INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial superantigens (SAGs) are a group of structurally related proteins which have the ability to strongly activate the immune system. The staphylococcal enterotoxins (SE) are the best characterized among the bacterial SAGs, and they have been shown to bind as unprocessed proteins to a region on the major histocompatibility (MHC) class II molecule which is distinct from the peptide binding groove (1, 30, 34). T cells bearing certain T-cell receptor (TCR) V families interact with the SAG-MHC class II complex, resulting in proliferation and the release of cytokines (1, 10, 30). A variety of techniques have been used to identify the regions of the SAGs involved in their various immunological activities (1, 5, 18, 24). However, the results are controversial. Some investigators localized the mitogenic activity of SEs to the N-terminal portion (5, 18, 24) while others blocked mitogenic function by deleting nine amino acid (aa) residues from the C-terminal region (22) or changed the V specificity by mutation of one aa residue from this region (23). Several studies have also localized MHC class II binding to the amino terminus (13, 18, 25, 26), but others have attributed MHC class II binding to the C-terminal portion of SEs (15, 27). Recently, the structures of SE type B (SEB) and toxic shock syndrome toxin and their complexes with MHC class II molecules were elucidated by X-ray crystallography and the residues involved in these interactions identified (2, 17, 19, 32). SEA is still the most potent SAG, and considerable evidence from competition studies suggests that SEA and SEB do not bind in the same way to MHC class II molecules. SEA inhibits SEB binding to MHC class II molecules, but excess SEB does not prevent SEA binding (7, 11). Furthermore, SEA requires a single zinc atom for high-affinity binding to MHC class II molecules (12), and two distinct but cooperative binding sites have been reported (1, 16). More recently, the crystal structure of SEA was described. The metal binding site was identified, and based on the similarity between SEA and SEB, the sites of interaction of SEA with TCR and MHC class II molecule were suggested (29). In order to establish a functional map of SEA, I used a simple approach based on the biological interaction of a panel of 20 specific anti-SEA monoclonal antibodies (MAbs) with SEA and the consequences of this interaction on mitogenic and binding activities of SEA. Subsequently, I studied the interaction of this panel with two large fusion proteins overlapping the SEA molecule. The first SEA-glutathione S-transferase (GST) fusion protein corresponds to the N-terminal region (aa 1 to 110) and the second corresponds to the C-terminal (aa 93 to 233) region. Furthermore, I studied the ability of the MAbs to present SEA in vitro to human and murine cells.

The results confirmed the presence of two functional binding sites of SEA on MHC class II molecules which differ by their requirement for T-cell activation. Antibodies recognizing MHC class II and TCR sites on SEA can present it to the immune cells and can start the cascade of cell activation and proliferation. All interaction sites of SEA depend only on the three-dimensional structure of SEA and not on its peptide sequence. Moreover, I confirmed the importance of the disulfide loop of SEA in TCR interaction and the presence of two different TCR-interacting sites.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Cloning and production of the N- and C-terminal regions of SEA. The pGEX-2T plasmid containing the SEA encoding gene was used as the template in a PCR (30 amplification cycles of 1 min of denaturing at 93°C, 1 min of annealing at 55°C, and an extension step of 1 min at 72°C) to produce the N- and C-terminal regions of SEA. The N-terminal region, corresponding to aa 1 to 110, was synthesized by using the 5' primer 5'-GTAAATGGATCCGAGAAAAGCGAAG3', which contained a BamHI site (underlined) placing the fragment of SEA in frame with the GST gene of pGEX-2T (Pharmacia Biotech Inc., Québec, Canada). The 3' primer was 5'-GGGAATTCTAACGTGGATCCAACCTTA3', containing an EcoRI site (underlined). For the C-terminal region (aa 93 to 223) the 5' primer used was 5'AATGGATCCGGTTATCAATGTGCGGG3', which contained a BamHI site, while the 3' primer was 5'GGGAATTCAACTTGTATATAAATATATATC3', containing an EcoRI site after the stop codon of the SEA gene. The PCR fragment was digested with BamHI and EcoRI and ligated into a BamHI/EcoRI-digested pGEX-2T plasmid. Escherichia coli DH5 was transformed with the plasmid, and transformants were picked and tested for the presence of the insert. Positive transformants were sequenced directly by the dideoxynucleotide method (28) by using the sequenase dideoxy termination sequencing kit from U. S. Biochemicals (Cleveland, Ohio). Several oligonucleotide primers that match the SEA sequence were used. Bacteria from an overnight culture were diluted (1:10) in fresh L broth medium and grown to mid-log phase (optical density at 600 nm [OD₆₀₀] = 0.8). Then isopropyl--D-thiogalactopyranoside was added, and the cultures were harvested after 3 to 5 h of incubation. Bacterial cell pellets were collected by centrifugation, resuspended in lysis buffer (50 mM Tris [pH 7.5], 50 mM NaCl, 5 mM MgCl₂, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride) and lysed by sonication at 4°C. Then the supernatants were dialyzed overnight to remove endogenous glutathione. The lysate was passed through a column of glutathione agarose (Pharmacia Biotech Inc., Québec, Ontario, Canada), which specifically binds the fusion protein (GST-SEA). After being washed, recombinant proteins were cleaved from GST by digestion with bovine thrombin (Sigma Diagnostics, Mississiauga, Ontario, Canada) overnight at 4°C according to manufacturer's instructions. Thrombin was removed by a 30-min incubation with p-aminobenzeamidine-agarose bead gels (Sigma Diagnostics) and centrifugation. Purified proteins were dialyzed against H₂O and filtered.

Anti-SEA antibodies. Anti-SEA MAbs were prepared in my laboratory according to classic methods (20, 21). MAb supernatant was used in the primary test; then the MAbs were produced as ascites, purified, and used in the following test. Rabbit polyclonal antibodies were prepared in our laboratory by the injection of rabbits with commercial SEA (Toxin Technology, Sarasota, Fla.).

ELISAs. (i) Direct enzyme-linked immunosorbent assay (ELISA). Microtiter plates coated with different recombinant proteins (1 µg/ml in 0.1 M carbonate-bicarbonate buffer, pH 9.6) were blocked with phosphate-buffered saline (PBS) containing 0.1% Tween-20 and 1% gelatin (PBS-Tween-gelatin), washed, and incubated with the MAbs at different concentrations for 1 h at 37°C. After being washed, the plates were incubated with peroxidase-labelled goat anti-mouse immunoglobulin G (Ig) (Bio-Rad) for another 1 h at 37°C. The enzyme activity was developed with 2,2-azino-di(-3 ethylbenzthiozoline 6-sulphonate) substrate, and the OD was measured.

(ii) Sandwich ELISA. Microtiter plates were coated with MAb (1 µg/ml) and then incubated with different concentrations of SEA recombinant proteins in PBS-Tween-gelatin for 1 h at 37°C. After being washed they were incubated with rabbit anti-SEA antibody, and the reaction was developed as described above with peroxidase-labelled goat anti-rabbit Ig.

Class II binding and inhibition assays. The ability of SEA to bind MHC class II molecules was assessed as previously described (31) with different MHC class II-positive cell lines (Raji and Doudi) obtained from the American Type Culture Collection (Rockville, Md.) and the HLA-DR1-transfected fibroblast cell line DAP-3. As a negative control I used the MHC class II-negative cell line RM3, derived from Raji (6) and kindly provided by R. P. Sékaly (Clinical Research Institute of Montreal, Montreal, Québec, Canada). SEA (20 µg) was iodinated as previously described (33). The binding tests were carried out as follows: 4 × 10⁵ cells were incubated with 20 ng of ¹²⁵I-labelled SEA in 200 µl of binding buffer (RPMI medium-2% fetal calf serum-0.1% NaN₃) for 1 h at 37°C. Cells were then pelleted through an oil cushion (84% silicon oil and 16% mineral oil), and their activity (counts per minute) was determined with a gamma counter. In inhibition tests, SEA was first preincubated for 1 h at 37°C with 50 µl of different MAbs, each at a concentration of 10 µg/ml. All tests were performed in triplicate, and the standard error of the mean (SEM) was less than 10% in all assays.

IL-2 assay. The ability of different MAbs to inhibit interleukin-2 (IL-2) production during SEA stimulation was tested with the murine T-cell line 3DT expressing V1 and V.8.1. A total of 8 × 10⁴ cells/well were incubated with SEA alone or were preincubated with 50 µl of MAbs (10 µg/ml) in 96-well plates for 24 h at 37°C in the presence of 2 × 10⁴ cells of the HLA-DR1-transfected fibroblast cell line DAP-3 as antigen-presenting cells (APC) or nontransfected DAP cells as negative control. Supernatants were collected and tested for their levels of IL-2 with the CTLL cell line. A total of 10⁴ cells/well of CTLL cells were cultured in 96-well plates for 24 h in the presence of 100 µl of supernatant or a standard curve of recombinant IL-2. Cells were then pulsed with 0.2 mCi of [³H]thymidine for 18 h and harvested, and the [³H]thymidine incorporated into cellular DNA was counted. Tests were conducted in triplicate, and the SEM was less than 10% in all assays.

Inhibition of cell proliferation by MAbs. The ability of the panel of MAbs to inhibit cell proliferation induced by SEA activation was assessed by culturing murine splenic cells (3.5 × 10⁵ cells/well in 96-well plates) in the presence of SEA alone or with different MAbs. The cells were cultured for 72 h and then pulsed for 18 h with 1 µCi of [³H]thymidine. Cells were harvested onto glass fiber filters, and the incorporation of [³H]thymidine was assessed. Alternatively, human peripheral blood mononuclear cells (PBMC) were isolated by the Ficoll-Hypaque method (8) from human peripheral blood obtained from healthy donors. These PBMC were incubated (10⁵ cells/well) as described above. Samples were assayed in triplicate, and the data are reported as mean counts per minute. The SEM was less than 10% in all cases.

Presentation of SEA by MAbs to human and murine cells. Microtiter plates (Nunc) were coated with rabbit anti-mouse IgG (5 µg/ml) (Bio-Rad), saturated with RPMI complete medium, washed five times with RPMI complete medium, and then incubated with different MAbs (10 µg/ml) in RPMI complete medium for 2 h at 37°C. After being washed, SEA in RPMI complete medium was added at different concentrations, and the mixtures were incubated for 2 h at 37°C. Unbound SEA was removed by extensive washing, the cells (human PBMC at 10⁵ cells/well or murine splenic cells at 2.5 × 10⁵ cells/well) were added, and the mixtures were incubated as for the activation test and then pulsed for 18 h with 1 µCi of [³H]thymidine. Cells were harvested onto glass fiber filters, and the incorporation of [³H]thymidine was assessed. All tests were performed in triplicate, and the SEM was less than 10% in all assays.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Inhibition of SEA binding on MHC class II molecules. One of the steps in the SAG cascade activation of the immune system is the binding of SAGs to MHC class II molecules found on APC. In order to determine the interaction sites of SEA with MHC class II molecules, 80 specific anti-SEA MAbs were generated. All react strongly with SEA as confirmed by ELISA and Western blotting. Of these, 20 MAbs were selected to be cloned and used in further experiments following the primary tests: (i) an ELISA on a SEA-coated plate and (ii) a SEA binding on MHC class II inhibition test on Raji cells. The characteristics of the panel of MAbs are given in Table 1. The MAbs were tested for the inhibition of ¹²⁵I-SEA binding to MHC class II molecules present on Raji, Daudi, and DR1-transfected DAP3 cell lines. The RM3 MHC class II-negative cell line was used as a negative control, and the results were compared to the inhibition obtained with cold SEA, defined as 100%. The same profile of inhibition was obtained with all three cell lines tested (Fig. 1). Thirteen MAbs (1F7, 1F9, 1H12, 2A11, 2H10, 4C12, 5H3, 7E10, 9B1, 9F12, 9H9, 11A8, and 13B12) inhibited between 75 and 100% of the SEA binding on all cell lines. Three MAbs (3B9, 3H11, and 6E7) partially inhibited (25 to 50%) the SEA binding on class II molecules. Four MAbs (4H8, 5A6, 6E3, and 8G6) had no significant effect on SEA binding.

View larger version (69K): [in this window] [in a new window]   FIG. 1.   Inhibition of SEA binding on MHC class II-positive cell lines by anti-SEA MAbs. A total of 20 ng of 125I-labelled SEA was preincubated with different MAbs for 1 h at 37°C. Then, 4 × 105 cells were added, and the mixture was incubated for the same time. Thereafter cells were pelleted and the activity of bound 125I was determined with a gamma counter. Data are means of duplicate measurements (counts per minute). The SEM was less than 10% in all assays.

Inhibition of IL-2 production. The complex formed by the association of SEA and the MHC class II molecule is able to interact with TCR and activate T cells bearing specific V to proliferate and secrete IL-2 and other cytokines. To test the ability of the panel of MAbs to inhibit IL-2 production I used the murine T-cell clone 3DT expressing V.1 and V8.1, which is stimulated normally by SEA presented by the murine fibroblasts DAP-3 transfected by human MHC class II (DR1). The results demonstrate that the interactions of anti-SEA MAbs with the SEA molecule have different effects on the production of IL-2 by the 3DT T-cell line (Fig. 2). Two MAbs (3B9 and 6E3) which have little or no effect on MHC class II binding were able to completely inhibit the production of IL-2 by this clone. Moreover, two other MAbs (5A6 and 8G6) from the same category inhibited 70% of IL-2 production. Figure 2 also shows that eight MAbs (1F7, 1F9, 1H12, 2A11, 2H10, 11A8, 3H11, and 13B12), which were good inhibitors for MHC class II binding, significantly inhibited IL-2 production by the 3DT T-cell line. The other MAbs, 4C12, 5H3, 7E10, 9F12 and 9H9, which strongly block MHC class II binding, had little or no effect on IL-2 production by this clone.

View larger version (40K): [in this window] [in a new window]   FIG. 2.   Inhibition of IL-2 production by 3DT murine T-cell line after stimulation with SEA by anti-SEA MAbs. 3DT cells (0.8 × 105/well) were cultured for 24 h in the presence of DR1-transfected cells (0.2 × 105/well) and stimulated by various concentrations of SEA with or without anti-SEA MAbs. Then, 100 ml of supernatant was collected and added to CTLL cells (104/well). Twenty-four hours later, thymidine was added and CTLL proliferation was determined by measuring [3H]thymidine incorporation after overnight incubation. The data are means (counts per minute) of triplicate measurements. The SEM was less than 10% in all assays.

Inhibition of cell proliferation. In order to generalize the results obtained with IL-2 inhibition, I tested the ability of MAbs to inhibit the proliferation of BALB/c murine splenic and human PBMC induced by SEA. Figure 3 shows that the best inhibition of cell proliferation by a MAb was obtained with BALB/c murine splenic cells, and the inhibition of cell proliferation was comparable to the inhibition of IL-2 production by the 3DT murine T-cell line. The two MAbs, 3B9 and 6E3, which inhibit completely IL-2 production, also completely inhibited BALB/c mouse cell proliferation. MAb 5A6 was also able to block 80% of cell proliferation, whereas MAb 8G6 was less efficient. MAbs inhibiting MHC class II molecule binding had different effects on cell proliferation. The majority can prevent this proliferation (1F7, 1F9, 1H12, 2A11, 2H10, 9B1, and 11A8), but as with IL-2 production, there are some MAbs which are able to strongly inhibit the binding of SEA on MHC class II molecules yet are unable to prevent cell proliferation (4C12, 5H3, 7E10, 9F12, and 9H9). The same profile of cell proliferation inhibition was seen in two human PBMC, but the inhibition was more evident with murine cells (Fig. 3).

View larger version (53K): [in this window] [in a new window]   FIG. 3.   Inhibition of murine splenic and human PBMC proliferation. BALB/C splenic cells (3.5 × 105/well) and purified human PBMC cells from donor 1 (HD1) or donor 2 (HD2) (105/well each) were stimulated for 72 h with different SEAs in the presence or absence of anti-SEA MAbs. Cellular proliferation was then determined by measuring [3H]thymidine incorporation. Data are means (counts per minute) of triplicate measurements. The SEM was less than 10% in all experiments.

View larger version (13K): [in this window] [in a new window]   FIG. 4.   Differential recognition of SEA by 3B9 and 6E3 MAbs. A SEA-coated plate was preincubated with different concentrations of either 6E3, 3B9, or 1H11 and then the mixture was incubated with biotinylated MAb 3B9. The reaction was developed with peroxidase-avidin.

Reactivity of MAbs with the two SEA fusion proteins. Previous studies using peptides or mutation or fusion proteins have localized the interaction sites of SAGs with MHC class II and TCR to the N- or C-terminal region of SEA (5, 13, 18, 22, 24-26). In the present study, I investigated whether the N- or C-terminal domain of SEA alone was involved in its immune activity. We generated in the PGEX-2T plasmid two large GST-SEA fusion proteins: the N-terminal region containing aa 1 to 110 and the C-terminal region containing aa 93 to 233. Both included the intercysteine region of SEA to confer more stability to these molecules and to yield structures more closely related to the normal three-dimensional structure of SEA. The reaction of the MAbs with the two molecules was tested by Western blotting and ELISA. In the Western blot test, all MAbs which strongly reacted with the SEA molecule did not recognize either of the two fusion proteins (results not shown). Furthermore, in the ELISA, all MAbs gave negative results except for two: 6E3, which completely inhibits IL-2 production and cell proliferation, recognized both fusion proteins as well as GST (used as negative control) (Table 1), and 4H8, which had no activity on SEA immune functions, was able to recognize the C-terminal SEA fusion protein region. To identify the epitope recognized by 6E3 and present on both fusion proteins and GST, I tested the reactivity of this MAb with two peptides corresponding to the amino acids between the cysteine residues in the disulfide loops of SEA and SEB. The results show that the SEA but not the SEB disulfide loop is recognized by this MAb and indicate that the disulfide loop of SEA represents at least a part of the epitope recognized by MAb 6E3 and is involved in cell proliferation induced by this SAG. Preincubation of 6E3 with this peptide at a concentration of 1 mg/ml completely inhibited its interaction with the peptide but was unable to significantly prevent its interaction with SEA as shown by ELISA (data not shown).

Presentation of SEA to human and murine cells by MAbs. It has been reported that only the antibodies reacting with the MHC class II binding site of SEB can present the SAG to T-cell clones (14). Because cell activation by SAG is not limited to T cells, I investigated whether the panel of MAbs was able to present SEA to human PBMC and to mouse splenic cells. Results shown in Fig. 5 indicate that independent of the SEA site recognized by the MAbs, all MAbs can present SEA with different potential to human (Fig. 5A) and murine (Fig. 5B) cells with the exemption of the 4H8 antibody. Six MAbs were selected, of which three recognized the SEA-MHC class II interaction site (1H12, 3H11, and 9B1); two inhibited IL-2 production and cell proliferation without a strong effect on MHC class II binding, suggesting that they may recognize the SEA-TCR interaction site (3B9, and 6E3); and one (4H8) had no effect. These antibodies were tested for their ability to capture and present SEA. Figure 6A shows the amount of SEA which can be captured by the MAbs. Most of the antibodies showed comparably strong binding to SEA, except for 4H8, which binds weakly. This is due to the fact that most of the antibodies recognized the native form of SEA while 4H8 could not. The small difference found between 1H11 and 6E3 in binding to SEA did not affect the ability of these antibodies to present SEA to mouse splenic cells as shown in Fig. 6B. The difference between the capacity of different MAbs to present SEA to human and mouse cells and their different specificities for the functional site of SEA suggest that the presentation by the MAbs may be different from that mediated by the MHC class II molecule. To test this possibility, a SEA mutant was used to test the activation of BALB/c splenic cells by SEA presented by the MAbs. In this mutant the amino acid residue at position 60 is naturally changed from aspartic acid to asparagine, resulting in a SEA able to activate human but not murine cells (21). Results in Fig. 7 show that only the SEA presented by MAb 6E3, which recognizes the SEA-TCR interaction site, can induce a slight proliferation of murine cells while the SEA presented by MHC class II molecules or other MAbs has no proliferation effect. This may be the result of the activation of APC, and it suggests that the MAbs can play a supplementary role in the activation of the immune system by the SAG.

View larger version (48K): [in this window] [in a new window]   FIG. 5.   Presentation of SEA to human and murine cells by MAbs. (A) Purified human PBMC cells from donor 1 (HD1) or donor 2 (HD2) were activated by SEA presented by the different MAbs, and the result was expressed as a percentage of 100% of activation obtained with SEA presented by MHC class II molecule. Data are means of triplicate measurements. The SEM was less than 10% in all assays. (B) Activation of splenic BALB/C mouse by SEA presented by MAbs. Data are mean counts per minute of triplicate measurements. The SEM was less than 10% in all assays.

View larger version (22K): [in this window] [in a new window]   FIG. 6.   Comparison of the amounts of SEA fixed by different MAbs and levels of cell activation. (A) The different abilities of the six MAbs to interact with SEA after immobilization on the plate were measured by a sandwich ELISA in which the MAbs were immobilized on the plate and incubated in the presence of different amounts of SEA. Then, rabbit anti-SEA antibodies were added, followed by peroxidase-labeled goat anti-rabbit Ig. Results are presented as mean OD490 values. C, negative control. (B) Response of splenic BALB/c cells to activation by different concentrations of SEA presented by the MAbs described in the legend for Fig. 5. Isotypes C and M indicate the activation of cells in the presence of a nonrelated MAb used as a negative control and the medium alone, respectively.

View larger version (29K): [in this window] [in a new window]   FIG. 7.   Activation of splenic mice cells by inactive SEA presented by 6E3 MAb. A SEA mutant at position 60, which can activate human but not murine cells, was presented at a concentration of 30 ng/ml to BALB/c splenic cells by the different MAbs as described in the legend for Fig. 5. M, activation of cells with the medium alone.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial SAGs, and more specifically SE, are bifunctional proteins possessing one site of binding to MHC class II molecules on APC and another site which interacts with TCR to activate a large number of T cells bearing a specific V (10, 20, 34). Different methods have been used to determine the sites by which SAGs interact with both MHC class II molecules and TCR. These include peptide binding and inhibition assays (24, 27), site-directed mutagenesis, the use of recombinant proteins (1, 22), and recently determination of the crystallographic structure of SE alone or with MHC class II molecules (2, 17, 19, 29, 32). Controversial results have been obtained depending on the method used to determine these sites (5, 18, 24-26). In this work, I present a simple method based on antibody-antigen interaction to determine the epitope recognized by the anti-SEA MAb and to test the impact of this interaction on the functional activity of SEA using a panel of 20 anti-SEA MAbs. In addition, I looked for the presence of the epitope recognized by the MAb on two overlapping SEA recombinant molecules which represent the N- and C-terminal regions, respectively. I also used this panel of MAbs to present SEA to human and murine cells. The results concerning the interaction site of SEA with three different MHC class II molecules (DAP; DR1 and Raji; and DR3, DR6, and Daudi DR13) indicate that SEA binds by similar sites to these molecules, since the same MAb blocks SEA binding with the same order of magnitude on all the cell lines used (Fig. 1). In addition, the results confirm the presence of two binding sites on SEA which are functionally different. The first site, recognized by MAbs 1F7, 1F9, 2A11, 2H10, 9B1, and 11A8, appears to be very important for the other immunological activities of SEA; the inhibition of this binding site by any MAb prevents IL-2 production and cell proliferation induced by SEA (Fig. 2 and 3). By contrast, the second site, recognized by MAbs 4C12, 5H3, 7E10, 9F12, and 9H9, is not required for IL-2 production or cell proliferation. Furthermore, blocking this site does not interfere with the subsequent function of SEA (Fig. 2 and 3). No significant difference was found between the intensity of the SEA-MHC class II binding inhibition by the two groups of MAbs. All MAbs block this interaction completely (e.g., 1F7 and 5H3), but there is a qualitative difference involving the site of interaction and its role in SEA function. The presence of two SEA sites interacting with MHC class II molecules has been reported by different groups (1, 16); the first site is mediated by a zinc atom and binds with high affinity to the  chain of the DR1 class II molecule, while the second site binds with low affinity to the  chain of the same molecule. The binding of one SEA to the DR1  chain enhances the binding of a second SEA molecule to the DR1  chain (1, 16). To explain the functional difference between the two sites recognized by the panel of MAbs, I suggest that the first set of MAbs may interact with the high-affinity site of SEA molecule and the second set may interact with the low-affinity site.

For SEA interaction with TCR, the results show the presence of many MAbs directed against the SEA-TCR interaction sites. The most important of them (3B9 and 6E3) are able to completely block IL-2 production and cell proliferation. Previous results demonstrate that 3B9 recognizes a conformational epitope on SEA which was modified by a natural mutation on residue 60 (21). In addition, 3B9 can block 50% of the SEA binding on MHC class II molecules (Fig. 1). 6E3 recognizes an epitope different from that recognized by 3B9. It is not involved in MHC class II binding and reacts with the disulfide loop on the SEA molecule, as well as with the two SEA recombinant proteins and GST. Competition assays between biotinylated 3B9 and 6E3 demonstrate that the sites recognized by the two MAbs are different. These data strongly suggest the presence of two TCR-interacting sites on SEA. The fact that the two MAbs are able to completely block IL-2 production and cell proliferation suggests that the two putative TCR-SEA interacting sites are required to deliver the T-cell activation signal. The presence of two SEA-TCR interaction sites was also recently suggested by a study using cells transfected by the murine TCR V 20 (3).

The negative results regarding the interaction of the MAbs with the two SEA fusion proteins demonstrate that all of the functional sites on SEA depend on the conformation of SEA and require the integrity of this molecule. Although MAb 6E3 was able to react directly with the SEA disulfide loop peptide, this peptide cannot prevent its interaction with SEA, showing its high affinity for the three-dimensional structure of SEA. However, the presence of a small amount of SEA can inhibit all interactions of 6E3 with these molecules, and the same peptide can inhibit the weak interaction of 6E3 with GST (data not shown), suggesting that the intercysteine region is only a part of the epitope recognized by this MAb. These results are supported by the fact that SAGs act as unprocessed proteins in their binding to MHC class II molecules and in T-cell activation (1, 30, 34).

The fact that presentation of SEA by the MAbs is independent of the interaction site of SEA suggests that binding of SAG with MHC class II molecules is not always required to start the activation of the immune system. The incubation of the antibodies against the SEA-MHC sites with SEA in solution completely blocks the binding of SEA to MHC, and for some of them, as well as those against the SEA-TCR site, prevents all cell proliferation. The same MAbs are able to present SEA and induce cell proliferation when they are fixed to an anti-Ig antibody. The activation of murine splenic cells by the mutant SEA presented by the MAb 6E3 supports this idea and indicates that the presentation of SAG by MHC class II molecules is different from that by the MAbs. The differences between these results and those of others showing that only the MAbs directed against the MHC-SAG interacting site can present the SAG SEB to T cells (14) may be due to the utilization of different cells and different SAGs. Efforts to present SEA to two T-cell clones failed to induce activation, possibly due to the angry induction in the absence of MHC or other supporting cells. Also, in contrast to the other activation study which obtained a lower response when the SAG was presented by MAbs, some of the MAbs used here can present and induce a level of activation comparable to the level obtained by the classical activation through MHC (Fig. 7B). This may be due to the affinity of the interaction between the MAb and SEA, to the orientation of SEA after its binding on the MAb, or to the presence of a mixed population of lymphocytes which can participate in this response.

The use of MAbs to map the SEA interaction sites in a functional test has proved to be a good method in the absence of a better biological test or determination of the X-ray crystallographic structure of this molecule complexed with MHC class II molecule and TCR. Moreover, this method is closer to the physiological condition than other methods used until now. It is based on the biological interaction of different members from the same superfamily of Igs (MAb, TCR, and MHC class II molecules) which share many common structures (4, 9). Recently, the results of X-ray crystallography of the TCR-SEA complex have been suggested to resolve the problem of SEA-TCR interaction (2). X-ray crystallography of the SEA complexed with one of our two MAbs (3B9 and 6E3) may determine the site(s) of SEA-TCR interaction without the TCR molecule.