Human Leukocyte Antigen-DQ8 Transgenic Mice: a Model To Examine the Toxicity of Aerosolized Staphylococcal Enterotoxin B

Experimental design.To determine whether transgenic mice were also susceptible to aerosols of SEB, groups of HLA-DQ8 mice were exposed to 5, 30, 60, 120, 240, or 310 μg of aerosolized SEB toxin per kg. Because BALB/c mice have been widely used to examine the pathological effects of BSAgs, in our initial experiments we incorporated these mice as controls. Groups of BALB/c or HLA-DR2β/IEα mice were also examined at the two highest doses for comparative purposes (Table 1). The controls received SEB only, received LPS only, or were given 75 μg of LPS intraperitoneally 3 h after exposure to aerosolized SEB. All mice were monitored for 21 days after SEB challenge. Thereafter, to further examine mechanisms of SEB-induced lung pathologies and death, HLA-DQ8 mice were exposed by aerosol to 120 μg of SEB per kg and were killed at 4, 10, and 24 h postexposure. At the indicated times, lungs, spleens, and blood samples were collected for determination of gene expression and secretion of proinflammatory cytokines. Full pathology examination was also performed on a subgroup of these animals.

Animals.Pathogen-free 10- to 12-week-old BALB/c (H-2^d) mice were obtained from Charles River (NCI-Frederick, Frederick, Md.). HLA-DQ8/human CD4⁺ transgenic mice were created by microinjecting DNA fragments into embryos from C57BL/6 mice, as previously described (26, 33, 34). Briefly, insertion of HLA-DQα1*0301 and HLA-DQβ1*0302 gene fragments created the HLA-DQ8⁺ transgenic mice. The C57BL/6 embryos were inserted into (C57BL/6 × DBA/2) F₁ mice and backcrossed to B10.M mice. The HLA-DQ8 molecule was then introduced into murine class II-negative mice by mating the H-2-negative strain (H-2 Ab^o/o) with HLA-DQ8.B10.M mice. Human CD4⁺ Ab° mice were similarly created and subsequently crossed with HLA-DQ8⁺ B10.M mice. HLA-DR2β/IEα mice were created based on the same strategy and were used as littermate controls. All mice had no detectable (<1:50) serum titers against SEA, SEB, and SEC1 as measured by enzyme-linked immunosorbent assay (ELISA).

Mice were maintained under pathogen-free conditions and fed laboratory chow and water ad libitum. 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.

Reagents.Lyophilized SEB toxin was obtained from Porton Down (Wiltshire, United Kingdom). The toxin was over 95% pure as determined by polyacrylamide gel electrophoresis. N-terminal protein sequencing further showed that SEB was the only enterotoxin present in the preparation, and no other SEs or toxic shock syndrome toxin 1 was detected. The toxin was reconstituted before aerosolization with freshly prepared endotoxin-free phosphate-buffered saline (PBS). The SEB stock solution concentration was determined by protein assay (MicroBCA; Pierce, Rockford, Ill.). The endotoxin levels in the SEB stock were less than the detection limit, as determined by the Limulus amebocyte lysate QCL-100 endotoxin assay according to the instructions of the manufacturer (Biowhittaker, Frederick, Md.). Escherichia coli LPS (O55:B5) (Difco Laboratories, Detroit, Mich.) was reconstituted in sterile PBS and stored at −70°C.

Aerosol exposure system and aerosol characterization.Mice were exposed to SEB aerosols by using a whole-body dynamic exposure chamber housed within class III biological safety cabinets maintained under negative pressure (27, 28). Mice were contained in the whole-body chamber by four smaller, stainless-steel mesh cages. Each cage contained up to 10 mice, for a maximum of 40 mice per exposure. The total flow through the chamber was 19.0 ± 0.5 liters/min, and the pressure inside the chamber was maintained equal to that inside the safety cabinet. Aerosols were generated with a Collison nebulizer (BGI USA, Inc., Waltham, Mass.). Particle sizing of the experimental atmosphere generated revealed a mass mean aerodynamic diameter of 1 μm with a geometric standard deviation of 1.4. The exposure concentration was determined by constantly sampling the chamber with an all-glass impinger (Ace Glass, Vineland, N.J.). PBS with antifoam A (0.001%, wt/vol) (Sigma, St. Louis, Mo.) was used as the collection medium in the impinger. Starting solutions and all-glass impinger samples were measured by protein assay (Pierce MicroBCA).

The presented inhalation dose was determined by using the respiratory minute volume (V_m). The estimates were derived by Guyton's formula, expressed as V_m = 2.10 × W_b^0.75, where W_b is body weight in grams (for mice we used the average of group weights on the day of exposure) (27, 28). The presented dose was then calculated by multiplying the total volume (V_t) of experimental atmosphere inhaled by each animal (V_t = V_m× length of exposure) by the empirically determined exposure concentration (C_e) (presented dose = C_e× V_t).

Head-only aerosol challenges of rhesus macaques, performed as part of other studies (8, 9), were done in a manner similar to that for the mice with respect to particle size distribution, sampling, and toxin preparation. Primates received a presented aerosol dose of 5 to 7 50% lethal doses (LD₅₀) of SEB.

Telemetric temperature analysis.Implantable programmable temperature transponders (IPTT-100) were purchased from BioMedic Data System Inc. (Seaford, Del.). Individually sterilized transponder chips were implanted subcutaneously at least 2 weeks before initiation of the experiment. Mice were exposed by aerosol to approximately 120 μg of SEB per kg, and temperature was monitored via telemetry and recorded every hour.

Histopathology.Scheduled necropsies of control or transgenic mice were performed on the mice at 3 days postexposure. The lungs were fixed in 10% neutral-buffered formalin, routinely processed, cut at 5 to 6 μm, and stained with Mayer's hematoxylin and eosin. The lungs from two naive age-, sex-, and strain-matched controls were similarly processed and evaluated for comparative purposes.

Cytokine analysis.To determine cytokine gene expression levels in tissues, total RNA was prepared from the spleen and lung tissues (Qiagen RNA Isolation Midi-Kit). The isolated RNA was reverse transcribed to produce cDNA (Invitrogen, Carlsbad, Calif.). After amplification, PCR products were separated by electrophoresis on a 1.5% agarose gel. In all reverse transcription PCR (RT-PCR) experiments, the β-actin region was amplified and served as an internal control. The protein levels of the cytokines TNF-α, IL-6, and IFN-γ in pooled samples from four to six mice per time point were determined by ELISA according to the manufacturer's specifications (Quantikine murine immunoassay kits; R&D Systems, Minneapolis, Minn.).

Statistics.Survival data were analyzed by the probit method (LD₅₀ estimation) and nonparametric life-table methods by using the Mantel-Cox statistic (BMDP Dynamic, version 7.0; SAS Corp., Cary, N.C.). Descriptive statistics (mean ± standard deviation) were used to display the results of the serum cytokine analysis.

HLA-DQ8 transgenic mice succumb to SEB aerosol exposure.Previously, we reported that human MHC class II/human CD4 transgenic mice deficient in endogenous murine MHC class II and murine CD4 receptors were extremely sensitive to BSAgs, without the administration of potentiating agents (11, 38). In these mice, parenteral challenge with the BSAg caused in vivo proliferative responses, surges of proinflammatory cytokines in serum, expansion and retraction of specific T cells, and death (10, 11, 38). Figure 1 shows the results of the determination of the lethal dose in the transgenic mice. Previously, we showed that littermate transgenic HLA-DR2β/IEα mice respond to BSAgs in a manner similar to that of BALB/c or C57BL/6 mice (11, 38). HLA-DR2β/IEα mice, on the same genetic background and transgenic for only the β chain of human MHC class II receptor, were used as the controls for the HLA-DQ8 mice. Because BALB/c mice have been widely used to examine the pathological effects of BSAgs, in our initial experiments we incorporated these mice as controls (Table 1). Within hours after SEB exposure, all HLA-DQ8 transgenic mice were visibly distressed and showed signs of ruffled coat and lethargy. These signs of malaise persisted for 6 to 72 h after aerosol challenge or until death. As shown in Fig. 1, most of the lethality in the HLA-DQ8 mice occurred within the first 4 to 6 days after exposure to SEB. SEB administered at the highest dose (310 μg/kg) was 100% lethal to all HLA-DQ8 transgenic mice. When HLA-DQ8 transgenic mice were exposed to 240 or 120 μg of SEB per kg, 88 and 80% lethality was observed, respectively, while 60 μg/kg induced lethality in 40% of the mice. Mice were slower to succumb to SEB challenges of 120 and 60 μg/kg. The dosage of 30 or 5 μg of SEB per kg resulted in no lethality; nevertheless, the mice displayed signs of malaise and were visibly distressed for 12 to 48 h postexposure. Based on the lethality curves, the median dose resulting in 50% lethality was approximately 70 μg/kg, showing that HLA-DQ8 transgenic mice are highly susceptible to aerosols of SEB.

• Open in new tab
• Download powerpoint

FIG. 1.

Aerosol exposure to SEB is lethal to HLA-DQ8/human CD4 transgenic mice. HLA-DQ8/human CD4 transgenic mice were exposed to 5, 30, 60, 120, 240, or 310 μg of aerosolized SEB per kg. All mice were monitored for illness for 21 days after SEB challenge. Results are plotted on Kaplan-Meier survival curves as percent survival for each group over time (n = 10 to 20 per group). The data presented are a compilation of those from two to four experiments with each dose of SEB.

In sharp contrast to the case for HLA-DQ8 transgenic mice, SEB at 310 or 240 μg/kg was not lethal to HLA-DR2β/IEα mice (Table 1). The HLA-DR2β/IEα mice showed no signs of stress after aerosol exposure. These results demonstrate that even at high doses of SEB, both chains of HLA-DR are needed for toxin to be effective in the mouse model. As previously demonstrated by LeClaire et al. (21), we also observed no lethality in BALB/c mice exposed to SEB at 310 μg/kg, although 100% lethality was produced in the BALB/c mice when the aerosol exposure was combined with a potentiating dose of LPS (Table 1). As expected, lower doses of SEB or LPS given alone were not lethal to BALB/c mice (data not shown).

SEB aerosols induce lung pathology.One of the major concerns regarding SEB in aerosol form is its ability to produce severe lung damage. Enterotoxin-producing S. aureus strains are associated with toxic shock syndrome, which includes systemic and respiratory complications such as cardiomyopathy, renal failure, electrolyte imbalances, disseminated intravascular coagulation, encephalopathy, and adult respiratory distress syndrome (2, 13, 18, 25, 31). Since in past studies SEB aerosol exposure was shown to induce lethal pulmonary lesions in nonhuman primates, we sought to examine whether aerosol delivery of SEB to the HLA-DQ8 mice induced respiratory damage (35, 39). Therefore, the effects of SEB on the lungs of HLA-DQ8 transgenic mice were examined and compared to those histological lesions induced in rhesus monkeys, a well-established nonhuman primate model for aerosol exposure to SEs. In contrast to mice sham exposed to saline, the mice exposed to SEB aerosols displayed substantial lung damage, which was characterized by the filling of alveolar spaces, interstitium, and lymphatics with edema, fibrin, and a cellular infiltrate composed of lymphocytes, macrophages, and neutrophils (Fig. 2A and B). The pathological changes observed in the HLA-DQ8 mice that were exposed to SEB aerosols were similar to those found in rhesus monkeys exposed to lethal doses of this toxin by the same route of exposure (Fig. 2C and D).

• Open in new tab
• Download powerpoint

FIG. 2.

Aerosol exposure to SEB induces pathological lesions in HLA-DQ8 transgenic mice. (A) Lung from a normal HLA-DQ8/human CD4 mouse. (B) Lung from an HLA-DQ8/human CD4 mouse 3 days after exposure by aerosol to 120 μm of SEB per kg. (C) Lung from a normal rhesus monkey. (D) Lung from an SEB-exposed rhesus monkey 4 days postchallenge with 5 to 7 LD₅₀ of SEB delivered by aerosol. Hematoxylin and eosin staining was used. Magnification, ×200. These data are representative of findings from two animals, and the experiment was repeated twice with similar results.

Aerosol exposure to SEB elicits proinflammatory cytokines.Acute induction of proinflammatory cytokine genes and secretion of proteins, including TNF-α, IL-1, IL-6, and IFN-γ, in the sera and spleens of mice parenterally administered SEB has been reported, and these manifestations are closely associated with toxin-induced lethal shock (32). Although there is a wealth of information available on cytokine induction after parenteral administration of BSAgs, little is known regarding the modulation of proinflammatory cytokines in the blood, lungs, and spleen after SEB aerosol challenge. Analyses of these tissue and fluid samples indicated an intense and rapid response to SEB, as several proinflammatory genes were found to be highly elevated in the lungs and spleens of HLA-DQ8 mice. The lungs of the HLA-DQ8 mice showed increases in TNF-α mRNA at all of the time points tested except in untreated mice, in which no cytokine mRNAs were detected (Fig. 3 and data not shown). The TNF-α mRNA induction stayed at high levels even at 24 h after SEB aerosol challenge, while increases in the cytokine mRNA levels of IL-2, IL-4, IL-6, and IFN-γ were not observed in the lungs of SEB-exposed HLA-DQ8 mice (Fig. 3). HLA-DR2β/IEα mice challenged with aerosol SEB showed only transient increases in IL-2, IL-4, IL-6, and IFN-γ mRNAs in the lungs (Fig. 3). The SEB-exposed HLA-DR2β/IEα mice had sustained levels of TNF-α, which were similar to those detected in HLA-DQ8 mice (Fig. 3). In contrast to the case for lungs, increased levels of TNF-α mRNA were noted at 4 and 10 h postexposure but were not detected at 24 h in spleens of SEB-exposed HLA-DQ8 mice (Fig. 4). The mRNA levels of IL-2 and IFN-γ increased in the spleens of SEB-exposed HLA-DQ8 transgenic mice after 4 h and remained elevated at 10 and 24 h after SEB exposure (Fig. 4). Increases in the mRNA expression of IL-6 genes were detected at 4 and 10 h postexposure but had subsided by 24 h after SEB aerosol challenge of the HLA-DQ8 mice (Fig. 4). We observed no enhancement of any cytokine mRNA prior to intoxication or of IL-4 mRNA throughout the experiment in the transgenic mice (Fig. 4). HLA-DR2β/IEα mice challenged with aerosol SEB showed only transient increases in IL-2 and TNF-α mRNAs in their spleens and no detectable mRNA expression of IL-4 or IL-6 (Fig. 4). IFN-γ mRNA expression was induced at 10 and 24 h after exposure to aerosolized SEB in the HLA-DRβ/IEα mice (Fig. 4).

• Open in new tab
• Download powerpoint

FIG. 3.

Exposure of HLA-DQ8 mice to aerosolized SEB induces cytokine mRNA changes in the lungs. Total RNA was isolated from the lungs of HLA-DR2β/IEα (unfilled bars) or HLA-DQ8 (filled bars) mice exposed by aerosolization to 120 μg of SEB per kg after 4, 10, and 24 h. RNAs from three to five animals were pooled, reverse transcribed, and cDNA amplified for IL-2, IL-4, IL-6, IFN-γ, TNF-α, or β-actin. RT-PCR products were resolved on an ethidium bromide-stained agarose gel. The data are expressed as the relative abundance of each cytokine mRNA, which was determined by comparing the signal of the RT-PCR product for the cytokine mRNA with that of the loading control of β-actin. These data are representative of those from three experiments with a similar design and outcome.

• Open in new tab
• Download powerpoint

FIG. 4.

Exposure of HLA-DQ8 mice to aerosolized SEB induces cytokine mRNA changes in the spleen. Total RNA was isolated from the spleens of HLA-DR2β/IEα (unfilled bars) or HLA-DQ8 (filled bars) mice exposed by aerosolization to 120 μg of SEB per kg after 4, 10, and 24 h. RNAs from three to five mice were pooled, reverse transcribed, and cDNA amplified for IL-2, IL-4, IL-6, IFN-γ, TNF-α, or β-actin. RT-PCR products were resolved on an ethidium bromide-stained agarose gel. The data are shown as the relative abundance of each cytokine mRNA, which was determined by comparing the signal of the RT-PCR product for the cytokine mRNA with that of the loading control of β-actin. These data are representative of those from three experiments with a similar design and outcome.

Since elevations in the mRNA expression of the cytokines varied in the spleen and lungs, we examined whether there were changes in serum cytokine levels. As shown in Fig. 5, substantial amounts of IL-2 and IL-6 were detected in sera of transgenic mice 4 h after SEB exposure, but these quickly declined. A robust and long-lasting IFN-γ response was observed after SEB challenge, contrasting with no apparent elevation in detectable serum TNF-α (Fig. 5). HLA-DRβ/IEα mice showed no sign of cytokine level increases, and no remarkable elevations in serum cytokines were detected at any time after SEB exposure (Fig. 5).

• Open in new tab
• Download powerpoint

FIG. 5.

Exposure to SEB aerosols increases serum cytokine levels in HLA-DQ8 mice. Circulating cytokine levels in HLA-DR2β/IEα (unfilled bars) or HLA-DQ8 (filled bars) mice were measured at 4, 10, or 24 h after exposure to aerosolized SEB (120 μg/kg). The levels of IL-2, IL-6, IFN-γ, and TNF-α were determined by ELISA. The bars indicate the means for four animals, and the error bars indicate the standard deviations. These data are representative of those from three experiments with a similar design and outcome.

Temperature changes in SEB-exposed HLA-DQ8 mice.One of the biological manifestations of toxic shock syndrome is temperature elevation (9, 17); hence, body temperature may provide a biomarker for SEB toxicity. In contrast to that in humans and nonhuman primates, toxic shock in mice is accompanied by hypothermia (37). To examine the effects of SEB exposure on the body temperature of HLA-DQ8 mice, telemetric implants were placed into the mice 14 days before SEB exposure. After SEB challenge, body temperature was continuously monitored for 14 days or until the animal died. Two days after SEB aerosol exposure, HLA-DQ8 mice receiving a dose of 120 μg/kg displayed sudden drops in body temperature (Fig. 6). The mice seemed to temporarily recover from hypothermia within 96 to 120 h after challenge, but then their body temperatures dropped swiftly within the next 12 h or until death. In these experiments, a dose of 120 μg/kg caused 100% lethality in HLA-DQ8 mice. Transgenic control HLA-DRβ/IEα mice showed no sign of hypothermia (Fig. 6) and had normal physical activity. No changes in body temperature were observed when HLA-DRβ/IEα mice were monitored for an additional 4 days (data not shown). These results suggest that, like in rhesus monkeys, temperature modulation is manifested in HLA-DQ8 mice soon after exposure to SEB aerosols, and temperature shifts may be indicative of BSAg-induced toxic shock (Fig. 6) (9).

• Open in new tab
• Download powerpoint

FIG. 6.

Exposure of HLA-DQ8 mice to aerosolized SEB results in dramatic decreases in body temperature. Mice were implanted with telemetric chips to measure their body temperature. HLA-DRβ/IEα (open circles) or HLA-DQ8 (filled circles) transgenic mice were exposed to 120 μg of SEB per kg. The body temperatures of the mice were monitored after exposure. The results are plotted as the mean of the body temperature (n = 5 per group, except at time points beyond 120 h, when the SEB-treated HLA-DQ8 mice began to succumb to toxic shock). The error bars indicate the standard deviations of the group means. These data are representative of those from three experiments with a similar design and outcome.