Mitogenic Activities of Amino Acid Substitution Mutants of ... - NCBI

INFECTION AND IMMUNITY, Aug. 1996, p. 3007–3015 0019-9567/96/$04.0010 Copyright q 1996, American Society for Microbiology

Vol. 64, No. 8

Mitogenic Activities of Amino Acid Substitution Mutants of Staphylococcal Enterotoxin B in Human and Mouse Lymphocyte Cultures ROGER J. NEILL,1* MARTI JETT,1 ROBBIN CRANE,1 JAMES WOOTRES,1 CHRISTOPHER WELCH,1 DAVID HOOVER,2 AND PETER GEMSKI1 Division of Pathology1 and Division of Communicable Diseases and Immunology,2 Walter Reed Army Institute of Research, Washington, D.C. 20307-5100 Received 13 February 1996/Returned for modification 21 March 1996/Accepted 16 May 1996

Site-directed mutagenesis has been used to introduce amino acid substitutions at specific residues of the staphylococcal enterotoxin B (SEB) gene cloned from Staphylococcus aureus 10-275. The mitogenic activities of these derivatives were determined in two assay systems: (i) mouse spleen cells and (ii) a mixture of human peripheral blood mononuclear cells and lymphocytes. Substitution of either His-12, His-32, His-121, His-166, Lys-152, or Gly-205 did not significantly alter the mitogenic activity from that of the wild-type toxin in either proliferation assay. Substitution of either residue Asn-23, Phe-44, or Cys-93 reduced the mitogenicity of SEB by a degree that depended upon the assay system used. Similar to the results reported by others measuring toxin activation of mouse lymphoid cells, we found that substitutions of these three residues of SEB caused at least 800-fold reductions of mitogenic activity from that of the wild-type toxin. When tested for toxicity in vivo in D-galactosamine-treated mice, the reduced activities of these mutant toxins, however, were not as pronounced. In contrast, when tested in the human cell mitogenicity assay, these mutant toxins were active. Small alterations in activity (two- to fivefold reduction) were observable only at low concentrations. Our findings reveal the importance of using human lymphocytes in addition to the traditional mouse spleen cell assay when assessing biological activities of staphylococcal enterotoxins.

T-cell receptor (TCR) of T cells expressing specific Vb types to which the toxin binds (11). The structural domains of the SEs that are involved in the receptor interactions leading to T-cell activation have been addressed by analysis of the activities of toxin peptide fragments (21, 27, 45) and genetically engineered amino acid substitution or deletion mutants (9, 17, 18, 20, 23, 29, 40). Evidence from these studies indicate that both amino- and carboxy-terminal regions of the toxin are involved in interactions with the MHC class II and TCRs. Analysis of SEA (15) and SEC1 (22) mutants with substitutions of cysteine residues revealed a role for the disulfide bond of SEs in mitogenic activity. Recent analysis of the three-dimensional structures of SEB (58) and of the MHC class II receptor complexed with SEB (25) have clarified these findings by describing the conformational structure of SEB and identifying complexes of residues involved in receptor interaction. The question of which functional domains of SE are associated with its emetic activity has also been addressed. Carboxymethylation of the histidine residues of SEB abrogated its emetic activity (48). Chemical and enzymatic modification of SEs suggested that neither the disulfide bond nor cystine loop was required for emetic activity (54). However, analysis of SEC mutants with substitutions of cysteine residues led to the proposal that there is a role for the disulfide bond in stabilizing a structure involved in emesis (22). Induction of emesis by SEs has been thought to be a consequence of their superantigen activation of T cells (37). However, analysis of substitution mutations of certain residues of SEB (17, 29) and of SEA (17) and substitution mutations of cysteine residues of SEC1 (22) were interpreted to suggest a dissociation between mitogenicity and emesis. Several studies evaluating the affect of specific mutations in the SE toxin sequence upon mitogenic activity have relied

Staphylococcal enterotoxin B (SEB) is one member of a class of serologically distinct but structurally and functionally related staphylococcal enterotoxins (SEs) secreted by some strains of Staphylococcus aureus (3). Although not identical in amino acid sequence, these proteins reflect several serotypes and constitute a group of toxins that share significant sequence homologies in various regions of the toxin molecule (4, 24, 37). The SEs were originally characterized as inducing emesis and diarrhea in human and nonhuman primates (2). In addition to this distinguishing enterotoxic property, the SEs also exhibit several biological activities that place them within a functionally and genetically related group of pyrogenic exotoxins whose activities impact several organ systems of susceptible hosts and contribute to clinical syndromes that can culminate in lethal shock (50). Among the biological activities reported for SEs are pyrogenicity (8, 12), enhancement of endotoxic shock (57), induction of immunosuppression (52), induction of a variety of cytokines (14, 51), mitogenesis of T cells (32, 44), cytotoxicity to human proximal kidney cells (10), and modulation of phospholipid metabolism resulting in increased levels of cyclooxygenase and lipoxygenase metabolites of the arachidonic acid pathway (7, 26, 43). The nature of the activity of SEs on cells of the immune system puts them in a class of molecules designated as superantigens (37). They induce mitogenesis and cytokine secretion in subclasses of T cells in toxin-specific patterns by interacting with both the major histocompatibility complex (MHC) class II receptor of macrophages and the

* Corresponding author. Mailing address: Division of Pathology, Walter Reed Army Institute of Research, Bldg. 40, Rm. 1051, Washington, DC 20307-5100. Phone: (202) 782-4411. Fax: (202) 7820947. 3007

3008

INFECT. IMMUN.

NEILL ET AL.

upon measurements determined in assays of mouse lymphoid cells (15, 17, 29). SEB is less active against murine cells than human lymphocytes (60). As a consequence, mutant SEBs may behave differently in the two assay systems. To assess this possibility, in the present study we have constructed mutant SEB derivatives by oligonucleotide-directed site-specific mutagenesis and have compared their mitogenic activities in both mouse and human cell assays. We have found that the relative mitogenic activities of several SEB mutants differ significantly, depending whether mouse or human cells are used in the assay. Our findings reveal the importance of using human lymphocytes in addition to the traditional mouse spleen cell assay when assessing biological activities of SEs. MATERIALS AND METHODS Medium and toxins. For toxin production, S. aureus strains were grown in CTG medium that contained 2.5% Casamino Acids (Difco, Detroit, Mich.), 0.3% tryptone (Difco), 0.5% NaCl, 0.26% Na2PO4, 0.13% NH4Cl, 0.001% thiamine, 0.001% nicotinic acid, and 0.2% glucose. For a positive control for the toxin in the enzyme-linked immunosorbent assay (ELISA) to measure SEB concentration (42), we used an SEB preparation (lot 14-30) provided by the U.S. Army Research Institute of Infectious Diseases, Frederick, Md., that had been previously purified by the method of Schantz et al. (49) from strain 10-275. Cloning of SEB gene from strain 10-275. For this study we cloned the SEB gene from the DNA of S. aureus 10-275, which is a derivative of strain S6 (53). The restriction enzyme map and nucleotide sequence for the SEB gene from S. aureus S6 have been previously characterized (28, 47). With this information, the 10-275 SEB gene was cloned as a KpnI-ClaI restriction fragment into plasmid pUC19 digested with KpnI and AccI, yielding plasmid p19KC. This SEB gene was next subcloned into a shuttle vector plasmid (pSV2) that was constructed for this study so as to facilitate expression of toxin in either Escherichia coli or S. aureus. pSV2 was derived by modification of plasmid pRIT5 (Pharmacia LKB Biotechnology, Piscataway, N.J.) as follows. (i) Plasmid pRIT5 was first digested with HincII and SmaI and then religated to remove these restriction sites. (ii) The region of the modified pRIT5, bound by restriction sites EcoRI and ScaI, was replaced with the EcoRI-ScaI fragment of pUC18 containing the multiple cloning site, thus yielding plasmid pSV2. The SEB gene was removed from p19KC as an EcoRI-HindIII fragment and inserted into EcoRI-HindIII-digested pSV2. Construction of SEB mutant toxins. We separated by restriction enzyme digestion the portion of the SEB gene that encodes mature toxin sequences (amino acid residues 3 through 239) from those SEB sequences which control its expression and introduced them into phagemid vector pTZ19 (Bio-Rad Laboratories, Richmond, Calif.). Single-stranded DNA isolated from this phagemid clone was subjected to oligonucleotide-directed site-specific mutagenesis. This was accomplished by following the manufacturer’s instructions (MUTA-GENE, version 2; Bio-Rad Laboratories) for incorporation of approximately 27-baselong synthetic oligonucleotides that were complementary to the SEB gene sequence (28), except for specific base changes encoding amino acid substitutions (see Table 1). Incorporation of specific mutations was verified by reactions with specific oligonucleotide probes and by standard DNA sequencing procedures. Mutant toxins containing two mutations were constructed by site-specific incorporation of the second mutation into the SEB gene containing the first mutation as described above. The resulting mutant gene sequences were then recloned into pSV2 containing SEB sequences controlling expression to reconstitute a complete toxin gene containing one or two amino acid substitutions. Plasmids encoding wild-type or mutant SEB genes were electroporated (30) into S. aureus RN4220 (31) for expression of toxin in culture fluid. Production of mutant toxin proteins was established by ELISA (42). Purification of toxins. Derivatives of strain RN4220 harboring clones of wildtype or mutant genes of SEB were grown in CTG medium. Cells were removed by centrifugation, and the culture fluid containing toxin was sterilized by filtration. Toxin was recovered from the culture filtrate by precipitation with ammonium sulfate to 80% saturation. The precipitate was recovered by centrifugation and resuspended with water at 1/100 the volume of the culture. Trifluoroacetic acid was added to a final concentration of 0.3%, and after 3 h at 48C, insoluble material was removed by centrifugation. The supernatant was dialyzed against water to remove salt. The toxin preparation was further purified by cationexchange high-pressure liquid chromatography (56). The sample was eluted from an SP-5PW column (7.5 by 75 mm; Waters Chromatography, Milford, Mass.) (equilibrated at pH 6.0) with a sodium chloride gradient at a flow rate of 1 ml/min. Peak toxin-containing fractions (0.5 ml) were pooled, and protein concentrations were determined by measuring A277 (49) and by a dye-binding protein assay (Bio-Rad Laboratories) and verified by ELISA. Proliferation assays using human cells and mouse splenocytes. The human cell proliferation assay was performed as described previously (27). Human peripheral blood mononuclear cells and small lymphocytes were obtained by leukophoresis of healthy donors and purified by centrifugation over lymphocyte

TABLE 1. Amino acid substitution mutants of SEB Toxin designation

Description or amino acid substitution

SEB

Wild-type clone

H12L N23I H32L F44S C93S H121D H121Q K152Y H166D G205R

Histidine 123leucine Asparagine 233isoleucine Histidine 323leucine Phenylalanine 443serine Cysteine 933serine Histidine 1213aspartate Histidine 1213glutamine Lysine 1523tyrosine Histidine 1663aspartate Glycine 2053arginine

N23I/F44S C93S/F44S H121D/H166D

Double mutant Double mutant Double mutant

Codon change (wild type3mutant)

CAC3CTG AAT3ATT CAT3CTG TTT3TCT TGT3TCT CAT3GAT CAT3CAG AAG3TAT CAC3GAC GGA3AGA

separation medium (Organon Teknika, Durham, N.C.). Lymphocytes and monocytes were further purified by use of counterflow centrifugation-elutriation with Ca- and Mg-free Dulbecco’s phosphate-buffered saline (Sigma Chemical Co., St. Louis, Mo.) as the eluant. This procedure usually yielded cell preparations with more than 95% viability and less than 5% contamination with other cell types. Lymphocytes (2.0 3 105 cells per well) and monocytes (0.5 3 105 cells per well) were mixed in tissue culture medium consisting of RPMI 1640 (Biofluids, Gaithersburg, Md.) containing 10% human AB serum (Sigma) and plated in 96-well plates (Costar, Cambridge, Mass.). Following the addition of toxin diluted in tissue culture medium, cells were incubated for 60 h in a humidified atmosphere (5% CO2) with 1 mCi of [methyl-3H]thymidine added for the final 15 h of incubation. Cells were harvested on glass-fiber mats with a cell harvester (Skatron, Sterling, Va.), and radioactivity was determined with a beta counter (LKB/ Wallace, Gaithersburg, Md). Eight replicates per sample were performed for each assay, and the SEB dose response was determined with each experiment. For the mouse cell assay (55), splenocytes were obtained from the spleens of C3H/HeJ or BALB/c mice and cultivated as described above. Splenocytes in tissue culture medium were added to 96-well plates (5 3 105/well), and toxininduced proliferation was isotopically measured as described above. Toxic activity in D-galactosamine-treated mice. Lethal sensitivity to wild-type or mutant SEB toxins was tested in BALB/c mice (males, ca. 20 g) treated with D-galactosamine as described by Miethke et al. (39). D-Galactosamine (50 mg/ ml) and toxin samples were diluted in sterile pyrogen-free water. At the time of challenge, mice received intraperitoneal injections (0.5 ml each) of D-galactosamine (25 mg) followed immediately by a dose of toxin sample. Mice were monitored twice daily, and survival of mice was recorded at 72 h after injection. Animal care was in accordance with the Guide for Care and Use of Laboratory Animals (41a).

RESULTS Mitogenic activities of SEB mutants. As part of our studies on the relation of structure and function of SEB, we chose to introduce amino acid substitutions into selected residues of SEB in order to assess the effects of these changes on biological activity. The rationales for our choices of residues for mutagenesis are described in the Discussion. Table 1 is a summary of the specific SEB mutants that were constructed for this study by oligonucleotide-directed mutagenesis. The residue number indicates the position of the amino acid relative to the amino-terminal end of the mature toxin. Codon sequences for the wild-type residues were obtained from the reported sequence of SEB (28) and were verified by DNA sequencing of the SEB gene used in this study. The yield of mutant toxin detected in supernatant culture fluids was generally within twofold of that detected for the wild-type SEB-producing strain grown under similar conditions. As a measure of the effect of particular amino acid substitutions on toxin-mediated T-cell activation, purified mutant SEB toxin proteins were tested for their ability to induce T-cell

VOL. 64, 1996

MITOGENIC ACTIVITIES OF SEB MUTANTS

3009

FIG. 1. Comparison of the mitogenic activities of wild-type SEB and mutant F44S toxin in cultures of human peripheral blood lymphocytes and monocytes or in cultures of mouse splenocytes.

proliferation. Mitogenic activity was determined in both human lymphocyte and murine splenocyte assays as a means of comparing the relative effects of the mutations on different species of target cells. Such a comparison is illustrated in Fig. 1. Kappler and coworkers (29) had previously reported that a serine substitution of phenylalanine 44 of SEB resulted in a mutant that was severely reduced in its ability to activate murine T cells, as measured by interleukin-2 production. We constructed a derivative of SEB containing the same substitution mutation (F44S) and compared its mitogenic activities versus that of wild-type SEB in mouse and human cell assays. Mutant F44S was a less effective mitogen (about fivefold-reduced activity) at lower concentrations (,50 ng/ml) than SEB in the human cell assay. At higher concentrations, F44S approaches wild-type activity. As has been reported previously (60), the mouse cell assay is approximately 10 to 20 times less reactive to SEB as a mitogen than the human cell assay (Fig. 1). In contrast to the human cell assay, mutant F44S exhibits no appreciable activity

in the mouse assay even at concentrations 100 times greater than necessary to detect activity in the human cell assay (Fig. 1). A similar analysis was performed on the mitogenic activities of all the mutant toxins we constructed compared with that of wild-type SEB. Figure 2 is a composite of an extensive series of dose-response experiments, comparable to the one illustrated in Fig. 1, that measured proliferation induced by wild-type and mutant SEB toxins in both the human lymphocyte and mouse spleen cell assays. The bar graphs represent different regions of the toxin dose-response curve, providing the activities of the mutant SEBs relative to that of wild-type SEB at specific toxin concentrations. This approach permits detection of small but reproducible differences in toxin activity that may be concentration dependent. The concentrations chosen for analysis were those at which SEB mitogenic activity is at its maximal (Fig. 2A), half-maximal (Fig. 2B), and one-fifth maximal (Fig. 2C) levels. The activity of SEB at each concentration is set at 100%, whereas the mutant toxin activities are calculated rela-

3010

NEILL ET AL.

INFECT. IMMUN.

VOL. 64, 1996


3011

FIG. 2. Relative mitogenic activities of SEB mutants in human and mouse cell assays. Data were extracted from a series of dose-response curves for each toxin determined as shown in Fig. 1. The results are averages of six experiments of eight replicates each for the human cell assay and three experiments for the mouse cell assay. Mitogenic activity is expressed as a percentage of the activity exhibited by wild-type SEB under the same conditions, which is set at 100%. (A) Maximum mitogenic activity achieved by each toxin. In the human assay, maximum activity was determined at 1 mg/ml. For the mouse cell assay, maximum activity was determined at 4.0 mg/ml. (B) Mitogenic activity of each toxin at the concentration (0.040 mg/ml for the human cell assay and 0.74 mg/ml for the mouse cell assay) at which SEB is at 50% of its maximum activity. (C) Mitogenic activity of each toxin at the concentration (0.009 mg/ml for the human cell assay and 0.218 mg/ml for the mouse cell assay) at which SEB is at 20% of its maximum activity. Each error bar represents the standard error of the mean. Values which are significantly different (P # 0.04, Student’s t test) from the SEB control value are indicated with an asterisk.

tive to this level. The results shown are averages of six experiments of eight replicates each for the human cell assay and three experiments for the mouse cell assay. Many of the amino acid substitution mutants exhibited some minor differences in activity compared with wild-type SEB. These differences were often concentration dependent and were not always the same in both assays. Some of the mutations cause profound changes when tested in the mouse assay but not in the human cell assay. In the human cell assay six of the mutant toxins (H32L, H121D, H121Q, K152Y, H166D, and double mutant H121D/ H166D) had activities comparable to that of SEB at all three concentrations examined. At the concentration giving the maximum level of SEB activity (Fig. 2A), only two mutants (double mutants N23I/F44S and C93S/F44S) showed significant (more than twofold) reduction in activity from that of wild-type SEB. At low levels of toxin (Fig. 2C), two mutants (N23I and G205R) exhibited 2- to threefold reduction in activity, whereas five mutants (H12L, F44S, C93S, N23I/F44S, and C93S/F44S) were 4- to 10-fold less active. These patterns of activity by the mutant toxins in the human cell assay were not duplicated in the mouse splenocyte assay (Fig. 2). Whereas H12L and G205R exhibited small reductions (2.5- to 4-fold) in activity at low concentrations in the human lymphocyte assay, they did not exhibit this degree of reduction in the mouse cell assay. Several mutants (H121D, H121D

H166D, and G205R) exhibited slight (30 to 60%) increases in activity (Fig. 2A and B) in the mouse assay but not in the human cell assay. At the lower toxin concentration, substitution of histidine 121 with aspartic acid (H121D) but not glutamine (H121Q) caused a fourfold reduction in activity from that of the wild type (Fig. 2C). As in the human cell assay, mutants K152Y, H32L, H121Q, and H166D tested in the mouse splenocyte assay had activities comparable to that of the wild-type toxin (Fig. 2). The most profound differences in behavior between the two assays were exhibited by mutant toxins encoding substitution N23I, F44S, or C93S. Unlike the results from the human cell assay, where mitogenic activity was readily detectable and nearly equivalent to SEB at maximal levels (Fig. 2A), toxins with any of these three mutations exhibited no measureable mitogenic activity at 4 mg/ml (Fig. 2A) in the mouse spleen cell assay. In an effort to determine the extent of decrease of mitogenic activity in the mouse splenocyte assay by mutant toxins with N23I, F44S, or C93S substitutions, we tested their activity at increased concentrations. The results in Fig. 3 indicate that mutants N23I, F44S, C93S and the double mutant N23I/F44S do not exhibit activity comparable to wild-type activity even at 800-fold-greater concentration. Mitogenic activity of SEB mutant F44S in mixtures of mouse and human cells. In order to address the question of which cell type in mouse splenocytes was not responsive to the

3012

NEILL ET AL.

INFECT. IMMUN.

FIG. 3. Relative mitogenic activities of mutant SEB proteins at high doses in the mouse spleen cell assay. Each result is the average of eight replicates. Each error bar represents the standard error of the mean.

SEB mutant F44S, we performed mixed culture experiments. The mitogenic activity of SEB mutant F44S was determined in mouse splenocyte preparations which were supplemented either with monocytes (Fig. 4A) or lymphocytes (Fig. 4B) obtained from human peripheral blood. Mitogenic activity was detected only in cultures of mouse splenocytes supplemented with human monocytes (Fig. 4A) which encode the MHC class II receptor. The recovered activity represents approximately 70% of the activity induced by wild-type SEB in the mouse splenocyte assay (data not shown). The slight increase in activity detected in the cultures supplemented with human lymphocytes (Fig. 4B) very likely reflects the presence of monocytes in the lymphocyte preparations (6% monocytes and 94% lymphocytes). Toxicities of SEB mutants in D-galactosamine-treated mice. We wished to determine if the degree to which C93S, F44S, and N23I caused severe reduction in mitogenic activity in vitro using mouse spleen cells was reflected similarly in vivo. For this purpose we employed the D-galactosamine-sensitized mouse model as an indicator of toxin-mediated T-cell activation in vivo (39). Table 2 summarizes the results of evaluating the toxic sensitivities of D-galactosamine-treated mice to mutant toxins F44S, N23I, and C93S. At the 20-mg dose of toxin, wild-type SEB induced 80% mouse lethality, similar to results reported previously (39). In contrast, mutant toxins F44S, N23I, and C93S caused no lethality at this dose. At 200 mg, however, N23I and F44S induced lethality (100 and 80%, respectively) comparable to that of the wild-type toxin at 20 mg, whereas C93S induced 60% lethality. This approximately 10fold reduction in toxin-mediated lethality in vivo is significantly less pronounced than the .800-fold reduction in toxin-mediated T-cell activation measured in vitro.

DISCUSSION In this study we have examined the effect of substitution of selected amino acids in SEB (Table 1) upon its mitogenic activity against lymphoid cells from two different species: humans and mice. We found that changes in activity that are associated with some of these mutations were not comparable between the two assays. Our rationale for preparing mutant toxins with substitutions of histidine residues of SEB was based in part on past studies of the chemical modification of SEB that established that carboxymethylation of histidine residues of SEB abolished its emetic activity (48). Moreover, mutagenesis studies (5, 6) indicated that certain histidine residues were important in the biological activity of toxic shock syndrome toxin 1 (TSST-1), a related superantigen produced by S. aureus. In addition, unlike other histidine residues of SEs, histidine 121 is highly conserved among the enterotoxins (37), and such conservation may reflect association with biological function. However, the results of our studies with the genetic substitutions that we constructed (H12L H32L, H121D, H121Q, and H166D) in four of the five histidine residues of the SEB molecule indicate that these four histidines of SEB are not of key importance for mitogenic activity in either the human or mouse cell assay (Fig. 2). The lack of association of histidine residues of SEB with mitogenic activity seen in this study supports an earlier observation that SEB containing carboxymethylated histidines retained mitogenicity for monkey lymphocytes (1). Because not all replacements for residues will have the same effect, it is possible that the choice of other amino acids for substitution of histidine residues of SEB might lead to constructs with significant alterations in mitogenic activity. The

VOL. 64, 1996

FIG. 4. Relative mitogenic activity of the SEB mutant F44S in mouse spleen cells supplemented with either human monocytes (A) or human lymphocytes (B). Toxin-induced proliferation was measured in mouse splenocyte cultures (5.5 3 105 cells per well) supplemented with either human monocytes (0.5 3 105 cells per well) or human lymphocytes (2.0 3 105 cells per well) at ratios comparable to those used in the human cell proliferation assay. Symbols for panel A, ■, human monocytes plus mouse splenocytes; h, human monocytes; F, mouse splenocytes. Symbols for panel B: ç, human lymphocytes plus mouse splenocytes; É, human lymphocytes; F, mouse splenocytes.

reduction in activity seen in the human cell assay at the low toxin concentration exhibited by an aspartic acid substitution (H121D) for His-121 was not exhibited when glutamine (H121Q) was used for substitution (Fig. 2). An alanine substitution of histidine 135 severely reduces mitogenic activity of TSST-1 when assayed in mouse or human cells (6, 13). His-135 of TSST-1 is equivalent to His-166 of SEB in sequence alignment and in three-dimensional structure (37, 46). Two other residues were selected for mutagenesis as likely candidates for involvement in T-cell proliferation. First, glycine 205 (Gly-205) represents a highly conserved residue among the SEs (37) and is adjacent to sequences whose equivalents in SEA and SEE are important in TCR interaction (23, 40). Recent X-ray crystallographic analysis of the three-dimensional structures of SEB and SEC places Gly-205 in a complex of residues thought to constitute the TCR binding site (58, 59). Second, lysine 152 (Lys-152) was selected because it is highly conserved among the SEs and TSST-1 (37). Moreover, a synthetic peptide encoding residues 130 to 160 of SEB, which includes Lys-152, inhibited mitogenesis in a human cell assay


3013

(27), thereby implicating a role for this region of the molecule in mitogenic activity. Despite these expectations, these two residues do not appear to play an important role in mitogenicity (Fig. 2). Although the substitution mutations in SEB that we have made in the histidine residues and Lys-152 and Gly205 do not reveal that they play a critical role in mitogenesis, the small changes in activity associated with some of the mutations may reflect small effects on toxin interactions with receptors such as with individual Vb types within the larger repertoire of T cells activated by SEB. Changes in the activation of subpopulations of T cells resulting from mutagenesis of specific residues in SEB, SEA, and SEE have been reported (19, 23, 29, 40). Several of the substitution mutants of SEB that we tested exhibited profound differences in the degree of reduction in activity, depending upon the species of lymphocytes used for assay. Past studies utilizing an isoleucine substitution of asparagine 23 of SEB, designated BC-6 (29), and serine or alanine substitutions of the cysteine residues of SEA (15) indicated that these residues are important for MHC class II or TCR interactions. We constructed SEB mutants containing either an isoleucine substitution of asparagine 23 (N23I) or a serine substitution for cysteine 93 (C93S). When tested in the mouse cell assay, our mutants N23I and C93S exhibited at least an 800-fold decrease in activity compared with wild-type SEB activity (Fig. 2). In the human cell assay, however, the maximum activities of mutants N23I and C93S were comparable to that of SEB (Fig. 2). Small decreases in activity (two- to fivefold) were observed for these mutants but only at low concentrations, and these decreases were far less than the decrease measured in mouse cells. Similar to the results we observed for the SEB mutant C93S, serine substitution for Cys-93 of SEC1 did not substantially alter its mitogenic activity from that of wild-type SEC1 when proliferation was measured with human lymphocytes (22). Diseases mediated by toxin superantigens such as SEB are thought to be a consequence of events involving the toxin’s stimulation of T cells (36, 37). Thus, functional sites on these toxins associated with T-cell activation should also be associated with disease. However, recent reports have suggested that domains of the toxin affecting T-cell stimulation, as measured by mitogenic activity, may be separable from other biological activities of the toxins. For example, the results of mutational analysis of TSST-1 indicated that residues associated with mitogenic activity in mouse splenocytes were distinct from those associated with TSST-1-mediated lethality in rabbits (41). In addition, SEC1 mutants containing alanine substitutions of cysteine residues retained significant mitogenic activity in a human cell assay but lost emetic activity (22).

TABLE 2. Toxicities of SEB mutants in D-galactosamine-treated mice Toxin

Dose (mg)

No. of surviving mice (n 5 5)

SEB

20 50

1 0

N23I

20 200

5 0

F44S

20 200

5 1

C93S

20 200

5 2

3014

NEILL ET AL.

Characterizations of amino acid substitution mutants of SEB and SEA also have led to suggestions of dissociation of the T-cell activation and emetic properties expressed by these two toxins (17). One such mutant, a derivative of SEB, designated BR-358 (29) and containing a serine substitution of phenylalanine 44, is reduced at least 100-fold in its ability to induce interleukin-2 production by murine T cells and 100- to 1,000fold in its ability to bind to human MHC class II receptor (29). Subsequent studies examined the emetic activity of mutant BR-358 (17) as well as that of mutant SEA toxin SEA-F47S (16), which contains a serine substitution in the analogous phenylalanine residue of SEA. Both mutants retained emetic activity in monkeys despite essentially no murine T-cell stimulatory activity (16, 17, 29). This finding led to the suggestion that the toxin domains of SEB and SEA associated with emetic activity and with T-cell stimulation can be dissociated (17) or that there is a lack of correlation between mitogenicity and emesis (16). The results of this present study reveal a potential risk in relying upon the mouse splenocyte assay to evaluate the mitogenic capacity of SEB. In this study we constructed a derivative of SEB, designated F44S, that contains a serine substitution for phenylalanine 44 and is analogous to mutant BR-358 prepared by Kappler et al. (29). F44S is nonmitogenic in the murine splenocyte assay (Fig. 1, 2, and 3), confirming the previous report of the loss of activation of murine T cells as a consequence of this substitution mutation (29). In order to more closely reflect toxin behavior in primates, we extended the characterization of the mitogenic properties of F44S to include behavior in assays containing human lymphocytes and monocytes. When mitogenicity was evaluated with human cells for measurements rather than mouse splenocytes, it became readily evident that F44S retains significant mitogenic activity (Fig. 1 and 2). The loss of activity in the mouse splenocyte assay can be recovered in part by the addition of human monocytes (Fig. 4A), which suggests that the inactivity of F44S in the mouse cell assay is the result of lack of binding of the mutant toxin to mouse monocytes. These data indicate that the substitution of serine for Phe-44 in SEB does not serve to dissociate its mitogenic activity from its emetic activity. Although this study did not evaluate the mitogenic activities of SEA mutants, the results suggest that comparison of mitogenic activity determined by assays of rodent cell cultures with results of emetic activity derived from primates requires cautious interpretation. Such differences in apparent activity for F44S, N23I, and C93S, dependent upon the species of cells used to test proliferation, may be a reflection of the suggestion (33, 35) that superantigens evolved for optimal interaction with a particular target host. TSST-1 (33) and SEC (35) prepared from S. aureus strains isolated from sheep exhibited optimal mitogenic activity for ovine lymphocytes compared with human or bovine lymphocytes. Mutations in the toxin may have a more profound effect in those culture assays that are not optimal for measuring activity. Our results (Fig. 2 and 3 and Table 2) with SEB mutants C93S, F44S, and N23I suggest that lethality in the in vivo D-galactosamine-treated mouse model, which is dependent upon T-cell activation (39), may detect residual mitogenic activity not readily apparent in the in vitro mouse splenocyte mitogenicity assay. In a similar observation regarding the sensitivities of assays, amounts of SEA insufficient to induce detectable mitogenesis of mouse cells in vitro were sufficient to mediate T-cell deletion in vivo in mice (38). Thus, the mouse splenocyte assay alone may be an incomplete measure of a toxin’s mitogenic potential. In this study, we have also attempted to identify possible synergistic effects on the mitogenic activity of two substitution

INFECT. IMMUN.

mutations in the same SEB molecule. The presence of two substitution mutations of histidine residues in one molecule (H121D/H166D) did not significantly reduce activity in either assay. Asn-23 and Cys-93 have been implicated in TCR interactions (15, 29), whereas Phe-44 has been implicated in MHC class II binding (29). Double mutants N23I/F44S and C93S/ F44S exhibited decreases in activity greater than any of the mutants with single mutations, indicating a cumulative effect of these mutations on activity in the human cell assay. This study has revealed the importance of using human lymphocytes in addition to the traditional mouse spleen cell assay when addressing structure-function relationships of SEB and assessing the biological activities of SEs. This conclusion is based on our finding that certain mutant SEB toxins appear to be nonmitogenic when measured in the mouse splenocyte assay but mitogenic when tested in human immune cell culture. Indeed, this variability in measurable mitogenic activity depending which host cells are used for assay is not limited to SEB. Compared with SEB, SEA is a more potent mitogen of mouse splenocytes (60). A naturally occurring mutant derivative of SEA containing an asparagine substitution for aspartic acid at position 60 was recently reported to exhibit severely reduced mitogenic activity in a mouse cell assay but not in a human cell assay (34). Thus, the determination of the full effect of mutations in SEs on their mitogenic functions requires analysis by the human cell assay in addition to the mouse spleen cell assay. ACKNOWLEDGMENT We thank John Iandolo for providing S. aureus RN4220. REFERENCES 1. Alber, G., D. K. Hammer, and B. Fleischer. 1990. Relationship between enterotoxic- and T lymphocyte-stimulating activity of staphylococcal enterotoxin B. J. Immunol. 144:4501–4506. 2. Bergdoll, M. S. 1970. Enterotoxins, p. 265–326. In S. J. Ajl, T. C. Montie, and S. Kadis (ed.), Microbial toxins. Academic Press, Inc., New York. 3. Bergdoll, M. S. 1983. Enterotoxins, p. 559–598. In C. S. F. Easmon and C. Adlam (ed.), Staphylococci and staphylococcal infections. Academic Press, New York. 4. Betley, J. J., D. W. Borst, and L. B. Regassa. 1992. Staphylococcal enterotoxins, toxic shock syndrome toxin and streptococcal pyrogenic exotoxins: a comparative study of their molecular biology. Chem. Immunol. 55:1–35. 5. Blanco, L., W. M. Choi, K. Connolly, M. R. Thompson, and P. F. Bonventre. 1990. Mutants of staphylococcal toxic shock syndrome toxin 1: mitogenicity and recognition by a neutralizing monoclonal antibody. Infect. Immun. 58: 3020–3028. 6. Bonventre, P. F., H. Heeg, C. Cullen, and C.-J. Lian. 1993. Toxicity of recombinant toxic shock syndrome toxin 1 and mutant toxins produced by Staphylococcus aureus in a rabbit infection model of toxic shock syndrome. Infect. Immun. 61:793–799. 7. Boyle, T., V. Lancaster, R. Hunt, P. Gemski, and M. Jett. 1994. Method for simultaneous isolation and quantitation of platelet activating factor and multiple arachidonate metabolites from small samples: analysis of effects of Staphylococcus aureus enterotoxin B in mice. Anal. Biochem. 216:373–382. 8. Brunson, K. W., and D. W. Watson. 1974. Pyrogenic specificity of streptococcal exotoxins, staphylococcal enterotoxin, and gram-negative endotoxin. Infect. Immun. 10:347–351. 9. Buelow, R., R. E. O’Hehir, R. Schreifels, T. J. Kummerehl, G. Riley, and J. R. Lamb. 1992. Localization of the immunologic activity in the superantigen staphylococcal enterotoxin B using truncated recombinant fusion proteins. J. Immunol. 148:1–6. 10. Chatterjee, S., and M. Jett. 1992. Glycosphingolipids: the putative receptor for Staphylococcus aureus enterotoxin B in human kidney proximal tubular cells. Mol. Cell. Biochem. 113:25–31. 11. Choi, Y., B. Kotzin, L. Herron, J. Callahan, and P. Marrack. 1989. Interaction of Staphylococcus aureus toxin “superantigens” with human T cells. Proc. Natl. Acad. Sci. USA 86:8941–8945. 12. Clark, W. G., and H. L. Borison. 1963. Pyrogenic effect of purified staphylococcal enterotoxin. J. Pharmacol. Exp. Ther. 142:237–241. 13. Drynda, A., B. Konig, P. F. Bonventre, and W. Konig. 1995. Role of a carboxy-terminal site of toxic shock syndrome toxin 1 in eliciting immune responses of human peripheral blood mononuclear cells. Infect. Immun. 63:1095–1101. 14. Grossman, D., J. G. Lamphear, J. A. Mollick, M. J. Betley, and R. R. Rich.

VOL. 64, 1996

15. 16. 17.

18. 19.

20.

21.

22.

23. 24. 25.

26.

27.

28. 29. 30. 31.

32. 33.

34.

35. 36.

1992. Dual roles for class II major histocompatibility complex molecules in staphylococcal enterotoxin-induced cytokine production and in vivo toxicity. Infect. Immun. 60:5190–5196. Grossman, D., M. Van, J. A. Mollick, S. K. Highlander, and R. R. Rich. 1991. Mutation of the disulfide loop in staphylococcal enterotoxin A: consequences for T cell recognition. J. Immunol. 147:3274–3281. Harris, T. O., and M. J. Betley. 1995. Biological activities of staphylococcal enterotoxin type A mutants with N-terminal substitutions. Infect. Immun. 63:2133–2140. Harris, T. O., D. Grossman, J. W. Kappler, P. Marrack, R. R. Rich, and M. J. Betley. 1993. Lack of complete correlation between emetic and T-cell stimulatory activities of staphylococcal enterotoxins. Infect. Immun. 61:3175–3183. Harris, T. O., W. O. Hufnagle, and M. J. Betley. 1993. Staphylococcal enterotoxin type A internal deletion mutants: serological activity and induction of T-cell proliferation. Infect. Immun. 61:2059–2068. Hayball, J. D., J. H. Robinson, R. E. O’Hehir, A. Verhoef, J. R. Lamb, and R. A. Lake. 1994. Identification of two binding sites in staphylococcal enterotoxin B that confer specificity for TCR Vb gene products. Int. Immunol. 6:199–211. Hedlund, G., M. Dohlsten, T. Herrmann, G. Buell, P. A. Lando, S. Segren, J. Schrimsher, H. R. MacDonald, H. O. Sjogren, and T. Kalland. 1991. A recombinant C-terminal fragment of a staphylococcal enterotoxin A binds to human MHC class II products, but does not activate T-cells. J. Immunol. 147:4082–4085. Hoffmann, M. L., L. M. Jablonski, K. K. Crum, S. P. Hackett, Y. I. Chi, C. V. Stauffacher, D. L. Stevens, and G. A. Bohach. 1994. Predictions of T-cell receptor and major histocompatibility complex-binding sites on staphylococcal enterotoxin C1. Infect. Immun. 62:3396–3407. Hovde, C. J., J. C. Marr, M. L. Hoffmann, S. P. Hackett, Y. I. Chi, K. K. Crum, D. L. Stevens, C. V. Stauffacher, and G. A. Bohach. 1994. Investigation of the role of the disulphide bond in the activity and structure of staphylococcal enterotoxin C1. Mol. Microbiol. 13:897–909. Hudson, K. R., H. Robinson, and J. D. Fraser. 1993. Two adjacent residues in staphylococcal enterotoxins A and E determine T cell receptor Vb specificity. J. Exp. Med. 177:175–184. Iandolo, J. J. 1989. Genetic analysis of extracellular toxins of Staphylococcus aureus. Annu. Rev. Microbiol. 43:375–402. Jardetzky, T. S., J. H. Brown, J. C. Gorga, L. J. Stern, R. G. Urban, Y. Chi, C. Stauffacher, J. L. Strominger, and D. C. Wiley. 1994. Three-dimensional structure of a human class II histocompatibility molecule complexed with superantigen. Nature (London) 368:711–718. Jett, M., W. Brinkley, R. Neill, P. Gemski, and R. Hunt. 1990. Staphylococcus aureus enterotoxin B challenge of monkeys: correlation of plasma levels of arachidonic acid cascade products with occurrence of illness. Infect. Immun. 58:3494–3499. Jett, M., R. Neill, C. Welch, T. Boyle, E. Bernton, D. Hoover, G. Lowell, R. E. Hunt, S. Chatterjee, and P. Gemski. 1994. Identification of staphylococcal enterotoxin B sequences important for induction of lymphocyte proliferation by using synthetic peptide fragments of the toxin. Infect. Immun. 62:3408–3415. Jones, C. L., and S. A. Khan. 1986. Nucleotide sequence of the enterotoxin B gene from Staphylococcus aureus. J. Bacteriol. 166:29–33. Kappler, J. W., A. Herman, J. Clements, and P. Marrack. 1992. Mutations defining functional regions of the superantigen staphylococcal enterotoxin B. J. Exp. Med. 175:387–396. Kraemer, G. R., and J. J. Iandolo. 1990. High-frequency transformation of Staphylococcus aureus by electroporation. Curr. Microbiol. 21:373–376. Kreiswirth, B. N., S. Lofdahl, M. J. Betley, M. O’Reilly, P. Schlievert, M. S. Bergdoll, and R. P. Novick. 1983. The toxic shock syndrome toxin exotoxin structural gene is not detectably transmitted by a prophage. Nature (London) 305:709–712. Langford, M. P., G. J. Stanton, and H. M. Johnson. 1978. Biological effects of staphylococcal enterotoxin A on human peripheral lymphocytes. Infect. Immun. 22:62–68. Lee, P. K., B. N. Kreiswirth, J. R. Deringer, S. J. Projan, W. Eisner, B. L. Smith, E. Carlson, R. P. Novick, and P. M. Schlievert. 1992. Nucleotide sequences and biologic properties of toxic shock syndrome toxin 1 from ovine and bovine associated Staphylococcus aureus. J. Infect. Dis. 165:1056–1063. Mahana, W., R. Al-Daccak, C. Leveille, J.-P. Valet, J. Hebert, M. Ouellette, and W. Mourad. 1995. A natural mutation of the amino acid residue at position 60 destroys staphylococcal enterotoxin A murine T-cell mitogenicity. Infect. Immun. 63:2826–2832. Marr, J. C., J. D. Lyon, J. R. Roberson, M. Lupher, W. C. Davis, and G. A. Bohach. 1993. Characterization of novel type C staphylococcal enterotoxins: biological and evolutionary implications. Infect. Immun. 61:4254–4262. Marrack, P., M. Blackman, E. Kushnir, and J. Kappler. 1990. The toxicity

Editor: V. A. Fischetti


3015

of staphylococcal enterotoxin B in mice is mediated by T cells. J. Exp. Med. 171:455–464. 37. Marrack, P., and J. Kappler. 1990. The staphylococcal enterotoxins and their relatives. Science 248:705–711. 38. McCormack, J. E., J. E. Callahan, J. Kappler, and P. C. Marrack. 1993. Profound deletion of mature T cells in vivo by chronic exposure to exogenous superantigen. J. Immunol. 150:3785–3792. 39. Miethke, T., C. Wahl, K. Heeg, B. Echtenacher, P. H. Krammer, and H. Wagner. 1992. T cell-mediated lethal shock triggered in mice by the superantigen staphylococcal enterotoxin B: critical role of tumor necrosis factor. J. Exp. Med. 175:91–98. 40. Mollick, J. A., R. L. McMasters, D. Grossman, and R. R. Rich. 1993. Localization of a site on bacterial superantigens that determines T cell receptor B chain specificity. J. Exp. Med. 177:283–293. 41. Murray, D. L., G. S. Prasad, C. A. Earhart, B. A. B. Leonard, B. N. Kreiswirth, R. P. Novick, D. H. Ohlendorf, and P. M. Schlievert. 1994. Immunobiologic and biochemical properties of mutants of toxic shock syndrome toxin-1. J. Immunol. 152:87–95. 41a.National Institutes of Health. 1985. Guide for care and use of laboratory animals. National Institutes of Health publication no. 85-23. National Institutes of Health, Bethesda, Md. 42. Neill, R. J., G. R. Fanning, F. Delahoz, R. Wolff, and P. Gemski. 1990. Oligonucleotide probes for detection and differentiation of Staphylococcus aureus strains containing genes for enterotoxins A, B, and C and toxic shock syndrome toxin 1. J. Clin. Microbiol. 28:1514–1518. 43. Parsonnet, J. 1989. Mediators/pathogenesis. Mediators in the pathogenesis of toxic shock syndrome: overview. Rev. Infect. Dis. 11:S263–S269. 44. Peavy, D. L., W. H. Adler, and R. T. Smith. 1970. The mitogenic effects of endotoxin and staphylococcal enterotoxin B on mouse spleen cells and human peripheral lymphocytes. J. Immunol. 105:1453–1458. 45. Pontzer, C. H., J. K. Russell, and H. M. Johnson. 1989. Localization of an immune functional site on staphylococcal enterotoxin A using the synthetic peptide approach. J. Immunol. 143:280–284. 46. Prasad, G. S., C. A. Earhart, D. L. Murray, R. P. Novick, P. M. Schlievert, and D. H. Ohlendorf. 1993. Structure of toxic shock syndrome toxin 1. Biochemistry 32:13761–13766. 47. Ranelli, D. M., C. L. Jones, M. B. Johns, G. J. Mussey, and S. A. Kahn. 1985. Molecular cloning of staphylococcal enterotoxin B gene in Escherichia coli and Staphylococcus aureus. Proc. Natl. Acad. Sci. USA 82:5850–5854. 48. Reck, B., P. H. Scheuber, W. Londong, B. Sailer-Kramer, K. Bartxch, and D. K. Hammer. 1988. Protection against the staphylococcal enterotoxininduced intestinal disorder in the monkey by antiidiotypic antibodies. Proc. Natl. Acad. Sci. USA 85:3170–3174. 49. Schantz, E. J., W. G. Roessler, J. Wagman, L. Spero, D. A. Dunnery, and M. S. Bergdoll. 1965. Purification of staphylococcal enterotoxin B. Biochemistry 4:1011–1016. 50. Schlievert, P. M. 1993. Role of superantigens in human disease. J. Infect. Dis. 167:997–1002. 51. Scholl, P. R., N. Trede, T. A. Chatila, and R. S. Geha. 1993. Role of protein tyrosine phosphorylation in monokine induction by the staphylococcal superantigen toxin shock syndrome toxin-1. J. Immunol. 148:2237–2241. 52. Smith, B. G., and H. M. Johnson. 1975. The effect of staphylococcal enterotoxins on the primary in vitro immune response. J. Immunol. 115:575–578. 53. Spero, L., A. Johnson-Wineger, and J. J. Schmidt. 1988. Enterotoxins of staphylococci, p. 131–163. In M. C. Hardegree and A. T. Tu (ed.), Handbook of natural toxins, vol. 4. Marcel Dekker, Inc., New York. 54. Spero, L., J. F. Metzger, J. R. Warren, and B. Y. Griffin. 1975. Biological activity and complementation of the two peptides of staphylococcal enterotoxin B formed by limited tryptic hydrolysis. J. Biol. Chem. 250:5026–5032. 55. Spero, L., B. A. Morlock, and J. F. Metzger. 1978. On the cross-reactivity of staphylococcal enterotoxins A, B, & C. J. Immunol. 120:86–89. 56. Strickler, M. P., R. J. Neill, M. J. Stone, R. E. Hunt, W. Brinkley, and P. Gemski. 1989. Rapid purification of Staphylococcus aureus enterotoxin B by high-pressure liquid chromatography. J. Clin. Microbiol. 27:1031–1035. 57. Sugiyama, H., E. M. McKissic, Jr., M. S. Bergdoll, and B. Heller. 1964. Enhancement of bacterial endotoxin lethality by staphylococcal enterotoxin. J. Infect. Dis. 114:111–118. 58. Swaminathan, S., W. Furey, J. Pletcher, and M. Sax. 1992. Crystal structure of staphylococcal enterotoxin B, a superantigen. Nature (London) 359:801–806. 59. Swaminathan, S., W. Furey, J. Pletcher, and M. Sax. 1995. Residues defining Vb specificity in staphylococcal enterotoxins. Nature Struct. Biol. 2:680–686. 60. Uchiyama, T., Y. Kamagata, S.-J. Yan, A. Kawachi, H. Fujikawa, and H. Igarashi. 1989. Relative strength of the mitogenic and interleukin-2-productioninducing activities of staphylococcal exotoxins presumed to be causative exotoxins of toxic shock syndrome: toxic shock syndrome toxin-1 and enterotoxins A, B and C to murine and human T cells. Clin. Exp. Immunol. 75:239–244.