Generation of protective immunity by inactivated recombinant staphylococcal enterotoxin B vaccine in nonhuman primates and identification of correlates of immunity

Available online at www.sciencedirect.com R

Clinical Immunology 108 (2003) 51–59

www.elsevier.com/locate/yclim

Generation of protective immunity by inactivated recombinant staphylococcal enterotoxin B vaccine in nonhuman primates and identification of correlates of immunity James W. Boles, M. Louise M. Pitt, Ross D. LeClaire, Paul H. Gibbs, Edna Torres, Beverly Dyas, Robert G. Ulrich, and Sina Bavari* United States Army Medical Research Institute of Infectious Diseases, Frederick, MD 21702, USA Received 19 April 2002; accepted with revision 13 March 2003

Abstract At this time there are no vaccines or therapeutics to protect against staphylococcal enterotoxin B (SEB) exposure. Here, we report vaccine efficacy of an attenuated SEB in a nonhuman primate model following lethal aerosol challenge and identify several biomarkers of protective immunity. Initial in vitro results indicated that the mutation of key amino acid residues in the major histocompatibility complex (MHC) class II binding sites of SEB produced a nontoxic form of SEB, which had little to no detectable binding to MHC class II molecules, and lacked T-cell stimulatory activities. When examined in a mouse model, we found that the attenuated SEB retained antigenic structures and elicited protective immune responses against wild-type SEB challenge. Subsequently, a vaccine regimen against SEB in a nonhuman primate model was partially optimized, and investigations of immune biomarkers as indicators of protection were performed. SEB-naı¨ve rhesus monkeys were vaccinated two or three times with 5 or 20 ␮g of the attenuated SEB and challenged by aerosol with wild-type SEB toxin. Unlike exposure to the native toxin, the vaccine did not trigger the release of inflammatory cytokines (TNF␣, IL6, or IFN␥). All rhesus monkeys that developed anti-SEB serum titers ⱖ104 and elicited high levels of neutralizing antibody survived the aerosol challenge. These findings suggest that the attenuated SEB is fully protective against aerosolized toxin when administered to unprimed subjects. Moreover, experiments presented in this study identified various biomarkers that showed substantial promise as correlates of immunity and surrogate endpoints for assessing in vivo biological responses in primates, and possibly in humans, to vaccines against SEs. © 2003 Elsevier Science (USA). All rights reserved. Keywords: Bacterial superantigens; Vaccine; SEB; Biomarkers; Aerosol

Introduction Staphylococcus aureus-produced bacterial superantigens (BSAgs)—staphylococcal enterotoxins (SEs), and toxic shock syndrome toxin 1— belong to a growing family of pyrogenic exotoxins [1,2]. The unprocessed toxins act as ligands for major histocompatibility complex (MHC) class II molecules, and form a ternary complex with the variable ␤ chain of the T-cell receptor (TCR) when presented by antigen-presenting cells (APC). This unusually promiscuous cross-ligation of T cells and APC may allow these * Corresponding author. Fax: ⫹1-301-619-2348. E-mail address: [email protected] (S. Bavari).

virulence determinants to help the bacterium escape cell surface proteins that serve as restricting elements for the cell-mediated immune recognition and trigger the overproduction of pyrogenic and inflammatory cytokines [1–3]. Following BSAg encounter, T cells may become unresponsive to subsequent challenge by the same BSAg, but not by other members of the BSAg family. Furthermore, triggered T cells down-regulate their TCR and may exit the peripheral circulation [4]. As a result these cells would not be readily available to contest viral or other infections. As an opposing hazard, a pool of silent autoreactive T cells could be overstimulated by SAgs and may find inappropriate targets to attack, causing autoimmune disorders [2,5]. More importantly staphylococcal enterotoxin A and B (SEA

1521-6616/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S1521-6616(03)00066-4

52

J.W. Boles et al. / Clinical Immunology 108 (2003) 51–59

and SEB, respectively) release during sepsis or intentional delivery by aerosol causes increased lung endothelial cell injury and enhanced development of pathologies associated with acute respiratory distress syndrome, leading to death [6 –11]. Because inhaled SEs cause remarkable toxicity when delivered via aerosol, SEB is considered a significant threat as a biological weapon and a “dual-threat” agent [12–14]. Consequently, there is a great need to develop protective vaccines and/or therapeutic regimens for use to protect both military and civilian populations. It is now clear that BSAgs bind as unprocessed proteins outside of the peptide-binding groove of MHC class II molecules and in the case of SEB, little to no contact with the bound peptide has been observed [15–18]. Furthermore, the integral binding sites of SEB to MHC class II consist of two structurally conserved surfaces [18]. Based on these observations, and our previous experience, we hypothesized that silenced BSAg proteins, i.e., BSAgs that contain mutations of critical amino acids that participate in BSAg molecular interactions with class II molecules, should provide an excellent avenue to generate efficacious vaccines against BSAg-associated pathology [19]. Previously we showed that SEB vaccine containing three site mutations in hydrophobic binding loop, polar binding pocket, and disulfide loop (L45R, Y89A, and Y94A, respectively) retained its antigenic characteristics [20]. This attenuated SAg-produced anti-SEB antibodies in mice and rhesus monkeys [20]. Here, we further expanded our previous work and showed efficacy of various vaccination regimens of an attenuated SEB mutant protein containing SEB L45R Y89A Y94A in a primate model against aerosolized wildtype (wt) SEB toxin. Additionally, in this study, we show a clear correlation between antibody titers and neutralizing antibody and survival in rhesus monkeys. These studies have identified immune-correlate biomarkers in primates that could be measured in the human population, thereby predicting not only the vaccine efficacy, but also pointing to the human dose schedules that may be most effective.

Materials and methods Bacterial superantigens and lipopolysaccharides (LPS) SEA, SEB, SEC1, and TSST1 were purchased from Toxin Technology (Sarasota, FL). Each toxin was judged to be greater than 95% pure by electrophoresis on 5–20% gradient sodium dodecyl sulfate (SDS)–polyacrylamide gels. The toxins were prepared in phosphate-buffered saline (PBS; 140 mM NaCl, 50 mM Na2H2PO3, pH 7.4). Escherichia coli 055:B5-derived LPS was obtained from Difco Laboratories (Detroit, MI) and reconstituted with PBS. Aliquots were stored at ⫺70°C for future use.

Animals Pathogen-free BALB/c mice, 10 to 12 weeks old, were obtained from Charles River (NCI-Frederick, Frederick, MD). Mice were maintained under pathogen-free conditions and fed laboratory chow and water ad libitum. Rhesus monkeys (Macaca mulatta), weighing 4 – 8 kg, had full access to filtered tap water ad libitum, and were fed approved commercially available food, including fresh fruits. All animal manipulations were performed under anesthetic dose (3– 6 mg/kg intramuscularly) of Telazole. Research was conducted in compliance with the Animal Welfare Act and other federal statutes and regulations relating to animals and experiments involving animals and adhered to principles stated in the Guide for the Care and Use of Laboratory Animals, National Research Council, 1996. The facility where this research was conducted is fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. Attenuated SEB production Engineered recombinant SEB containing three engineered site mutations (SEB L45R/Y89A/Y94A) was cloned and expressed in our laboratory as described elsewhere [21]. Briefly, recombinant SEB L45R/Y89A/Y94A (referred to as rSEB) was partially purified by tangential flow filtration and differential ammonium sulfate precipitation. The rSEB was further purified using hydrophobic interaction, cation exchange, and size exclusion chromatography. The final product was over 95% pure, as determined by SDS–polyacrylamide gel electrophoresis (SDS–PAGE), high-pressure size exclusion chromatography, and capillary zonal electrophoresis. The bacterial toxins and the rSEB contained less than 1 endotoxin units/500 ␮l, as determined by limulus lysate assay. In vitro characterization of rSEB protein: binding to MHC class II receptors and T-cell assays The binding assay was performed as previously described [22]. Briefly, cells from human B lymphoblastoid cell line LG2 (5 ⫻ 105 cells) were incubated (30 min at 37°C) with various dilutions of rSEB or wild-type SEB and the unbound proteins were removed by several washes. The bound SEB was detected with 5 ␮g of affinity-purified anti-SEB antibody. The cells were washed and the bound antibody was detected with fluorescein isothiocyanate (FITC)-labeled goat anti-rabbit IgG and the cells were analyzed by a FACSort flow cytometer (Becton Dickinson, Mountain View, CA). For T-cell proliferation assays, mouse splenic (5 ⫻ 105 cells) and human or rhesus monkey peripheral blood mononuclear cells (1 ⫻ 105 cells) were resuspended in medium

J.W. Boles et al. / Clinical Immunology 108 (2003) 51–59

containing 5% fetal bovine serum (FBS), and 100 ␮l of the cell suspension was added to triplicate wells of 96-well flat bottom plates in the presence of different concentrations of SEA, SEB, SEC1, TSST-1, or rSEB. The cultures were pulsed (18 h) after a 72-h incubation with 1 ␮Ci/well of [3H]thymidine (Amersham, Arlington Heights, IL) and incorporated radioactivity was measured by liquid scintillation. Cytokine assay Purified peripheral blood mononuclear cells from healthy individuals were cultured in presence of rSEB or wt SEB for 20 h. Cell-free supernatants were harvested and levels of TNF␣ and IFN␥ were determined by ELISA using commercially available kits (R&D systems, Minneapolis, MN). Mice were injected intraperitoneally (ip) with wt SEB or rSEB and were given a potentiating dose of LPS. Serum was collected at 4 and 20 h after LPS injection. Sera from five treated mice were pooled and serum-borne cytokines were measured by ELISA (Quantikine Murine Immunoassay kits, R&D systems), according to the manufacturer’s specifications. In vitro and in vivo passive protection assay For in vitro passive transfer studies, we used the Tlymphocyte proliferation assay. Briefly, blood was obtained from an unvaccinated donor rhesus monkey and mononuclear cells were isolated by buoyant density centrifugation (Ficoll-Paque, Amersham Pharmacia Biotech AB) and washed three times. The cells were resuspended in medium containing 5% FBS, and 100 ␮l (1 ⫻ 105 cells) of the cell suspension was added to triplicate wells of 96-well flatbottom plates in presence of 5% rhesus monkey serum. The cells were incubated (37°C, 5% CO2) for 1 h and then cultured (37°C, 5% CO2) with graded concentrations of SEB for 3 days. The cultures were pulsed (18 h) with 1 ␮Ci/well of [3H]thymidine (Amersham) and incorporated radioactivity was measured by liquid scintillation. All cultures were performed in triplicate. The data are presented as percentage of SEB stimulation, where inhibition ⫽ 100 ⫺ [(experimental counts per minute for serum from 6 weeks after the last vaccination)/(serum before vaccination counts per minute) ⫻ 100]. For in vivo passive transfer studies, 50 LD50 (⬇10.0 ␮g/mouse) of SEB was incubated with 200 ␮l of serum obtained from vaccinated monkeys or 200 ␮l of PBS at 37°C for 30 min [23]. Mice were injected ip with the mixture and 3 h later with 75 ␮g of LPS, as previously described [19,23]. Deaths were recorded after 3– 4 days. Challenge controls were mice injected with either LPS or SEB only (no deaths were observed).

53

Vaccination and challenge protocol rSEB was diluted in 50 mM glycine, pH 8.5/0.9% NaCl formulated with Alhydrogel adjuvant at the ratio of 7:1 (weight/weight, adjuvant:vaccine). For mouse vaccination protocol, mice were injected intramuscularly (im) with 0.5, 5, or 20 ␮g of rSEB in 200 ␮l of Alhydrogel adjuvant (or with adjuvant alone) and boosted at 14 and 28 days in the same manner as described for the primary injection. Serum antibody titers were determined as described elsewhere [19]. Mice were challenged ip 2 weeks after the second boost with 2 ␮g, approximately 10 LD50 of wild-type SEB and 3 h later with 75 ␮g of LPS, as previously described [19,23]. Deaths were recorded after 3– 4 days. Challenge controls were aged-matched mice injected with either LPS or SEB only (no deaths were observed). For rhesus monkeys vaccination protocol, 2 weeks before rSEB injection, rhesus monkeys were bled, and their serum antibody titers against SEs and TSST-1 were determined. Monkeys that had undetectable (⬍1:50) serum titers against the BSAgs were used in this study. rSEB was diluted in 50 mM glycine, pH 8.5/0.9% NaCl formulated with Alhydrogel adjuvant at the ratio of 7:1 (weight/weight, adjuvant:vaccine). Rhesus monkeys were injected im two or three times, at 4-week intervals, with rSEB 5 or 20 ␮g adsorbed onto 0.5 ml of aluminum hydroxide suspension (Alhydrogel). The animals were challenged with aerosolized SEB 6 weeks after the last injection of rSEB or adjuvant. Adjuvant control monkeys were injected twice or three times with 0.5 ml of Alhydrogel. Anesthetized monkeys were exposed to an aerosol generated by a Collison nebulizer in a head-only chamber containing approximately 75 lethal concentration (LCT)50 of wild-type SEB 6 weeks after the last vaccination. Aerosol concentration was determined by impinging a known volume of the aerosol during the exposure and determining the protein concentration (Micro BCA) based on a SEB standard curve. Dose was determined by the volume of aerosol individual monkeys inhaled by indirect plethysmography (Buxco, Sharon, CT) and calculated exposure based on aerosol concentration, respired volume, exposure time, and body weight.

Results Evaluation of rSEB binding to MHC class II, T-cell stimulation, and cytokine induction We examined cell surface binding of rSEB to MHC class II of human lymphoblastoid B cell line LG2 by using flow cytometery, and compared to wt SEB. As seen in Fig. 1A, substantial SEB binding to LG2 cells was detected at 10 ␮g/ml after a short incubation with the wt toxin. Higher binding of SEB was observed at 250 ␮g/ml. By comparison, there was no detectable binding of rSEB to LG2 cells greater than the control values.

54

J.W. Boles et al. / Clinical Immunology 108 (2003) 51–59

Fig. 1. (A) Binding of wt SEB (filled) and rSEB (unfilled) to human MHC class II. Different concentrations of wt SEB or rSEB were incubated with LG2 cells. Cell surface-bound mutant or toxin was detected with affinity-purified rabbit anti-SEB antibodies. The complexes were detected with goat anti-rabbit labeled FITC and the samples were analyzed by FACS analysis. (B) Human T-cell responses to wt SEB and rSEB. Human peripheral blood mononuclear cells were incubated for 3 days in triplicate wells in 96-well plates containing varying concentrations of SEB or rSEB. The cells were incubated for an additional 12 h with [3H] thymidine and incorporated radioactivity was measured using a scintillation counter. (C) rSEB lacks capacity to induce cytokine production. Human peripheral blood mononuclear cells were incubated with 100 ng/ml of SEB toxin or rSEB for 20 h. Cell-free supernatants were used to measure TNF␣ and IFN␥ by ELISA. These experiments were repeated three times using three different donors with similar results. Results represent the mean (⫾SD) of triplicate determination.

As a second measure of rSEB attenuation, the effect of different concentrations of recombinant protein on the proliferative responses of human mononuclear cells was examined (Fig. 1B). The cells were exposed to SEB or rSEB for 72 h and T-cell proliferation was measured. Maximal response against wt SEB was observed at 0.1 and 1 ␮g/ml. Compared to wt SEB, rSEB induced little to no proliferation at these doses and only minimal activity was observed at 10 ␮g/ml. Therefore, both lack of MHC class II binding and attenuation of T-cell responses by rSEB suggested that SAg activity of rSEB had been substantially reduced, and that rSEB was was no longer a SAg. Next, we examined the in vitro cytokine responses of human mononuclear cells to rSEB and wt SEB toxin (Fig. 1C). SEB at 100 ng/ml induced massive release of proinflammatory cytokines TNF␣ and IFN␥. rSEB, on the other hand, failed to induce detectable enhancement of cytokine production over the medium control. Similarly, rSEB was also inactive when it was added to mononuclear cells from

rhesus monkey (data not shown). These data clearly demonstrated that attenuation of SEB SAg activity, via specific site-directed mutageneses, resulted in drastic reduction of human T-cell and APC responses. Toxicity and vaccine efficacy of rSEB in a mouse model As in vitro studies indicated that rSEB was nonsuperantigenic, its biological effects in vivo were examined. Mice (n ⫽ 10/group) were injected ip with wt SEB (0.2, 0.5, 1, 2, or 20 ␮g/mouse) or rSEB (2 or 20 ␮g/mouse), and a potentiating dose of LPS, and lethality was scored (Fig. 2). All mice injected with 1, 2, and 20 ␮g of wt SEB succumbed to the lethal effects of the toxin within 36 h after administration, while 0.5 and 0.2 ␮g of SEB induced lethality in 70 and 40% of mice, respectively. In contrast to wt SEB, none of the mice injected with either 2 or 20 ␮g of rSEB died after 72 h. In fact, these mice were observed for an additional 21 days, and during this time, there was no additional

J.W. Boles et al. / Clinical Immunology 108 (2003) 51–59

55

sure. Survival (Fig. 3, filled symbols) was 40% for mice vaccinated with 0.5 ␮g of rSEB, 80% for those vaccinated with 5 ␮g rSEB, and 100% for mice injected with 20 ␮g rSEB. Interestingly, mice that had low titers against SEB succumbed to the lethal effects of the toxin. Mice that developed titers ⱖ104 survived the challenge and approximately only half of mice that developed titers below 104 survived. These data suggest that in a mouse model, there is a positive correlation between antitoxin titers and survival, and that the levels of anti-SEB titers may give us a substantial guidance and readout for efficacy of rSEB vaccine. rSEB vaccine efficacy in nonhuman primate model Fig. 2. Biological effects of rSEB. Eight to ten-week-old female Balb/C mice (groups of 10) were injected ip with varying doses of SEB or rSEB and 4 h later with a potentiating dose of LPS (75 ␮g). Kaplan–Meier estimates of mean survival times indicates statistical differences between wt SEB from the rSEB doses (P ⫽ 0.009 for wt doses of 0.2 and 0.5 ␮g/mouse, and P ⫽ ⬍0.0001 for wt doses of 1, 2, and 20 ␮g/mouse).

mortality in either of these experimental groups. Overall, mice injected with rSEB displayed no visible signs of morbidity, and no visible differences were observed between rSEB-injected mice and controls. In a parallel experiment, mice were injected with 2 ␮g of wt SEB or 20 ␮g of rSEB, and cytokine levels of TNF-␣, IFN-␥, and IL-6 were examined. As expected, mice injected with wt SEB showed a surge in TNF-␣, IFN-␥, and IL-6 release, while those injected with rSEB displayed no elevation in these cytokines when compared to controls (data not shown). Next, mice in groups of 10 were injected three times on Days 0, 14, and 28 with 0.5, 5, or 20 ␮g of the rSEB or adjuvant alone (as a control). Serum titers against SEB were determined on Day 38 using ELISA (Fig. 3) [19]. All mice vaccinated with rSEB developed detectable titers against wt SEB. In 8 of 10 mice, 0.5 ␮g of rSEB vaccine elicited antibody titers against SEB exceeding 103. In the same group, 2 mice had titers of 103. All mice vaccinated with 5 and 20 ␮g of rSEB had titers that were ⱖ5 ⫻ 103. In the 5 ␮g rSEB-vaccinated group, 3 of the 10 mice developed antibody titers which reached 104. Mice vaccinated with three doses of 20 ␮g of rSEB developed the highest titers, and 10 of 10 of the vaccinees in this group had titers that reached 105. Mice injected with the adjuvant only had no detectable increases of titers against SEB. These data indicated that mice could mount a significant humoral immune response against the attenuated bacterial SAg vaccine and there was a direct relationship between rSEB dose and the titers against wt SEB. In order to determine whether the vaccination elicited serum antibodies that were capable of neutralizing the wildtype SEB, mice were challenged on Day 42 with approximately 10 LD50 (2 ␮g) of wt SEB and were observed for 14 days (Fig. 3). All mice injected with adjuvant only died (open symbols) within the first 24 –30 h of wt SEB expo-

Rhesus monkeys were chosen for rSEB vaccine efficacy, as they have been used extensively to dissect the pathophysiological actions of BSAgs, and because this animal exhibits immunological responses to SEB toxicities similar to those of humans [6,7]. The rSEB vaccine was evaluated for protection against aerosolized wt SEB, as this would be an optimal form of delivery as a biological weapon and poses even more rigorous test of efficacy than natural forms of exposure. Six weeks after the last rSEB or adjuvant-only vaccination, rhesus monkeys were challenged with a lethal aerosol concentration dose of approximately 75 LCT50 of SEB. These animals were examined postvaccination individually for (1) survival after challenge; (2) antibody titers against wt SEB; (3) ability of serum taken at the time of challenge to inhibit SEB-induced proliferation in rhesus T cells in vitro; and (4) ability of serum to passively protect mice from lethal SEB challenge. Maximum protection was observed in animals injected with three doses of 20 ␮g

Fig. 3. Immunogenicity and vaccine potential of rSEB in mice. Eight to ten-week-old female Balb/C mice (groups of 10) were vaccinated with 0.5, 5, or 20 ␮g of rSEB, adsorbed to adjuvant or adjuvant only, three times (0,14, and 28 days). Ten days after the last boost, mice were bled, and end-point titers (reciprocal serum dilution resulting in OD reading twice above the negative) were determined by ELISA. Controls were ELISA wells containing either no toxin or no serum. On Day 42, mice were challenged with 10 LD50 of wt SEB and 3 days later lethality was scored. Each symbol represents a single mouse. Filled symbols represent live and unfilled symbol represent dead mice. Probit analysis of survival rates indicates statistical association with dose (P ⫽ 0.002), with ED50 ⫽ 2.6 ␮g rSEB (1.2, 4.9; 95% CL).

56

J.W. Boles et al. / Clinical Immunology 108 (2003) 51–59

Table 1 Attenuated rSEB protein protects rhesus monkeys against aerosol challenge of SEB Vaccinea

Survival (live/total)b

Adjuvant 5 ␮g given twice 5 ␮g given three times 20 ␮g given twice 20 ␮g given three times

0/2 5/8 5/8 5/8 8/8

a Rhesus monkeys were injected im with 5 or 20 ␮g rSEB vaccine in Alhydrogel adjuvant at the ratio of 7:1 (weight/weight, adjuvant:vaccine) two or three times 4 weeks apart. b Rhesus monkeys were challenged 6 weeks after the last vaccination by aerosol with 75 LCT50 of SEB, P ⫽ 0.20, comparing active treatments with high dose by Fishers exact tests.

(Table 1). All rhesus monkeys vaccinated with three doses of 20 ␮g of rSEB survived. Monkeys receiving two doses of 20 ␮g twice, 5 ␮g twice, or 5 ␮g three times were only partially protected. Correlates of protection Serum samples obtained on the day of SEB challenge from all vaccinated rhesus monkeys and were examined to determine antibody titers in vitro and in vivo neutralization activity. Among all groups of vaccinees, rhesus monkeys vaccinated with three doses of 20 ␮g of rSEB developed the highest antibody titers in serum, and neutralized T-cell responses against wt SEB (Fig. 4). In this group, six of eight vaccinees produced endpoint anti-SEB titers reaching 104, while two of the eight vaccinated animals produced titers that exceeded 104 (Fig. 4A). Other vaccine dose regimens

elicited antibody titers that were substantially below those elicited by three doses of 20 ␮g of rSEB. In a parallel experiment, the ability of the same serum to inhibit the mitogenic effects of SEB on donor rhesus T cells was tested (Fig. 4B). Serum from animals receiving three doses of 20 ␮g of rSEB inhibited SEB-induced T-cell activation by ⬎95%. Serum titers and neutralizing ability of the serum obtained from animals vaccinated with other three regimens were significantly lower and varied substantially within each group. For example, only two of eight rhesus monkeys vaccinated with two doses of 20 ␮g of rSEB had serum anti-SEB titers reaching 104. The other six animals in this group had titers below 104. When serum from this group of vaccinees was mixed with SEB, we observed proliferative responses against the toxin that were substantially unfocused and varied from 10 to 90% inhibition. Similarly, serum obtained from rhesus monkeys vaccinated with 5 ␮g either two or three times had low antibody titers and lacked good neutralizing activity. In these groups, 50% of the vaccinated monkeys had serum titers below 104 and displayed neutralizing titers varying from 3 to 97%. Rhesus monkeys given adjuvant only did not develop antiSEB titers and their serum lacked any in vitro neutralizing activity. All rhesus monkeys that developed T-cell neutralizing titers of ⬎80% survived the aerosol SEB challenge. The in vivo neutralizing activity of serum obtained from vaccinated monkeys on the day of challenge was also tested (Table 2). Serum from vaccinees was incubated with a lethal dose of SEB (10 ␮g), and this mixture was subsequently injected into SEB-naı¨ve mice. Serum from adjuvant vaccinated monkeys failed to protect mice against SEB in vivo; all of these mice died when challenged with SEB. By comparison, all mice (24/24, survival/total) survived when

Fig. 4. rSEB vaccine evoked high protective antibody titers in rhesus monkeys. Rhesus monkeys were vaccinated im with either 5 or 20 ␮g of rSEB in Alhydrogel adjuvant; two or three times; 4 weeks a part. Control rhesus monkeys were given Alhydrogel only. Monkeys were bled 6 weeks after the last vaccination and were exposed to 75 LCT50 of wt SEB delivered by aerosol. Each symbol represents an individual rhesus monkey. Filled symbols denote live and unfilled symbols represent dead animals. (A) End-point serum titers against wt SEB (reciprocal serum dilution resulting in OD reading twice above the negative controls) were determined by ELISA. Controls were ELISA wells containing either no toxin or no serum. Fisher exact test comparing survival rates for serum dilution exceeding 104, P ⬍ 0.0001. (B) The ability of serum taken from the vaccinated monkeys at the time of challenge to inhibit SEB-induced T-cell proliferation of SEB-naı¨ve rhesus monkey T cells was measured. Fisher’s exact test comparing survival rates for percentage inhibition of T-cell stimulation exceeding 80%, P ⬍ 0.001.

J.W. Boles et al. / Clinical Immunology 108 (2003) 51–59 Table 2 Passive transfer of rhesus monkey serum raised against rSEB protects against SEB-induced lethality Conditions

Passive protection of mice live/total

P value compared to (rSEB vaccination) 20 ␮g given three times

PBS Adjuvant 5 ␮g given twice 5 ␮g given three times 20 ␮g given twice 20 ␮g given three times

0/10 0/10 10/24 17/24 15/24 24/24

⬍0.001 ⬍0.001 ⬍0.001 0.01 0.02 —

Note. Pooled sera from vaccinated or control rhesus monkeys and 10 ␮g (approximately 50 LD50) of SEB were incubated for 1 h at 37°C. Mice were given the mixture, and then they were injected with a potentiating dose of LPS (75 ␮g). Lethality was recorded 4 days after the challenge dose. (All treatments were statistically less protective than 20 ␮g given three times and the three active treatments with partial protection were not statistically different.)

passively treated with serum obtained from rhesus monkeys previously vaccinated with three doses of 20 ␮g rSEB. Serum obtained from other vaccine regimens produced approximately 40 –70% protection against SEB in vivo. In these groups, 5 ␮g given three times protected 17/24 mice, while 10/24 and 15/24 were protected when serum samples from rhesus monkeys vaccinated twice with 5 or 20 ␮g were tested, respectively. These results suggest that antibody titers and in vitro and in vivo neutralization data may be used to provide a surrogate endpoint of human clinical efficacy of rSEB vaccine.

Conclusions SEs are believed to be part of the biological weapons arsenals of certain countries, and to pose a real bioterrorism threat. At present there are no licensed vaccines available to protect humans against the toxic effects of BSAgs, such as SEA and SEB. We showed previously that inactivated BSAgs may be used as vaccine candidates (19). In those studies, the approach to producing an effective and safe SAg-vaccine candidate centered on designing attenuated nonsuperantigenic proteins with the ability to retain immunogenicity and protection against wt SAg. We determined that analogues of SEA, in which the MHC binding sites were silenced by site-directed mutagenesis, retained immunogenicity, and vaccinated mice were fully protected against SEA challenge [19]. In contrast, mutagenesis of the TCR binding face of SEA produced a construct that contained some residual in vitro and in vivo toxicity. Based on these studies, we hypothesized that attenuated SAg that did not bind MHC class II, or triggered T-cell responses and protected T cells from activation by a mechanism that relied on elicitation of neutralizing antibodies, would be an excellent vaccine candidate.

57

Studies presented in this article were prompted by the observation that most BSAgs use common structural strategies for binding to the ␣ chain of MHC class II receptors [1–3]. A leucine residue (SEB L45) within a hydrophobic loop has been identified as one of the key residues that control BSAg interaction with class II molecules [17,18,24]. Mutation of SEB L45, or corresponding residues in other SAgs, substantially reduces SEB binding and decreases the ability of the BSAg to activate T cells [20,25]. To further suppress the activity of SEB, and for added safety, we generated an analogue of SEB with additional amino acid residue mutations in a disulfide loop and polar binding pocket. This approach further reduced the superantigenic characteristics of SEB (data not shown). The goal of this study was to further examine the suitability of an attenuated SEB protein for vaccine purposes and examine it in a mouse model and in a more relevant primate model. Before exploring the vaccine potential of the rSEB, we sought to examine the effects of the mutations on binding to MHC class II and T-cell activation. Results from these studies indicated that the nonsuperantigen analogue of SEB did not bind MHC class II receptors, and lacked T-cell stimulatory activity. Furthermore, the rSEB was not toxic in mice, and vaccinated mice developed robust protective immune response against SEB challenge. Humans and other primates are more sensitive to bacterial SAgs than most other species because of their greater MHC class II binding affinity to these virulence factors. Among the primates that we tested, rhesus monkey lymphocytes were shown to respond to the lowest concentrations of BSAgs [26]. Because of this level of sensitivity, this animal was chosen as the most relevant model to examine therapeutics and vaccines against BSAgs. After rSEB vaccination in primates, there were no inflammatory cytokines or thermogenic responses detected (data not shown). However, when different rSEB vaccine regimes were compared for their ability to produce neutralizing antibody, fluctuations were observed. All rhesus monkeys vaccinated with three doses of 20 ␮g of rSEB generated high neutralizing antibody responses that protected T cells from the mitogenic effects of SEB. It is noteworthy that the same vaccine regime also protected mice from SEB-induced death. The other goals of this work were to identify correlates of immunity for protection against bacterial SAg. Recently, we showed a clear correlation between human antibody titers and the inhibition of T-cell response to BASgs, and used a potentiated mouse model to demonstrate that high-titer pooled human sera could protect mice against SEB challenge [27]. Results from these previous studies clearly indicate that anti-SEB antibody is the sole protective factor in the pooled sera [27]. In this study, we extended those observations and showed a clear correlation between antiSEB titers and survival when mice or rhesus monkeys were challenged by SEB toxin. All mice and rhesus monkeys, which elicited anti-SEB titers exceeding 104, survived SEB

58

J.W. Boles et al. / Clinical Immunology 108 (2003) 51–59

challenge, while titers below 104 were only partially protected. It is worth noting that this 104 titer was of a level reported in SEB-immune humans [27]. Although in this study we did not examine the Ig isotype after rSEB vaccination, previously Woody et al. reported high levels of IgG2a and IgG2b following vaccination with L45R in mice [25]. The lymphocyte stimulation provides another functional bioassay for measuring the neutralizing ability of anti-SEB antibodies, and could provide a key biomarker for assessing biological responses in humans. We noted a direct correlation between high titers (above 104) and the ability of the same serum to protect donor T cells from SAg-induced activation. Vaccinees survived SEB challenge when donor serum neutralized SEB-induced T-cell activation by ⬎80%. Only passive transfer of sera from rhesus monkeys vaccinated with three doses of 20 ␮g of rSEB was 100% protective in mice, while other rSEB vaccine regimens showed variable protective effects. These findings clearly show a direct correlation between vaccine dose in primates, serum antibody titers, and protection. Therefore, our data point to serum antibody as the main protective factor against BSAgs. In summary, the status of SEs as biowarfare agents and increasing civilian concerns over drug-resistant toxicogenic strains of these pathogens have accelerated the need for effective vaccines and therapeutics against these infections. Our studies focused on a nonsuperantgenic form of SEB for vaccine purposes in primates, and used serum antibody to determine if differences existed between dosing regimens. Our results indicate that serum antibody titers and neutralization could be used as correlates of protection in primates and most likely in humans. In this study, naı¨ve rhesus monkeys were used for the rSEB vaccine trials, and it was shown that vaccination with the attenuated enterotoxin could completely protect them against SEB challenge. Because most humans have already been exposed to SEs, rSEB vaccine would most likely boost some levels of preexisting immunity. This prior exposure should enhance vaccineinduced immunity, and perhaps lower the dose and dosing regimens needed to protect humans. It is important that low antibody titers to BSAgs have been associated with the recurrence of toxic shock syndrome and that this population may benefit from vaccination with anti-SAg vaccine.

References [1] J. Fraser, V. Arcus, P. Kong, E. Baker, T. Proft, Superantigens— powerful modifiers of the immune system, Mol. Med. Today 6 (2000) 125–132. [2] R.G. Ulrich, S. Bavari, M.A. Olson, Bacterial superantigens in human disease: structure, function and diversity, Trends Microbiol. 3 (1995) 463– 468. [3] D.L. Stevens, Invasive streptococcal infections, J. Infect. Chemother. 7 (2001) 69 – 80.

[4] F. Niedergang, A. Hemar, C.R. Hewitt, M.J. Owen, A. Dautry-Varsat, A. Alcover, The Staphylococcus aureus enterotoxin B superantigen induces specific T cell receptor down-regulation by increasing its internalization, J. Biol. Chem. 270 (1995) 12839 –12845. [5] M.M. Dinges, P.M. Orwin, P.M. Schlievert, Exotoxins of Staphylococcus aureus, Clin. Microbiol. Rev. 13 (2000) 16 –34. [6] C.F. Weng, J.L. Komisar, R.E. Hunt, A.J. Johnson, M.L. Pitt, D.L. Ruble, J. Tseng, Immediate responses of leukocytes, cytokines and glucocorticoid hormones in the blood circulation of monkeys following challenge with aerosolized staphylococcal enterotoxin B, Int. Immunol. 9 (1997) 1825–1836. [7] J. Tseng, J.L. Komisar, R.N. Trout, R.E. Hunt, J.Y. Chen, A.J. Johnson, L. Pitt, D.L. Ruble, Humoral immunity to aerosolized staphylococcal enterotoxin B (SEB), a superantigen, in monkeys vaccinated with SEB toxoid-containing microspheres, Infect. Immun. 63 (1995) 2880 –2885. [8] B.T. Peterson, E.J. Miller, D. Morris, Neutrophil influx and migration in rabbit airways in response to staphylococcal enterotoxin-A, Exp. Lung Res. 25 (1999) 41–54. [9] N. Fujisawa, S. Hayashi, A. Kurdowska, J.M. Noble, K. Naitoh, E.J. Miller, Staphylococcal enterotoxin A-induced injury of human lung endothelial cells and IL-8 accumulation are mediated by TNF-alpha, J. Immunol. 161 (1998) 5627–5632. [10] E.J. Miller, A.B. Cohen, B.T. Peterson, Peptide inhibitor of interleukin-8 (IL-8) reduces staphylococcal enterotoxin-A (SEA) induced neutrophil trafficking to the lung, Inflamm. Res. 45 (1996) 393–397. [11] E.J. Miller, S. Nagao, F.K. Carr, J.M. Noble, A.B. Cohen, Interleukin-8 (IL-8) is a major neutrophil chemotaxin from human alveolar macrophages stimulated with staphylococcal enterotoxin A (SEA), Inflamm. Res. 45 (1996) 386 –392. [12] J.M. Madsen, Toxins as weapons of mass destruction. A comparison and contrast with biological-warfare and chemical-warfare agents, Clin. Lab. Med. 21 (2001) 593– 605. [13] D.R. Franz, P.B. Jahrling, A.M. Friedlander, D.J. McClain, D.L. Hoover, W.R. Bryne, J.A. Pavlin, G.W. Christopher, E.M. Eitzen Jr., Clinical recognition and management of patients exposed to biological warfare agents, JAMA 278 (1997) 399 – 411. [14] D.R. Franz, P.B. Jahrling, D.J. McClain, D.L. Hoover, W.R. Byrne, J.A. Pavlin, G.W. Christopher, T.J. Cieslak, A.M. Friedlander, E.M. Eitzen Jr., Clinical recognition and management of patients exposed to biological warfare agents, Clin. Lab. Med. 21 (2001) 435– 473. [15] S. Swaminathan, W. Furey, J. Pletcher, M. Sax, Crystal structure of staphylococcal enterotoxin B, a superantigen, Nature 359 (1992) 801– 806. [16] A.C. Papageorgiou, H.S. Tranter, K.R. Acharya, Crystal structure of microbial superantigen staphylococcal enterotoxin B at 1.5 A resolution: implications for superantigen recognition by MHC class II molecules and T-cell receptors, J. Mol. Biol. 277 (1998) 61–79. [17] J. Kim, R.G. Urban, J.L. Strominger, D.C. Wiley, Toxic shock syndrome toxin-1 complexed with a class II major histocompatibility molecule HLA-DR1, Science 266 (1994) 1870 –1874. [18] T.S. Jardetzky, J.H. Brown, J.C. Gorga, L.J. Stern, R.G. Urban, Y.I. Chi, C. Stauffacher, J.L. Strominger, D.C. Wiley, Three-dimensional structure of a human class II histocompatibility molecule complexed with superantigen, Nature 368 (1994) 711–718. [19] S. Bavari, B. Dyas, R.G. Ulrich, Superantigen vaccines: a comparative study of genetically attenuated receptor-binding mutants of staphylococcal enterotoxin A, J. Infect. Dis. 174 (1996) 338 –345. [20] R.G. Ulrich, M.A. Olson, S. Bavari, Development of engineered vaccines effective against structurally related bacterial superantigens, Vaccine 16 (1998) 1857–1864. [21] J.D. Coffman, J. Zhu, J.M. Roach, S. Bavari, R.G. Ulrich, S.L. Giardina, Production and purification of a recombinant Staphylococcal enterotoxin B vaccine candidate expressed in Escherichia coli, Protein Expr. Purif. 24 (2002) 302–312. [22] R.G. Ulrich, S. Bavari, M.A. Olson, Staphylococcal enterotoxins A and B share a common structural motif for binding class II major

J.W. Boles et al. / Clinical Immunology 108 (2003) 51–59 histocompatibility complex molecules, Nat. Struct. Biol. 2 (1995) 554 –560. [23] S. Bavari, R.G. Ulrich, R.D. LeClaire, Cross-reactive antibodies prevent the lethal effects of Staphylococcus aureus superantigens, J. Infect. Dis. 180 (1999) 1365–1369. [24] A.C. Papageorgiou, K.R. Acharya, Microbial superantigens: from structure to function, Trends Microbiol. 8 (2000) 369 –375. [25] M.A. Woody, T. Krakauer, R.G. Ulrich, B.G. Stiles, Differential immune responses to staphylococcal enterotoxin B mutations in a

59

hydrophobic loop dominating the interface with major histocompatibility complex class II receptors, J. Infect. Dis. 177 (1998) 1013–1022. [26] S. Bavari, R.E. Hunt, R.G. Ulrich, Divergence of human and nonhuman primate lymphocyte responses to bacterial superantigens, Clin. Immunol. Immunopathol. 76 (1995) 248 –254. [27] R.D. LeClaire, S. Bavari, Human antibodies to bacterial superantigens and their ability to inhibit T-cell activation and lethality, Antimicrob. Agents Chemother. 45 (2001) 460 – 463.