SEE LYMPH NODES -t SEE SEE + IL-1 IL-l SEE + IL-2 IL-2 SEE + IL-4 IL-4 SEE + IL-5 IL-5 SEB + IL-6 IL-6 SEE + IFN-y IFN-7 SEB + TNFa TNFa

m I

I m 0

10

20

30

40

CPM i 1Ci3 (*SE) FIG. 2. Staphylococcal enterotoxin B stimulates IL-2 production by purified murine T cells. Purified lymph node T cells (4 X lo5 cells/well) were cultured with SEB or Con A at 10 rg/ml in the presence or absenceof IL- I (100 U/ml), IL-2 (1000 U/ml), IL-4 (1000 U/ml), IL-6 ( 1000 U/ml), IFN-7 (1000 U/ml), IL5 (1000 U/ml), TNF= (1000 U/ml), or syngeneicaccessorycells (8 X 10scells/well). After 24 hr, supematants from the cultured cells were harvested and then incubated with CTLL-2 at a final concentration of 25% for an additional 24 hr. IL-2 production was assessedas described under Materials and Methods. Each bar is expressedas the mean (&SEM).

TAUB AND ROGERS

272

the TCR or simply an accessorycell contamination within our “purified” lymph node T cell preparations. Various T cell clones and hybridomas were cultured with SEA or SEB in the presence of a variety of cytokines for 48 hr, after which the proliferative response was assessed.The THr clone A.E7, specific for PCC in the context of I-Ek and bearing the V,3 allele, demonstrated a significant proliferative response to both SEA and SEB when cultured with IL-2 and (to a lesserextent) IL-4 at 100 U/ml (Fig. 3). However, no proliferative response was observed with IL-l, IL-6, IFN-7, IL-5, TNF,, or TNF, in the presence or absence of enterotoxin. IL-2 and IL-4 alone demonstrated a small proliferative response upon addition to the resting A.E7 clone. On the other hand, the TH2 clone DlO.G4.1, specific for conalbumin in the context of IEk and bearing the V&2 TCR element, showed a significant proliferative responseto SEB but not SEA (data not shown) when cultured with IL-2 at 10 U/ml and IL-l, IL4, and IL-6 at 100 U/ml (Fig. 3). No significant proliferative response was observed when cytokine was added to the cultures alone. Several other T cell clones and hybridoma lines were examined for their ability to be costimulated by enterotoxin. The data in Table 1 show that the V&bearing TH, cell clone GLL8 was stimulated to proliferate in response to SEB in the presence of IL-2 and IL-4. No significant proliferative response was observed with SEB alone or in the presence of other cytokines. The VB8-bearing TH2 clone 1O-5 17 was also stim-

30 25 20 15 10 5 0

DlO.G4.1 25

IL-l

IL-~

IL-4

IL-6

IFNY

IL-5

TN5

TNF,

CYTOKINE ADDED FIG. 3. Staphylococcal enterotoxins directly stimulate responsive V&earing T cell clones. The T cell clones A.E7 (top) and DIO.G4.1 (bottom) (4 X 10’ cells/well) were extensively rested and cultured with SEA (m) or SEB (I@ at 10 &ml, or medium ( ) in the presence or absence of IL-1 (100 U/ml), IL-2 (10 U/ml), IL-4 (100 U/ml), IL6 (100 U/ml), IFN-y (1000 U/ml), IL-5 (1000 U/ml), TNE, (1000 U/ml), TNF, (1000 U/ml), or syngeneic accessory cells. T cell proliferation was quantitated after 48 hr by [3H]TdR incorporation as described under Materials and Methods. The response with accessorycells of DlO.G4.1 was 58,276 + 2507 cpm with SEB and 2105 + 375 with SEA. The response with accessorycells of A.E7 was 68,425 f 5698 cpm with SEB and 84,295 + 9325 with SEA. Each bar is expressedas the mean (+_SEM).

+ + -

+ + -

10-5-17

BDK 23.1

75” 204 78’ 108

2247 f 55’ 950 f 235 2525 f 15’ 314k 85

3884 + 3257 + 3216 + 2815 f

Media

IL-2

94854 -t 56343 + 72344 k 37246 +

5634’ 8397 8627’ 3575

2 1026 t 8426 k 36953 k 4030 +

4386’ 1456 1458’ 1989

IL-2/4 response (cpm + SE)

526” 1237 1245” 525

Proliferative response (cpm f SE)

16552 -c 2375’ 237 + 23 44208 f 4874’ 117f 88

5250 k 3269 f 5154 + 3126 f

IL- 1

4236b 2743 1546 4257

16321 f 3247’ 4371 + 575 12497 + 2178’ 651 + 80

89643 + 72875 + 33187 + 9864 +

IL-4

6946 f 1221 -t 10606 + 759 Ik

1276b 215 1149’ 103

5539 zi 125b 418Ok 75 7173 + 5863 3781 f 215

IL-6

675 215 1308 750

3212 + 750” 1768 + 525 8506 t 2109’ 61k 23

3768 t 4765 + 3079 t 2778 +

TNF,

‘P<0.001.

Note. The T cell clones 10-5-17 and GLL8 and hybridomas BDK 23.1 and 22.Dll (4 X IO5cells/well) were cultured with SEB at 10 &ml of medium in the presence or absenceof IL- I(100 U/ml), IL-2 (10 U/ml), IL-4 (100 U/ml), IL-6 (100 U/ml), or TNF, (1000 U/ml). T cell proliferation was assessedafter 48 hr by [3H]TdR incorporation as described under Materials and Methods. In T cell hybridoma experiments, IL-2 production was measured after 24 hr using the IL-2-dependent CT-20 line as described under Materials and Methods. Each data point is expressed as the mean (+SEM). a P < 0.05. bP
22.Dll

GLLS

SEB

T cell

Costimulation of Murine T Cell Clones and Hybridomas with Staphylococcal Enterotoxin B

TABLE 1

2 z CA

2 id g

%

?

5 ?

8 $ El 2

8 3

274

TAUB

AND

ROGERS

ulated by SE3 in the presenceof IL-2 and IL-4. Interestingly, no sign&ant proliferative response was observed in the presence or absence of enterotoxin and IL-l with the 10-5-l 7 TH2 cell clone. Several T cell hybridomas were also tested for their ability to secretelymphokine in responseto soluble enterotoxin. The VB8-bearing T cell hybridomas BDK 23.1 and 22.Dll produced small quantities of IL-2 in response to SEB alone in culture. However, these enterotoxins in combination with IL- 1, IL-2, IL-4, IL-6, and TNF, induced significantly greater IL-2 production after 24 hr. No increase in proliferation or in IL-2 production was observed with enterotoxin in the presence of IL-5, IFN-7, or TNFs (data not shown). In hybridoma experiments where IL-2 and IL-4 were utilized, titrations of supernatants generated from mixtures of SEB and IL 2 or SEB and IL-4 with IL-2/IL-4-reactive cells demonstrate approximately a lo-fold increase in titers of lymphokine compared to IL-2 or IL-4 alone in culture. These two T cell hybridoma lines were also found to be only very weakly reactive with SEA (data not shown). These results demonstrate that these enterotoxins appear to costimulate T cell clones and hybridomas in an V,restricted manner. Further studies have shown that stimulation of several T cell lines with SEB and various cytokines results in a substantial increase in the level of IL-2 production. Unit analysis of culture supematants of the A.E7 T cell clone incubated with SEB in combination with IL-2 or IL-4 revealed an increased IL-2 production compared to IL-2 (120 U/ml vs 40 U/ml) or IL-4 (80 U/ml vs 2 U/ml) alone. The cytokines IL-l, IL6, and TNF, resulted in little to no IL-2 production by the A.E7 clone in either the presence or the absenceof SEB. A further analysis of the supematants collected from the BDK 23.1 T cell hybridoma incubated with SEB and IL- 1, IL-2, IL-4, or IL-6 also demonstrated a substantial increase in IL-2 production compared to the cytokines IL1 (44 U/ml vs 8 U/ml), IL-2 (164 U/ml vs 88 U/ml), IL-4 (44 U/ml vs 10 U/ml), or IL-6 (24 U/ml vs 8 U/ml) alone. These data demonstrate that soluble SEB can directly interact with murine T cell clones and hybridomas to induce not only proliferation but also lymphokine secretion.

StaphylococcalEnterotoxin B Binds Directly to Purified T Cells and V&ompatible T Cell Clones In order to determine whether T cells can be directly bound by staphylococcal enterotoxins, purified lymph node T cells were pretreated witb SEB for 8 hr, extensively washed, and placed back in culture in the presence of cytokines or accessory cells. The results in Fig. 4 demonstrate that SEB-pulsed T cells exhibit a significant proliferative responsewhen accessorycells were added into the culture. Similar proliferative responseswere observed when IL- 1, IL-2, IL-4, and IL-6 were placed into culture with SEB-pulsed T cells. SEB-pulsed T cells alone or in the presence of IG5, TNF,, IFNy, or purified T cells demonstrated little to no mitogenic activity. T cells which were not pretreated with SEB exhibited no proliferative response. Similar pretreatment studies were performed with T cell clones and hybridomas. The DlO.G4.1 and A.E7 T cell clones and the BDK 23.1 T cell hybridoma were pretreated with SEB for 4 hr at 37°C. As above, these cells were extensively washed and placed into culture with or without purified accessorycells. The results demonstrate that the SEB-pulsed D 10.G4.1 and A.E7 T clones exhibited significant proliferative activity when untreated accessorycells were added to the culture (Table 2, top). The SEB-pulsed BDK 23.1 T cell hybridoma also exhibited significant IL-2 production

COSTIMULATORY

ACTIVITY

OF ENTEROTOXINS

275

LYMPH NODES + MEDIUM LYMPH NODES + SEB T CELLS + SEB SEB (8h) + IL-1 MEDIUM (8h) + IL-l SEB (8h) + IL-2 MEDIUM (8h) + IL-2 1 SEB (8h) + IL-4 m MEDIUM (8h) + IL-4 1 SEB (8h)

+ IL-5

MEDIUM (8h)

+ IL-5

1

SEB (8h)

+ IL-6

1

MEDIUM (8h)

+ IL-8

w

+ IFN-?,

)

+ IFN--y

1

SEE (8h) MEDIUM

(8h)

SEB (8h) MEDIUM

(8h)

+ TNFa + TNFa 0

10

20

30

40

CPM x 1ti3 (&SE) FIG. 4. Pretreatment of purified T cells with SEB induces cytokine responsiveness.Purified lymph node T cells (4 X 10’ cells/well) were cultured in the presence or absence of SEB at 10 &ml for 8 hr. After incubation, T cells were harvested and extensively washed with serum-containing medium. These SEBpulsed T cells were then recultured with IL- 1 ( 100 U/ml), IL-2 ( 1000 U/ml), IL-4 ( 1000 U/ml), IL-6 ( 1000 U/ml), IFNy (1000 U/ml). IL5 (1000 U/ml), TNF$ (1000 U/ml), TNF, (1000 U/ml), or syngeneicaccessory cells for an additional 48 hr. T cell proliferation was assessedafter 48 hr by [3H]TdR incorporation as described under Material and Methods. Each bar is expressedas the mean (kSEM).

when cultured with class II-positive accessorycells. The pattern of T cell activation by this enterotoxin pretreatment also appears to be largely V+estricted. These SEbound T cells appear to be activated to become more responsiveto exogenouscytokine signals provided by the accessorycells. Additional pretreatment studies were performed using Thy 1-depleted splenocytes as an enriched source of class II-bearing accessorycells. These enriched accessorycell populations were treated with various concentrations of SEB for 1 hr at 37°C and were then extensively washed. These toxin-pulsed accessorycells were combined with nonpulsed T cell clones or hybridomas and T cell activation was assessed.The results (Table 2, bottom) demonstrate that SEB-bound B cells are capable of binding and stimulating untreated T cells in culture. It should be noted that the response using pulsed accessorycells required approximately loo-fold less SEB compared to the responseselicited by enterotoxin-pulsed T cells. Together, these results suggestthat SEs are capable of binding directly to both T cell and B cell populations. SEB Directly Induces IL-2 Receptor Expression on the Surface of Purijied Murine T Cells The costimulatory activity of these enterotoxins observedin these latter experiments suggeststhat the enterotoxin may directly induce increased cytokine receptor levels. In an effort to assessthe activation state of enterotoxin-treated cells, purified T cells

276

TAUB AND ROGERS TABLE 2 Staphylococcal Enterotoxins Bind Directly to Both T Cells and Accessory Cells Responseof pulsed T cells”

T cell

SEB

SE concentration kJm1)

+APC

-APC

+APC

-APC

100 10 1 100 10 1 100 10 I

38462 + 3556 375 19 + 4238 ND 9858 + 254 1024 f 315 2114f 58 ND ND ND

1236 k 427 436 f 59 ND 5647 + 514 2136 + 417 1452 + 75 ND ND ND

21215 zk 1548 11236 + 2459 1023 + 147 38510 f 1548 32613 f 2146 8643 + 1624 24598 + 478 25138k 811 3898 + 262

548 + 74 216 + 15 198+ 16 1247 + 213 2148 f 348 1458 f 148 5398 + 299 4111 If: 615 4352 + 409

A.E7

DlO.G4.1 BDK 23.1

SEA

T cell responseto pulsed accessorycell&’ SEB concentration h/ml)

A.E7

DlO.G4.1

10 1 0.1 0.01 0

38456 f 2145 35423 zk 1234 21432 + 3421 14321 + 1432 2314k 512

48765 f 1435 41234 f 2154 23145 zk 3214 16547 + 3214 4532 + 756

BDK 23.1 43526 + 28765 f 14325 + 7685 2 2134 +

2143 1254 2134 545 256

a T cell clones A.E7 and DlO.G4.1 and hybridoma BDK 23. I were pulsed with various concentrations ofSEA or SEB for 4 hr at 37°C. Cells were then extensively washedin 10 ml ofa serum-containing medium. T cell-depleted splenocytes were irradiated (2000 rads) and used as nonpulsed APC. Enterotoxin-pulsed T cells (4 X 10’ cells/well) and nonpulsed APC (8 X 10’ cells/well) were then combined and T cell activation was assessed.T cell proliferation was assessedafter 48 hr by [3H]TdR incorporation. In T cell hybridoma experiments, IL2/IL4 production was measuredafter 24 hr using the IL2 dependent CT-20 line as described under Materials and Methods. Each data point is expressedas the mean (&SEM). b T cell-depleted splenocytes (APC) were irradiated (2000 rads) and pulsed with various concentrations of SEB for 1 hr at 37°C. SEB-treated cells were then washed five times with 10 ml of a serum-containing medium. SEB-pulsed APC (8 X lo5 cells/well) and nonpulsed T cell clones or hybridomas were combined and T cell activation was assessedas described above.

were cultured in the presenceor absenceof SEB and/or IL- 1 for 4 hr. The SEB-treated cells were then examined for the level of IL-2Ra expression. The results shown in Table 3 demonstrate that SEB-treated T cells in the absenceof accessorycells express increased levels of IL-2Ra chain on their surface compared to untreated T cell controls ( 11.2%vs 4%). SEB-treated T cells cultured in the presence of IL-l exhibited an even greater increase in IL-2R expression (24.1%) compared to the IL-l-treated control (8%). These results are consistent with the studies showing accessorycell-independent induction of both lymphokine production and proliferation of T cells. DISCUSSION Staphylococcal enterotoxin-induced T cell activation typically depends on an interaction among at least three molecules: MHC class II molecules, TCR V, gene

COSTIMULATORY

ACTIVITY

277

OF ENTEROTOXINS

TABLE 3 Staphylococcal Enterotoxins Induce IL-2R Expression on Purified Murine T Cells % positive Group

Untreated

SEB (100)

SEB (10)

SEB (10) + IL-I

IL-1

Control Anti-IL-2R Anti-CD3

0.3 3.5 97.0

1.3 11.6 98.0

1.7 11.2 97.0

1.3 21.6 98.0

1.4 10.8 98.0

Note. Purified lymph node T cells were cultured in the presenceor absenceof SEB or SEB plus IL- 1 ( 100 U/ml) for 4 hr at 37°C. The SE&treated cells were washed and stained with fluorescein-labeled anti-IL-2R (7D4) or anti-CD3 (145-2Clll) antibodies. The cells were washed again and analyzed by the fluorescenceactivated cell sorter facility of the Wistar Institute (Philadelphia, PA).

elements, and the toxin itself (28-32, 43). The association between MHC class II molecules and the SEs has been demonstrated on both a functional and a physical level. Several studies have shown that T cell proliferation induced by SEsrequires the presence of class II-bearing accessorycells and is inhibited by monoclonal antibodies to MHC class II antigens (28-32, 43, 56-59). Specific and high affinity binding has been observed for both SEA and SEB to human HLA-DR and the murine I-A and IE molecules (56-6 1). The specific binding of numerous other “superantigens” to MHC class II molecules has also been recently demonstrated (28). In view of the role of MHC class II molecules in enterotoxin-induced activation, most superantigens are presented by allogeneic cells or cells transfected with class II molecules regardless of isotype or allotype (67-69; unpublished observations). In SE-induced mitogenesis, the MHC class II molecule may be required to focus and/or assist in the high avidity binding of these enterotoxins to the TCR complex. However, previous studies from several laboratories have demonstrated that the addition of PMA or PMA together with IL-l and IL-6 may partially restore the mitogenic capacity of T cells to SEsand other superantigens in the absence of accessorycells (44, 45, 56, 70). These studies suggestthat a direct interaction between T cells bearing responsive TCR elements and the SEsis possible. In the present study, the ability of SEsto activate purified murine T cells and V,responsive T cell clones and hybridomas in the presenceof a secondary cytokine signal suggeststhat these enterococcal superantigens interact directly with T cells via the TCR complex in the absence of class II-bearing accessorycells. Our examination of a possible direct T cell:SE interaction revealed that highly purified T cells could be induced to expressIG2R, secretelymphokine, and proliferate upon addition of the SEs together with a costimulatory molecule(s). This activation was found to be highly SE and cytokine concentration dependent. Concentrations of 10 &ml or above of SE were required for optimal costimulatory activity. This stringent SE concentration requirement suggeststhat a low avidity interaction between the responsive T cell populations and enterotoxin can be bypassedby a saturating quantity of mitogen. The strength ofthe secondary signal provided also affectedthe costimulatory activity of these enterotoxins. Titrations of cytokines in the presence or absence of SEsdemonstrated that 1000 U/ml of IL-2, IL-4, and IL-6 and 100 U/ml of IL-1 were required for optimal purified lymph node T cell costimulation. Costimulation experiments using responsive TH2 cell clones and T cell hybridomas required lo-fold less

278

TAUB

AND

ROGERS

IL-4 and IL-6 and loo-fold less IL-2 for optimal proliferative activity than that for the purified lymph node T cells. Studies performed with TH, cell clones showed that only IL-2 and IL-4 were capable of providing a secondary signal in the presenceof enterotoxin. Due to the constitutive IL-2 receptor expression of THr cell clones (even on resting cells) and the fastidious requirements for costimulation, it was difficult to assesswhether increased proliferation of TH, cell clones was due to an increased IL-2 production over and above the exogenous added rIL-2 or to a direct effect on cell division. Titrations of supernatants generated from TH, cells have demonstrated increased titers of IL-2 activity after costimulation with SEs and IL-2 or IL-4. Several other responsive T cell clones and hybridomas (Table 1) also demonstrated a direct activation by staphylococcal enterotoxins. This difference in responsivenessmay reflect the avidity of an SE for an ol//? TCR complex or may reflect a lower density of the TCR and/or an accessorymolecule required for SE interaction. The TCR complex is clearly involved in the response to SEs (28-32, 43, 46). This interaction has been shown to be V,-restricted and varies, depending upon the stimulating toxin (28). Our present studies show that the direct activation of normal T cells and T cell clones and hybridomas to SEA and SEB is also V,restricted. SEB costimulated clones bearing VP3 and VP8 TCR gene elements, while SEA only stimulated the VB3-bearing T cell populations. Although the number of T cell clones tested in these studies was not exhaustive, the V, allele restriction of SEA was particularly striking. This V,restricted activation of T cells by the SEsis consistent with the view that these enterotoxins costimulate directly via the TCR complex. Control blocking experiments were performed to verify this specific interaction. Although not shown, neutralization analysis using the anti-V88 mAb F23.1, in an effort to block SE-induced activation of a VB8-bearing T cells, proved inconclusive due to the stimulatory nature of this antibody alone. Similar studies were performed with anti-CD4 mAb GK1.5 and anti-class II mAbs to assessany accessory molecule requirement in SE-induced costimulation. The anti-CD4 antibody reproducibly inhibited costimulation of T cells by the SEs. However, this antibody alone modulated the control cellular response to cytokines. This probably reflects the negative signaling by anti-CD4 that has been described previously (74). Anti-class II mAbs demonstrated no inhibitory activity on direct T cell activation by SEs(data not shown). These studies suggestthat SEsinteract directly with T cells to increase cytokine receptor expression and secretelymphokine in the absenceof class II molecules. Basedon the results shown above, SEswould appear to possessan epitope(s) which binds both the class II molecule and the TCR. This model is supported by the pretreatment studies demonstrated in Table 2. In these studies, SE-pulsed classII-bearing accessorycells were capable of presenting SEsto nonpulsed T cells, and SE-pulsed T cells were activated by nonpulsed accessorycells. Although the intensity of the response to SE-pulsed T cells was less than that of SE-pulsed accessory cells, these studies suggestthat the binding of SEs to the TCR will be affected by their binding to class II molecules and visa versa. The interaction between macrophages and T cells leading to T cell activation is a critical component in the initiation of the immune response.Macrophages serve their accessory cell function by providing at least two signals to the naive T cells: they present processedantigen to the T cell antigen receptor via an MHC classII molecule, and they secretecostimulatory cytokines such as IL-1 and IL-6. In some studies the

COSTIMULATORY

ACTIVITY

OF ENTEROTOXINS

279

addition of high concentrations of IL- 1, IL-2, IL-4, IL-6, or IL-7 in the presence of a lectin or anti-TCR mAb can replacethe accessorycell requirement in T cell stimulation. Becausethe requirement for class II antigen-bearing APCs is characteristic of superantigenic responses(28-32) it was important to determine whether this requirement could be bypassed via a costimulation assay. Early studies by Carlsson et al. (56) reported that immobilized SEA alone or in combination with rIL-1 was unable to stimulate T cells to secretelymphokine or proliferate; however, SEA in combination with PMA resulted in IL-2 secretion by resting human T cells. More recent studies by Gascoigne et al. (75) have demonstrated that an isolated V, chain of the TCR is capable of binding specifically to an SEA complexed to cell-surface MHC class II molecules, whereas enterotoxin and /3 chain alone demonstrated little to no direct interaction. In contrast, Fleischer et al. (45) demonstrated that SEA in the presence of phorbol esters induced T cell proliferation, although to a much lesser extent than in the presence of class II antigen-bearing accessory cells. However, the ability of enterotoxins to mobilize Ca2+ within T cell clones in the absence of accessory cells (44) also suggestedthat they interact directly with the T cell receptor. Yagi et al. (43) have also demonstrated that certain VB-bearing T cell clones were capable of proliferating directly to SEB at high concentrations. More recent studies by Fleischer and co-workers (76) have demonstrated that human T cells could be activated in the absence of class II molecules if the toxins exfoliating toxin A or SEB were cocrosslinked on beads together with anti-CD8 or anti-CD2 antibodies. These studies along with our findings that SEB plus IL- 1 induces increased IL-2R expression on T cells demonstrate that a direct interaction between SEs and T cells can occur in the absence of class II antigen-bearing accessorycells. Our demonstration of an SE-mediated increase in cytokine receptor expression is supported by the studies of O’Hehir et al. (77). In these studies, human T cells exposed to SEs in the absence of antigen-presenting cells failed to respond to their natural ligand presented in an immunogenic form, despite an enhanced proliferation to exogenousIL-2. We hypothesize that the requirement for classII molecules is to associate with the SE in a manner that results in a high avidity binding to the TCR required to produce the appropriate activation signals. In the absence of class II elements, these superantigens can still interact with the T cell with low avidity, and the activation signal delivered in the presenceof T cell growth factors triggers lymphokine production and IL-2 receptor expression and this eventually results in cellular proliferation. The demonstration of this direct interaction between T cells and SEsmay have important implications for their use as prototypes in the study of Mls antigens and in thymic and peripheral T cell tolerance (78,79) as well as for toxin-mediated diseasein mouse and man (80). ACKNOWLEDGMENT We thank Gregory Harvey for excellent editorial assistance.

REFERENCES 1. 2. 3. 4. 5.

Schwartz, R. H., Annu. Rev. Immunol. 3,237, 1985. MacDonald, H. R., and Nabholz, M., Annu. Rev. Cell. Biol. 2, 231, 1986. Ullman, K. S., Northrop, J. P., Verweij, C. L., and Crabtree, G. R., Annu. Rev. Immunol. 8,421, 1990. Altman, A., Coggeshall, K. M., and Mustelin, T., Adv. Zmmunol. 48, 227, 1990. Habu, S., and Raff, M. C., Eur. J. Immunol. 7, 451, 1977.

280

TAUB AND ROGERS

6. Hoffmann, M. K., Chun, M., and Hirst, J. A., Lymphokine Res. 5, 1, 1986. 7. BekolT,M., Kakivchi, T., Grey, H. M., J. Zmmunol. 134, 1337, 1985. 8. Peavy, D. L., Adler, W. H., and Smith, R. T., J. Zmmunol. 105, 1453, 1970. 9. Smith, B. G., and Johnson, H. M., J. Zmmunol. 115, 575, 1975. 10. Bekoff, M., Kubo, R., and Grey, H. M., J. Zmmunol. 137, 1411, 1985. 11. Halvorsen, R., Gaudemack, G., Leivestad, T., Vartdal, F., and Thorsby, E., Stand. J. Zmmunol. 26, 197, 1987. 12. Bentin, J., Vaughan, J. H., and Tsoukas, C. D., Eur. J. Zmmunol. 18,627, 1988. 13. Hara, T., and Fu, S. M., .Z Exp. Med. 161, 641, 1985. 14. Mizel, S. B., Zmmunol. Rev. 63, 51, 1982. 15. Uyttenhove, C., Coulie, P. G., and Van Snick, J., .Z Exp. Med. 167, 1417, 1988. 16. Garman, R. D., Jacobs, K. A., Clark, S. C., and RauIet, D. H., Proc. Natl. Acad. Sci. USA 84, 7629, 1987. 17. Garman, R. D., and Raulet, D. H., J. Zmmunol. 138, 1121, 1987. 18. Morrissey, P. J., Goodwin, R. G., Nordan, R. P., Anderson, D., Grabstein, K. H., Cosman, D., Sims, J. J., Lupton, S., Acres, B., Reed, S. G., Mochizuki, D., Eisenman, J., Conlon, P. J., and Namen, A. E., .Z Exp. Med. 169, 707, 1989. 19. Chazen, G. D., Pereira, G. M. B., LeGros, G., Gillis, S., and Shevach, E. M., Proc. Natl. Acad. Sci. USA 86,5923, 1989. 20. Hu-Li, J., Shevach, E. M., Mizuguchi, J., Ohara, J., Mossmann, T., and Paul, W. E., J. Exp. Med. 165, 157, 1987. 21. Kawakami, K., Yamamoto, Y., Kakimoto, K., and Onoue, K., J. Zmmunol. 142, 1818, 1989. 22. Bank, I., and Chess,L., J. Exp. Med. 162, 1294, 1985. 23. Williams, J. M., Deloria, D., Hansen, J. A., Dinarello, C. A., Loertscher, R., Shapiro, H. M., and Strom, T. B., J. Zmmunol. 135,2249, 1985. 24. Geppert, T. D., and Lipsky, P. E., J. Zmmunol. 138, 1660, 1987. 25. Vine, J. B., Geppert, T. D., and Lipsky, P. E., J. Zmmunol. 141,2593, 1988. 26. Quintans, J., Yokoyama, A., Evavold, B., Hirsch, R., and Mayforth, R. D., J. Mol. Cell. Zmmunol. 4, 225, 1989. 27. Geppext, T. D., and Lipsky, P. E., .Z Clin. Invest. 81, 1497, 1988. 28. Marrack, P., and Kappler, J., Science 248, 705, 1990. 29. White, J., Herman, A., Pullen, A. M., Kubo, R., Kappler, J. W., and Marrack, P., Cell56, 27, 1989. 30. Janeway, C. A., Jr., Yagi, J., Conrad, P. J., Katz, M. E., Jones,B., Vroegop, S., and Buxser, S., Zmmunol. Rev. 107,61, 1989. 31. Misfeldt, M. L., Znfect.Zmmun. 58,2409, 1990. 32. Janeway, C. A., Jr., Cel163, 659, 1990. 33. Cole, B. C., Kartchner, D. R., and Wells, D. J., J. Zmmunol. 144,425, 1990. 34. Norton, S. D., Schlievert, P. M., Novick, R. P., and Jenkins, M. K., J. Zmmunol. 144,2089, 1990. 35. Imanishi, K., Igarashi, H., and Uchiyama, T., J. Zmmunol. 145, 3 170, 1990. 36. Kotb, M., Tomai, M., Majumdar, G., and Beachey, E. H., J. Cell Biol. 109,29a, 1989. 37. Tomai, M., Kotb, M., Majumdar, G., and Beachey, E. H., J. Exp. Med. 172,359, 1990. 38. Abe, R., and Hodes, R. J., Annu. Rev. Zmmunol. 7, 683, 1989. 39. Abe, R., and Hodes, R. J., Zmmunol. Rev. 107, 5, 1989. 40. Abe, R., Vacchio, M. S., Fox, B., and Hodes, R. J., Nature 335, 827, 1988. 41. MacDonald, H. R., Schneider,R., Lees,R. K., Howe, R. C., Acha-Qrbea, H., Festenstein,H., Zinkemagei, R. K., and Hengartner, H., Nature 332,40, 1988. 42. Pullen, A. M., Marrack, P., and Kappler, J. W., Nature 335, 796, 1988. 43. Yagi, J., Baron, J., Buxser, S., and Janeway, C. A., Jr., J Zmmunol. 144, 892, 1990. 44. Fleischer, B., and Schrezenmeier, H., J. Exp. Med. 167, 1697, 1988. 45. Fleischer, B., Schrezenmeier, H., and Conradt, P., Cell. Zmmunol. 120, 92, 1989. 46. Lando, P. A., Dohlsten, M., Kalland, T., Sjogren, H.-O., and Carlsson, R. O., Stand. J. Zmmunol. 31, 257, 1990. 47. Callahan, J. E., Herman, A., Kappler, J. W., and Marrack, P., J. Zmmunol. 144, 2473, 1990. 48. Takimoto, H., Yoshikai, Y., Kishihara, K., Matsuzaki, G., Kuga, H., Gtani, T., and Nomoto, K., Eur. J. Zmmunol. 20, 617, 1990. 49. Kappler, J., Kotzin, B., Gelfand, E. W., Bigler, R. D., Boylston, A., Carrel, S., Posnett, D. N., Choi, Y., and Marrack, P., Science 244, 8 11, 1989.

COSTIMULATORY

ACTIVITY

OF ENTEROTOXINS

281

50. Choi, Y., Kotzin, B., Herron, L., Callahan, J., Marrack, P., and Kappler, J., Proc. Nutl. Acad. Sci. USA 86,8941,

51. 52. 53. 54. 55.

56. 57.

58. 59. 60.

61. 62.

63. 64.

65. 66. 67. 68.

1989.

Janeway, C. A., Jr., Chalupny, J., Conrad, P. J., and Buxser, S., J. Zrnmunogenet.15, 161, 1988. Vroegop, S. M., and Buxser, S. E., Znfict. Zmmun. 57, 1816, 1989. Pullen, A. M., Wade, T., Marrack, P., and Kappler, J. W., Cell 61, 1365, 1990. Choi, Y., Herman, A., DiGiusto, D., Wade, T., Marrack, P., and Kappler, J., Nature 346,471, 1990. Buxser, S., Bonventre, P. F., and Archer, D. L., Infect. Immun. 33, 827, 1981. Carlsson, R., Fischer, H., and Sjogren, H. O., J. Immunol. 140, 2484, 1988. Fraser, J. D., Nature 339, 22 1, 1989. Mollick, J. A., Cook, R. G., and Rich, R. R., Science 24, 8 17, 1989. Fischer, H., Dohlsten, M., Lindvall, M., Sjogren, H., and Carlsson, R., J. Immunol. 142, 315 1, 1989. Russell, J. K., Pontzer, C. H., and Johnson, H. M., Biochem. Biophys. Res. Commun. 168,696, 1990. Lee, J. M., and Watts, T. H., J. Immunol. 145, 3360, 1990. Scholl, P., Diez, A., Monrad, W., Parsonnet, J., Geha, R. S., and Chatila, T., Proc. Natl. Acud. Sci. USA 86,4210, 1989. Uchiyama, T., Imanishi, K., Saito, S., Araake, M., Yan, X.-J., Fujikawa, H., Igarashi, H., Kato, H., Obata, F., Kashiwagi, N., and Inoko, H., Eur. J. Immunol. 19, 1803, 1989. Karp, D. R., Teletski, C. L., Scholl, P., Geha, R., and Long, E. O., Nature 346, 474, 1990. Lynch, D. H., Cole, B. C., Bluestone, J. A., and Hodes, R. J., Eur. J. Immunol. 16, 747, 1986. Cole, B. C., David, C. S., Lynch, D. H., and Kartchner, D. R., J. Zmmunol. 144, 420, 1990. Herrmann, T., Accolla, R. S., and MacDonald, H. R., Eur. J. Immunol. 19, 2171, 1989. Scholl, P. R., Diez, A., Karr, R., Sekaly, R. P., Trowsdale, J., and Geha, R. S., J. Immunol. 144, 226,

1990. 69. Herman, A., Croteau, G., Sekaly, R.-P., Kappler, J., and Marrack, P., J. Exp. Med. 172, 709, 1990. 70. Kotb, M., Majumdar, G., Tomai, M., and Beachey, E. H., J. Immunol. 145, 1332, 1990. 71. Hecht, T. T., Longo, D. L., and Matis, L. A., J. Immunol. 131, 1049, 1983. 72. Kaye, J., Porcelli, S., Tite, J., Jones, B., and Janeway, C. A., Jr., J. Exp. Med. 158, 836, 1983. 73. Taplits, M. S., Henkart, P. A., and Hodes, R. J., J. Zmmunol. 141, 1, 1988. 74. Bank, I., and Chess,L., J. Exp. Med. 162, 1294, 1985.

75. Gascoigne, N. R., and Ames, K. T., Proc. Nutl. Acad. Sci. USA 88,613, 1991. 76. Fleischer, B., Gerardy-Schahn, R., Metzroth, B., Carrel, S., Gerlach, D., and Kohler, W., J. Immunol. 146, 11, 1991. 77. O’Hehir, R. E., and Lamb, J. R., Proc. Nutl. Acud. Sci. USA 87, 8884, 1990. 78. Kawabe, Y., and Ochi, A., J. Exp. Med. 172, 1065, 1990. 79. Rellahan, B. L., Jones, L. A., Kruisbeek, A. M., Fry, A. M., and Matis, L. A., J. Exp. Med. 172, 109 1, 1990. 80. Marrack, P., Blackman, M., Kushnir, E., and Kappler, J., J. Exp. Med. 171,455, 1990.