Staphylococcal enterotoxin A induction of pro-inflammatory cytokines and lethality in mice is primarily dependent on MyD88

Abstract

Staphylococcal enterotoxin (SE) -induced toxic shock is triggered by inflammatory cytokine signal amplification after SE binding to major histocompatibility complex class II molecules on antigen-presenting cells and T-cell receptors. Identifying host cellular elements contributing to this pro-inflammatory signal amplification is critical for developing a strategy for therapeutic intervention. Myeloid differentiation primary-response protein 88 (MyD88) is an intracellular signalling adaptor protein primarily known for mediating pro-inflammatory cytokine responses. We investigated the role of MyD88 in staphylococcal enterotoxin A (SEA) -treated cell cultures and mouse models of toxic shock. Our results demonstrated that elevated levels of tumour necrosis factor-α, interferon-γ, interleukin-1α/β (IL-1α/β), IL-2 and IL-6 production correlated with up-regulation of MyD88 after treatment of spleen cells and mice with SEA alone or in combination with lipopolysaccharide (LPS). The SEA-induced lethality was also observed in (LPS-independent) d-galactosamine-sensitized mice. While LPS potentiated SEA-induced cytokine responses, d-galactosamine treatment had no additive effect. Most importantly, our results demonstrated that MyD88^−/− mice were resistant to SEA-induced toxic shock and had reduced pro-inflammatory cytokine responses. These results suggest that SEA-induced lethality is primarily dependent on MyD88. Our findings offer an important insight on potential therapeutic treatment of SEA-induced toxic shock targeting MyD88.

Introduction

Certain bacterial exotoxins from Gram-positive bacteria such as Staphylococcus aureus and Streptococcus pyogenes cause toxic shock syndrome in humans and animals, which often progresses to sepsis, multi-organ dysfunction and death.^1–4Staphylococcus aureus is one of the most commonly isolated Gram-positive bacteria from patients with sepsis^5,6 and produces a wide array of toxins such as staphylococcal enterotoxin A (SEA), which belongs to a family of related enterotoxins.⁷ The toxin-mediated illness, toxic shock syndrome results from the ability of enterotoxins including SEA to act as a superantigen. Superantigens stimulate immune-cell expansion and rampant pro-inflammatory cytokine production in a manner that bypasses normal major histocompatibility complex (MHC) -restricted antigen processing. The SEA cross-links the MHC class II molecules present on antigen-presenting cells to T-cell receptor β-chains, triggering a release of pro-inflammatory cytokines from both cells, including tumour necrosis factor α (TNF-α), interferon γ (IFN-γ), interleukin-1 (IL-1) and IL-6.^8–12 In a mouse model, the biological effects of staphylococcal enterotoxins (SEs) are potentiated by lipopolysaccharide (LPS), a bacterial factor that binds to toll-like receptor 4 (TLR4) on the surface of cells. The SEA and LPS synergistically amplify the pro-inflammatory cytokines that lead to severe toxicity.^13–15 It appears that SEA is more potent than other toxin serotypes such as staphylococcal enterotoxin B (SEB) or staphylococcal enterotoxin C₁ (SEC₁).¹³ Although LPS enhances the lethality of SEA in vivo,^13–16 the mechanism underlying the synergistic induction of pro-inflammatory cytokine responses remains unclear. In earlier studies, binding of SE to MHC class II molecules on monocytes resulted in an increase in IL-1 and TNF-α.¹⁷ The SEB-dependent induction of IL-1 and TNF-α was thought to involve protein kinase C and protein tyrosine kinase.^17–19 Recent results indicate that IFN-γ as well as IL-1 rely on myeloid differentiation factor 88 (MyD88) -dependent pathways.²⁰ The adaptor protein MyD88 integrates and transduces intracellular signals generated by the TLR and interleukin receptor (IL-R) superfamily, critically regulating innate immunity and host defence.^21,22 However, the impact of MyD88-mediated signalling with respect to toxic shock remains unknown.

Conventionally, agonists (bacterial components) binding to host innate immune receptors, IL-1R, and all of the TLRs (excluding TLR3) prompt recruitment of the Toll-interleukin receptor (TIR) domain-containing adaptor molecule MyD88 through a TIR–TIR domain interaction. This interaction leads to downstream nuclear factor-κB (NF-κB) activation which permits the trans-activation of pro-inflammatory cytokine genes.^23–25 With regard to TLR and MyD88 signalling in human monocytes, it has been reported that ligation of MHC class II with SEB up-regulates monocyte membrane TLR4 expression.²⁶ In fact, because of the monocyte response to enterotoxin endotoxin pro-inflammatory effects, it has been suggested that monocytes are an early responder in Gram-positive toxic shock.²⁶ It is important to note that many septic patients are simultaneously exposed to multiple bacterial products from Gram-negative and Gram-positive pathogens, which augments the pro-inflammatory responses. An understanding of how these pathogen-derived exotoxins (e.g. SEA) and endotoxins (e.g. LPS) induce or intersect in host inflammation and amplify the response is important for identifying targets for therapeutic intervention.

Deficiency of MyD88 reportedly improves resistance against sepsis caused by polymicrobial infection.²⁷ Peritoneal macrophages from MyD88-deficient mice are unable to produce any detectable levels of TNF-α and IL-6 in response to S. aureus infection.²⁸ Despite the distinct functions of MyD88 in regulating the pro-inflammatory response, it remains unclear whether functional activation of the MyD88-mediated signalling contributes to SEA-induced toxic shock or co-operates with the TLR signalling cascade to amplify the inflammatory responses. As signalling by MyD88 appears to be a common link in many inflammatory processes, we examined whether MyD88 plays a dominant role in host-directed pro-inflammatory responses to SEA. Earlier studies showed that mice lacking MyD88 are defective in pro-inflammatory cytokine responses and they provide a useful model for evaluating innate immunity and immune regulation.^29,30 In this study, we investigated the role of MyD88 in SEA-induced pro-inflammatory cytokine responses and lethality using spleen cells and a mouse model of toxicity.

Materials and methods

Reagents

The SEA was purchased from Porton Down, Inc. (Salisbury, UK) and stored at −50°. It was endotoxin free and prepared under good manufacturing practice conditions. Escherichia coli LPS (055:B5) was purchased from Difco Laboratories (Detroit, MI). d-Galactosamine (d-Gal) was purchased from Sigma Chemical (St Louis, MO). A cytometric bead array (CBA) for cytokine analysis was purchased from BD Biosciences Pharmingen (San Diego, CA). An RNA-extracting reagent Tri-Reagent was obtained from Molecular Research Center Inc. (Cincinnati, OH) and Maloney murine leukaemia virus reverse transcriptase was purchased from Perkin Elmer (Waltham, MA). Primary anti-MyD88 antibody was obtained from AnaSpec, Inc. (San Jose, CA). β-Actin antibody was purchased from Cell Signalling (Danvers, MA) and Syber Green PCR master mix was obtained from BioRad (Hercules, CA).

Mice

Pathogen-free, C57BL/6 mice (6–8 weeks old) were obtained from Charles River (NCI-Frederick, Frederick, MD). MyD88 gene knockout (MyD88^−/−) mice of C57BL/6 background were obtained from Oriental Bio-service, Inc. (Ukyo-ku, Kyoto, Japan) and bred at Charles River Laboratories (Wilmington, MA).

SEA challenge of mice

To measure cytokines in serum and in vivo activation of MyD88 and NF-κB, C57BL/6 mice were injected intraperitoneally with SEA and 2 hr later with LPS (100 μg/mouse). In a d-Gal sensitization experiment, mice were injected with d-Gal (20 mg/mouse) 30 min before SEA injection. Blood samples were collected through the tail vein and pooled for measuring serum cytokines, after which mice were killed and their spleens were removed.

Spleen cell isolation and cultures

Spleens were removed aseptically from the mice after killing. Single-cell suspensions were prepared by lysing red blood cells using ammonium chloride–potassium lysing buffer (Cambrex, Walkersville, MD), followed by three washes with medium. Spleen cells from C57BL/6, or MyD88^−/− mice were cultured with SEA (200 ng/ml) for 2 hr, before adding LPS (1 μg/ml). Cultures were incubated (37°, 5% CO₂) for 16 hr, supernatant was collected, and cells were harvested for measuring transcriptional activation of the MyD88 gene, TNF-α and IL-1β.

Real-time polymerase chain reaction

For measuring transcriptional activation of MyD88, total RNA was extracted from spleen cells using Tri-Reagent and reverse transcribed into complementary DNA (cDNA) with Maloney murine leukaemia virus reverse transcriptase according to the manufacturer’s instructions (Perkin Elmer). Quantitative real-time polymerase chain reaction (PCR) was performed using cDNA collected as described above. Amplification was performed using 4 μl of a 1 μm final concentration of gene specific primers, MyD88 (forward 5′-CACTCGCAGTTTGTTGGATG-3′; reverse 5′-CGCAGGATACTGGGAAAGTC-3′), and β-actin (forward 5′- TCCTGTGGCATCCACGAAACT-3′; reverse 5′-GAAGCATTTGCGGTGGACGAT-3′), 10 μl of Power Syber Green PCR master mix, and 6 μl cDNA according to the manufacturer’s instructions. Finally, the 20 μl PCR was amplified using a 7900HT Fast Real-Time PCR System. The final normalized results were calculated by dividing the relative transcript levels of the test genes by the relative amount of the β-actin RNA.

Cytokine analysis

Cytokines in culture supernatants were measured using a CBA kit using capture beads coated with antibodies specific for cytokines and flow cytometry analysis according to the manufacturer’s method, as described elsewhere.³¹ Interleukin-1α was measured using an enzyme-linked immunosorbent assay (ELISA) kit according to the manufacturer’s procedure (R & D Systems, Minneapolis, MN).

NF-κB assay

To prepare nuclear extract, spleen cells (4 × 10⁶) were washed and resuspended with 1 ml cold phosphate-buffered saline and placed on ice for 15 min to allow the cells to swell. Fifty microlitres of 10% Nonidet P-40 was added to the cells and briefly centrifuged. The nuclear pellet was resuspended in lysis buffer (Active Motif) containing dithiothreitol and protease inhibitor for 30 min on ice, then centrifuged for 10 min at 14 000 g at 4°; extract was then collected for NF-κB assay. The TransAm Chemi kit (Active Motif, Carlsbad, CA) had NF-κB-binding oligonucleotides immobilized to a 96-well plate. To the 96-well plate, nuclear extracts of spleen cells were placed in duplicates for an hour. The plates were then washed several times and diluted NF-κB antibody was added to the plates for 1 hr at room temperature. The plates were washed and a diluted horseradish peroxidase-conjugated secondary antibody was added to the plate for an hour at room temperature. After incubation, the plates were washed with wash buffer and a chemiluminescent working solution was added to detect NF-κB recognition of an epitope on the p65 subunit.

Western blots

Mouse spleen cells (5 × 10⁶/ml) were stimulated with LPS (1 μg/ml) or SEA (200 ng/ml), or were left untreated for 2 hr. Cells were collected into fresh 1·5-ml centrifuge tubes and chilled on ice for 5 min before centrifuging. Membrane and cytoplasm separation were performed by resuspending the pellets in 50 μl lysis buffer (Active Motif) in the presence of dithiothreitol, protease inhibitors and phosphatase inhibitors on ice for 30–60 min. The membrane fraction was collected by centrifuging lysates at 14 000 g’s for 20 min. Supernatant contained the cytoplasmic fraction and the pellet contained the membrane fraction. Samples containing 10 μg of total cytoplasmic proteins were separated by gel electrophoresis and transferred to nitrocellulose membranes. Nitrocellulose membranes were blocked overnight in Tris-buffered saline containing 0·1% Tween-20 and 3% bovine serum albumin at 4°. Blots were extensively washed and probed with anti-MyD88 polyclonal antibody followed by horseradish peroxidase-conjugated secondary antibody. Blots were washed extensively and developed with chemiluminescent substrate in the presence of hydrogen peroxide using Immun-Star WesternC Chemiluminescent Kit (BioRad). An imaging system VersaDoc Model 4000 (BioRad) was used to capture the image.

Statistical analysis

The planned analyses using t-tests with step-down Bonferroni adjustment to compare geometric mean cytokine levels between groups were undertaken for determining statistical significance. Kaplan–Meier survival analysis (dependent variables: survival status, time to death) and log-rank test to compare survival curves (with step-down Bonferroni adjustment for pairwise comparisons) among groups were performed.

Results

Up-regulation of MyD88 correlates with cytokine production in response to SEA and LPS treatment of spleen cell cultures

To examine induction of the pro-inflammatory cytokine response to SEA, we treated C57BL/6 mouse spleen cells with SEA, LPS or SEA plus LPS for 16 hr and measured TNF-α, IFN-γ and IL-6 release into the culture supernatant. Compared with untreated cells, increased production of TNF-α, IFN-γ and IL-6 in culture supernatants was observed in cells after SEA treatment (P-values ≤ 0·0006) (Fig. 1a). A substantial increase of these cytokines was observed when treated with SEA plus LPS compared with SEA or LPS alone (P ≤ 0·0411). Along with cytokine production, transcriptional up-regulation of MyD88 and correspondingly of TNF-α and IL-1β messenger RNA were also observed (Fig. 1b). Concurrent with TNF-α transcripts, release of TNF-α was observed after SEA plus LPS treatment. This was accompanied by de novo synthesis of MyD88 protein after treatment with SEA, LPS, or in combination (Fig. 1c). MyD88 is generally recruited as a dimer to the receptor complex. Our results showed an increase in MyD88 dimer formation when stimulated with SEA or LPS alone or SEA plus LPS compared with untreated spleen cells. These results indicate that as a consequence of SEA stimulation, along with cytokine release, up-regulation of MyD88 protein occurred in spleen cells.

Figure 1.

Pro-inflammatory cytokine production by murine spleen cells in response to staphylococcal enterotoxin A (SEA) and lipopolysaccharide (LPS) correlates with up-regulation of MyD88. Spleen cells from C57BL/6 mice were cultured with SEA (200 ng/ml) or LPS (1 μg/ml) or SEA plus LPS as described in the Materials and methods. (a) Analysis of tumour necrosis factor-α (TNF-α), interleukin-6 (IL-6) and interferon-γ (IFN-γ) in culture supernatants. Cytokines were measured by a cytometric bead array method using capture beads coated with antibodies specific for cytokines and flow cytometry. Data are presented as the average of three independent experiments. Error bars represent ± SEM. The statistical significance assigned as follows *P ≤ 0·0006 between untreated (Unt) versus SEA and **P ≤ 0·0411 between SEA versus SEA plus LPS treatment group respectively. (b) Transcriptional activation of MyD88, TNF-α and IL-1β messenger RNA in spleen cells cultured for 2 hr with SEA and LPS was measured by a quantitative real-time polymerase chain reaction. Relative expression of MyD88, TNF-α and IL-1β was normalized to the expression of β-actin. Expression levels represent means ± SEM. The experiments were performed at least three times on independent samples. (c) Up-regulation of MyD88 protein in spleen cells after stimulation with SEA or LPS. C57BL/6 spleen cells were cultured with SEA, LPS, or kept untreated. Cell lysates were prepared and 10 μg protein were separated by sodium dodecyl sulphate–polyacrylamide gel electrophoresis, blotted to nitrocellulose and probed with MyD88-specific and β-actin-specific antibodies.

Induction of serum cytokine response and NF-κB activation in response to SEA in mice

We examined in vivo responses by co-administering SEA (0·1, 0·5 and 10 μg per mouse) and LPS (100 μg) to C57BL/6 mice and measuring both serum cytokine responses as well as activation of NF-κB in spleen cells at 2 hr after administration. Our results showed that SEA plus LPS increased TNF-α, IFN-γ, IL-6 and IL-1α in serum relative to those mice injected with only SEA or LPS (Fig. 2a). Total serum levels of TNF-α, IFN-γ and IL-6 released in vivo corresponded to the challenge dose of SEB (0·1 and 0·5 μg), whereas serum levels of cytokine dropped at the higher dose (10 μg) of SEA plus LPS (Fig. 2a). These mice were in very poor health at the time of blood collection (2 hr) after LPS treatment. In addition to MyD88 up-regulation and cytokine response, the transcription factor, NF-κB, a key component of the signal transduction pathway downstream of MyD88, was also examined in nuclear extracts of spleen cells. Increased activation of the NF-κB p65 subunit was observed, which correlated with the SEA challenge dose (Fig. 2b). Collectively, these results suggest that SEA treatment is followed by MyD88 activation, increased cytokine production and NF-κB activation.

Figure 2.

Staphylococcal enterotoxin A (SEA) -induced serum cytokine responses in mice and activation of nuclear factor-κB (NF-κB). C57BL/6 mice were injected with SEA alone (10 μg/mouse) or lipopolysaccharide (LPS) alone (100 μg/mouse) or SEA at 0·1, 0·5 or 10 μg/mouse followed by LPS (100 μg/mouse) 2 hr later. (a) Serum tumour necrosis factor-α (TNF-α), interferon-γ (IFN-γ), interleukin-6 (IL-6) and IL-1α were measured by a cytometric bead array kit and flow cytometry. (b) NF-κB activation. Transcription factor NF-κB activation was examined in nuclear extract of spleen cells. Data are representative of three independent experiments.

MyD88 deficiency impairs pro-inflammatory cytokine production in spleen cells in response to SEA and LPS

We next examined in vitro cytokine production in spleen cells of MyD88-deficient mice exposed to SEA in culture. Figure 3 illustrates a drastic reduction in the TNF-α, IL-6 and IFN-γ production by MyD88-deficient spleen cells in response to SEA, LPS or SEA plus LPS when compared with wild-type C57BL/6 mice (P ≤ 0·05). MyD88-deficiency therefore caused severe reduction of pro-inflammatory cytokine responses to SEA. Collectively, these results reveal that pro-inflammatory cytokine production was dependent upon MyD88.

Figure 3.

Impairment of tumour necrosis factor-α (TNF-α), interleukin-6 (IL-6) and interferon-γ (IFN-γ) production in spleen cells from MyD88^−/− mice. Spleen cells (5 × 10⁶/well) from MyD88^−/− or C57BL/6 (wild-type) mice were cultured in vitro with staphylococcal enterotoxin A (SEA; 200 ng/ml), lipopolysaccharide (LPS; 1 μg/ml) or SEA (200 ng/ml) plus LPS (1 μg/ml) for 16 hr. Culture supernatants were tested after (a) SEA; (b) LPS; (c) SEA plus LPS treatment for the presence of pro-inflammatory cytokines using a cytometric bead assay as described elsewhere.³¹ Data presented an average by compiling data from four independent experiments. Cytokine release levels represent means ± SEM. The statistical analysis for cytokines TNF-α, IFN-γ and IL-6 treated with (a) SEA (P ≤ 0·05); (b) LPS (P ≤ 0·045) and (c) SEA+LPS (P ≤ 0·05) showed a statistical significance between C57BL/6 versus MyD88^−/− mice.

MyD88-dependent pro-inflammatory cytokine response in mice contributes to lethal toxicity

To further confirm that MyD88-dependent pro-inflammatory cytokine responses contributed to toxic shock-induced death, we compared survival and serum cytokine responses in MyD88^−/− and C57BL/6 mice after exposure to SEA. Mice were injected with graded doses of SEA (0·5, 2·5 and 5 μg/mouse) followed by LPS 2 hr later. Mice were bled 3 hr after SEA challenge to measure serum cytokines and were observed for 96 hr for survival. Control mice injected with LPS or SEA alone did not die (Fig. 4a). In contrast to survival of wild-type mice (P < 0·0001), MyD88-deficient mice (P = 1·0000) were resistant to toxic shock-induced death even at the highest doses of SEA (Fig. 4a). Serum levels of TNF-α, IFN-γ, IL-2, IL-5 and IL-1α were lower in MyD88-deficient mice compared with C57BL/6 mice in response to SEA plus LPS (Fig. 4b). Although IFN-γ and IL-5 were detected in the serum of MyD88-deficient mice, the serum levels of IL-2, IFN-γ, TNF-α and IL-5 were threefold to fivefold less compared with levels in wild-type mice treated with SEA plus LPS. In contrast to SEA plus LPS treatment, there was not much serum TNF-α response with SEA alone in MyD88^−/− mice or in wild-type mice. Although the serum cytokine responses to different SEA doses remained almost unchanged in MyD88^−/− mice, there was an SEA dose-dependent effect in serum cytokine responses in C57BL/6 mice. These results suggested that reduced cytokine responses and resistance to SEA challenge were the result of MyD88.

Figure 4.

MyD88^−/− mice are resistant to staphylococcal enterotoxin A (SEA) intoxication. Lipopolysaccharide (LPS) -potentiated lethal SEA challenge in C57BL/6 and MyD88^−/− mice was used to evaluate the role of MyD88 in pro-inflammatory cytokine responses and toxicity. (a) MyD88 contributed to lethal toxicity. Mice were injected intraperitoneally with SEA (0·5, 2·5 or 5 μg/mouse) followed by LPS (100 μg/mouse), or with SEA or LPS alone. Mice were observed for survival. Control mice injected with 100 μg of LPS or 5 μg of SEA survived. Data represent one of two similar experiments. Data presented as a percentage survival of mice (n = 6 per group). The log-rank test for survival curves among all groups was significant (P < 0·0001) in C57BL/6 mice but no statistically significant difference was found in survival curves among all groups of MyD88^−/− mice (P = 1·0000). (b) Serum tumour necrosis factor-α (TNF-α), interferon-γ (IFN-γ), interleukin-2 (IL-2), IL-5 and IL-1α response in C57BL/6 and MyD88^−/− mice injected with SEA followed by LPS. Blood was collected 3 hr after the LPS injection and pooled. TNF-α, IFN-α, IL-2 and IL-5 were measured by cytometric bead assay as described earlier. IL-1α was measured by an enzyme-linked immunosorbence-based assay using anti-IL-1α antibody-coated plates according to the manufacturer’s protocol.

d-Galactosamine treatment has no effect on pro-inflammatory cytokine induction by spleen cells

Lipopolysaccharide contributes to SEA-induced production of pro-inflammatory cytokine responses via TLR4–MyD88. To differentiate SEA-induced toxicity from LPS, we examined SEA-induced pro-inflammatory cytokine responses independent of LPS. The d-Gal-sensitized mice have been used as a model to evaluate potential toxic shock syndromes triggered by SE.¹ As a control experiment, to corroborate previously published work and to use as a potentiator for an SEA toxicity assay in mice, we examined in vitro cytokine production in d-Gal-treated spleen cells cultured with SEA. The results shown in Fig. 5 indicate that besides IL-6, there was no apparent increase in cytokine production after d-Gal treatment. These results suggest that d-Gal is a viable potentiator of SEA-induced toxicity that does not contribute to cytokine production as described earlier.¹

Figure 5.

Staphylococcal enterotoxin A (SEA) but not d-galactosamine (d-Gal), induced cytokine production by spleen cells. Spleen cells (5 × 10⁶/well) from C57BL/6 mice were cultured with SEA (200 ng/ml), lipopolysaccharide (LPS), or d-Gal (0·1 or 5 mg/ml) combination for 16 hr. Culture supernatants were tested for pro-inflammatory cytokines by cytometric bead assay as described elsewhere.³¹

Lethal effect of SEA in d-Gal-treated mice

The sensitivity of mice to endotoxin can be greatly enhanced by impairing liver metabolism with d-Gal.³² Similar to LPS potentiation, d-Gal-sensitized mice have been used as a model system to evaluate potential toxic shock syndromes triggered by SE.^1,2 We evaluated the lethal toxicity of SEA in d-Gal-sensitized mice. Mice were injected with a dose of 20 mg d-Gal (i.p.), 30 min before SEA injection. Results shown in Fig. 6(a) indicate that d-Gal treatment, as was the case with LPS, generates high sensitivity to increasing doses of SEA (P ≤ 0·0280). Yet, d-Gal alone did not induce cytokine production or lethality. Besides IL-6, which is known to be released after d-Gal treatment and is important for liver regeneration, there was no obvious difference in the serum pro-inflammatory cytokine response between SEA and d-Gal plus SEA-treated mice (Fig. 6b). However, the d-Gal plus SEA-treated mice succumbed to lethal shock. These results indicate that the SEA-induced cytokine production contributes to toxicity in d-Gal-sensitized mice.

Figure 6.

Lethal effect of staphylococcal enterotoxin A (SEA) on d-galactosamine (d-Gal) -sensitized mice. C57BL/6 mice were injected intraperitoneally with d-Gal (20 mg/mouse) 30 min before SEA (2, 20 and 100 μg/mouse) challenge. Mice were observed for survival. (a) SEA dose-dependent lethality. Control mice injected with 100 μg of SEA or 20 mg of d-gal survived. The Kaplan–Meier survival curves among all groups showed P< 0·0001). (b) Serum cytokine response in mice injected with phosphate-buffered saline or SEA (20 μg/mouse), or d-Gal (20 mg/mouse) or d-Gal plus SEA/mouse. Blood was collected 3 hr after the SEA injection. Serum was tested for tumour necrosis factor-α (TNF-α), interferon-γ (IFN-γ), interleukin-6 (IL-6), IL-2, IL-12p70 and monocyte chemotactic protein 1 (MCP-1) by cytometric bead assay as described earlier.

MyD88 requirement for SEA toxicity

Our in vivo results demonstrated that MyD88-mediated pro-inflammatory cytokine responses contribute to SEA lethality in the presence of LPS or d-Gal. To further establish that MyD88-mediated signalling plays a vital role in SEA toxicity, we evaluated the requirement of MyD88 by comparing survival of d-Gal-sensitized C57BL/6 and MyD88^−/− mice using a significantly high dose of SEA (approximately 50 lethal dose 50%). At 6 hr, we recorded only 33% survival among wild-type C57BL/6 mice, whereas all of the MyD88^−/− mice survived (Fig. 7). By 24 hr, all C57BL/6 mice had died, whereas 66% of MyD88^−/− mice were alive (P < 0·0212). These results suggest that SEA-induced production of pro-inflammatory cytokines and toxicity primarily depend on MyD88.

Figure 7.

MyD88 is required for toxicity after staphylococcal enterotoxin A (SEA) exposure in d-galactosamine (d-Gal) -sensitized mice. MyD88^−/− mice showed delayed lethality and increased survival in response to high doses of SEA challenge. C57BL/6 and MyD88^−/− mice were injected with d-Gal (20 mg/mouse) 30 min before SEA (100 μg/mouse). Mice were examined for survival. Control C57BL/6 and MyD88^−/− survived d-Gal treatment. The Kaplan–Meier survival analysis provided evidence to suggest that there is a statistically significant difference in survival curves among all groups (P < 0·0001). Also, pairwise comparisons of survival curves between C57BL/6 and MyD88^−/− mice (C57BL/6 d-Gal + SEA 100 μg versus MyD88^−/−d-Gal + SEA 100 μg) was statistically significant (P< 0·0212).

Discussion

We present evidence that MyD88-mediated pro-inflammatory cytokine signalling contributes to SEA-induced toxic shock in mice. Our results demonstrate that MyD88^−/− mice were protected from a lethal SEA challenge and did not release pro-inflammatory cytokines into their serum. In contrast, the potent cytokine response of wild-type mice was significant and lethal. Spleen cells from the MyD88^−/− mice had impaired inflammatory cytokine production in response to SEA. These findings are in agreement with previous reports concerning the inability of peritoneal macrophages from MyD88-deficient mice to produce any detectable levels of TNF-α and IL-6 to S. aureus infection.²⁸ The signal transduction mechanism, particularly for IL-1 and TNF-α expression after SEB binding to MHC class II molecules on human monocytes, has been partially characterized and involves protein kinase C and protein tyrosine kinase.^18,19 However, what has not been defined is the dominant pro-inflammatory signalling mechanism that contributes to SEA induction of pro-inflammatory responses. In this study, our data suggest that MyD88-mediated signalling is critically involved in pro-inflammatory cytokine responses and SEA-induced toxic shock.

Signalling by MyD88 appears to be a common link in inflammatory processes.^21–23 The MyD88 integrates and transduces intracellular signals generated by the TLR and IL-R superfamily binding to various microbial components. The signalling begins through TIR domain interactions that are critical for several aspects of innate immune regulation.^33,34 The signal eventually results in massive release of inflammatory cytokines such as TNF-α, IL-1β, IL-2, IL-6, IFN-γ etc. These cytokine responses ultimately lead to septic shock and death. Besides TLR/IL-1R1 receptor interaction, MyD88 can associate with signalling proteins by means other than homophilic TIR–TIR or death-domain interactions.²⁰ Interactions have been detected between MyD88 and other proteins lacking TIR and death domains such as Bruton’s tyrosine kinase,³⁵ phosphatidyl-inositol-3-OH kinase,³⁶ and IFN regulatory factor 7.³⁷ Other results indicate that binding of SEA to MHC class II molecules on human monocytes activated MyD88-mediated induction of pro-inflammatory cytokines (Kissner et al. 2009, manuscript submitted). Along with transcriptional up-regulation of MyD88, de novo synthesis of MyD88 was observed in spleen cells after stimulation with SEB. Previous studies propose that SEB binding to MHC class II on monocytes elicits rapid transcription of IL-1β and TNF-α gene(s),^11,17 plus it increases membrane expression of TLR4²⁶ (Saikh et al. unpublished observation). Additionally, attenuation of several cytokines and chemokines released following LPS stimulation by MyD88-defective cells was also reported.^38,39 In line with these observations, our results provide a possible link between MyD88-mediated signalling and LPS potentiation of toxicity.¹³ Importantly, our results demonstrate that MyD88 was a required component of intracellular signalling in response to SEA as well as to SEA plus LPS. This signalling intersects at a number of points that include activation of both antigen-presenting cells and T cells, leading to the release of pro-inflammatory cytokines and toxic shock. Besides LPS potentiation of SE-induced toxic shock in mice, several other mouse models of toxic shock have employed sensitization methods^1,2 because mice are more resistant than humans to the toxic effects of SEs. Our in vitro and in vivo data show that d-Gal treatment does not play a role in cytokine induction after SEA exposure. In d-Gal-sensitized mice, SEA dose-dependent lethality strongly suggests SEA-induced toxicity. The effect of d-Gal was likely to augment sensitivity to endogenous mediators, rather than to affect their production. Hence the inflammatory cytokine response and lethality to SEA appeared to be a contribution of MyD88-mediated signalling because MyD88 knockout mice were resistant to SEA. Although low levels of IFN-γ and IL-5 were observed in MyD88 knockout mice, these animals were protected from toxic-shock-induced death. It is likely the low cytokine response by spleen cells of MyD88 knockout mice may be attributed to a signalling mechanism such as the responses mediated by protein tyrosine kinase C or tyrosine kinase that were reported earlier.^2,18

In summary, our results demonstrate that SEA toxic-shock-induced death in mice is critically dependent on the recruitment of MyD88. MyD88 represents an important intracellular target for therapeutic intervention against SEA exposure. Targeting of MyD88 to disrupt pro-inflammatory responses would probably provide critical leads towards developing efficacious therapeutics against toxic shock. Our current research effort is directed towards this objective.

Disclosures

Research was conducted in compliance with the Animal Welfare Act and other federal statutes and regulations relating to animals and experiments involving animals and it adheres 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. Views expressed in this paper are those of the authors and do not purport to reflect official policy of the U.S. Government.

Conflict of interest

The authors declare no financial or commercial conflict of interest.