Superantigens are critical for Staphylococcus aureus Infective endocarditis, sepsis, and acute kidney injury

Abstract

Infective endocarditis and kidney infections are serious complications of Staphylococcus aureus sepsis. We investigated the role of superantigens (SAgs) in the development of lethal sepsis, infective endocarditis, and kidney infections. SAgs cause toxic shock syndrome, but it is unclear if SAgs contribute to infective endocarditis and kidney infections secondary to sepsis. We show in the methicillin-resistant S. aureus strain MW2 that lethal sepsis, infective endocarditis, and kidney infections in rabbits are critically dependent on high-level SAgs. In contrast, the isogenic strain lacking staphylococcal enterotoxin C (SEC), the major SAg in this strain, is attenuated in virulence, while complementation restores disease production. SAgs' role in infective endocarditis appears to be both superantigenicity and direct endothelial cell stimulation. Maintenance of elevated blood pressure by fluid therapy significantly protects from infective endocarditis, possibly through preventing bacterial accumulation on valves and increased SAg elimination. These data should facilitate better methods to manage these serious illnesses.

Importance: The Centers for Disease Control and Prevention reported in 2007 that Staphylococcus aureus is the most significant cause of serious infectious diseases in the United States (R. M. Klevens, M. A. Morrison, J. Nadle, S. Petit, K. Gershman, et al., JAMA 298:1763-1771, 2007). Among these infections are sepsis, infective endocarditis, and acute kidney injury. Infective endocarditis occurs in 30 to 60% of patients with S. aureus bacteremia and carries a mortality rate of 40 to 50%. Over the past decades, infective endocarditis outcomes have not improved, and infection rates are steadily increasing (D. H. Bor, S. Woolhandler, R. Nardin, J. Brusch, D. U. Himmelstein, PLoS One 8:e60033, 2013). There is little understanding of the S. aureus virulence factors that are key for infective endocarditis development and kidney abscess formation. We demonstrate that superantigens are critical in the causation of all three infections. We show that their association results from both superantigenicity and direct toxic effects on endothelial cells, the latter likely contributing to delayed endothelium healing. Our studies contribute significantly to understanding the development of these illnesses and are expected to lead to development of important therapies to treat such illnesses.

FIG 1

Deletion of sec in S. aureus strain MW2 protects rabbits against lethality in the IE and sepsis model despite high-load bacteremia. (A) Percent survival of rabbits infected intravenously with 2 × 10⁷ to 3 × 10⁷ CFU of wild-type MW2, the MW2Δsec strain, the MW2Δsec strain with psec (ectopic expression), or the MW2Δselx strain after mechanical damage of the aortic valves. ***, P = 0.0009 (MW2 versus MW2Δsec) and P = 0.0002 (MW2 versus MW2Δsec + psec), log-rank, Mantel-Cox test. (B) Bacterial counts per milliliter of blood recovered from the rabbits postmortem. (C) Enlargement of the spleen resulting from S. aureus bacteremia. UI, uninfected. *, P = 0.0385, one-way ANOVA and nonparametric Kruskal-Wallis test. (B and C) Horizontal lines and error bars represent mean values ± standard errors of the means (SEM). P values of ≤0.05 are considered statistically significant. No P value means no statistical significance.

FIG 2

Deletion of sec in S. aureus strain MW2 protects rabbits against IE. (A) Representative images of vegetative lesions on aortic valve cusps caused by wild-type MW2; (B) total weight of vegetations dissected from aortic valves after intravenous inoculation with 2 × 10⁷ to 3 × 10⁷ CFU of wild-type MW2, the MW2Δsec strain, the MW2Δsec strain with psec (ectopic expression), or the MW2Δselx strain; (C) bacterial counts recovered from aortic valve vegetations shown in panel B. ***, P = 0.0002; **, P = 0.0011; one-way ANOVA and nonparametric Kruskal-Wallis test. Horizontal lines and error bars represent mean values ± SEM. P values of ≤0.05 are considered statistically significant.

FIG 3

Deletion of sec in S. aureus strain MW2 protects rabbits against development of acute renal infarction and abscess formation and partially against lung necrosis and hemorrhage in the infective endocarditis and sepsis model. (A) Representative images of kidneys harvested from infected rabbits demonstrating the severity of kidney injury by SEC-producing S. aureus strains. (B) Quantification of renal injury and abscess formation plotted as the percentage of the kidney surface area with visible kidney injury in rabbits infected intravenously with 2 × 10⁷ to 3 × 10⁷ CFU of wild-type MW2, the MW2Δsec strain, the MW2Δsec strain with psec (ectopic expression), or the MW2Δselx strain after mechanical damage of the aortic valves. *, P = 0.025, nonparametric (one-way ANOVA) Kruskal-Wallis test. Horizontal lines and error bars represent mean values ± SEM. P values of ≤0.05 are considered statistically significant. (C) Representative images of lungs harvested from infected rabbits. Gross examination shows an improvement in the pathology of the lungs in rabbits infected with the sec deletion strain.

FIG 4

Hematoxylin and eosin-stained section of vegetative lesions on aortic valves characteristic of staphylococcal endocarditis. (A) A 2× composite of a vegetative lesion containing heart muscle (asterisks), valves (solid lines), and aortic artery (dashed lines) with vegetations (block arrows) primarily centered on the valves. Foci of necropurulent inflammation were detected in the nearby adjacent heart muscle (long arrow). (B) A ×10 magnification of the vegetative lesion shown in panel A highlights the presence of extensive bacterial colonies within the valve leaflets (block arrows). Vegetations are composed of aggregates of fibrin (star) with pockets of platelets, erythrocytes, and large masses of bacterial cocci consistent with S. aureus (dashed outline) in addition to bacterial macroclusters centered on the valves.

FIG 5

Fluid replacement therapy inhibits vegetation formation and decreases bacterial burden. For fluid replacement, 25-ml volumes of sterile 0.9% saline solution were injected subcutaneously twice daily. (A) Bacterial counts per milliliter of blood recovered from the rabbits postmortem; (B) total weight of vegetations dissected from aortic valves after intravenous inoculation with 2 × 10⁷ to 3 × 10⁷ CFU of wild-type MW2; (C) bacterial counts recovered from aortic valve vegetations shown in panel B. **, P = 0.0074; *, P = 0.0312; nonparametric Mann-Whitney test. Horizontal lines and error bars represent mean values ± SEM. P values of ≤0.05 are considered statistically significant.

FIG 6

SEC induces IL-8 production in primary human aortic endothelial cells (HAECs), and inhibitors of various IL-8 transcription activating pathways in epithelial/endothelial cells alter the SEC-induced IL-8 production by primary HAECs. (A) IL-8 detected in culture supernates after 6 h of incubation of primary HAECs with increasing concentrations of SEC (**, P = 0.005). Panels B to D show IL-8 detected in culture supernates after 30 min of treatment with inhibitors and 6 h of incubation with 10 µg/ml of SEC. (B) TAPI-1 (pan-ADAM and matrix metalloproteinase inhibitor; *, P = 0.016); (C) gallein (Gβγ inhibitor; *, P = 0.011); (D) axitinib (vascular endothelial growth factor receptor inhibitor; **, P = 0.008). P values were determined with the one-way ANOVA and nonparametric Kruskal-Wallis test. P values of ≤0.05 are considered statistically significant.