Interactions of Streptococcus suis serotype 9 with host cells and role of the capsular polysaccharide: Comparison with serotypes 2 and 14

Abstract

Streptococcus suis is an important porcine bacterial pathogen and a zoonotic agent responsible for sudden death, septic shock and meningitis, of which serotype 2 is the most widespread, with serotype 14 also causing infections in humans in South-East Asia. Knowledge of its pathogenesis and virulence are almost exclusively based on these two serotypes. Though serotype 9 is responsible for the greatest number of porcine cases in Spain, the Netherlands and Germany, very little information is currently available regarding this serotype. Of the different virulence factors, the capsular polysaccharide (CPS) is required for S. suis virulence as it promotes resistance to phagocytosis and killing and masks surface components responsible for host cell activation. However, these roles have been described for serotypes 2 and 14, whose CPSs are structurally and compositionally similar, both containing sialic acid. Consequently, we evaluated herein the interactions of serotype 9 with host cells and the role of its CPS, which greatly differs from those of serotypes 2 and 14. Results demonstrated that serotype 9 adhesion to but not invasion of respiratory epithelial cells was greater than that of serotypes 2 and 14. Furthermore serotype 9 was more internalized by macrophages but equally resistant to whole blood killing. Though recognition of serotypes 2, 9 and 14 by DCs required MyD88-dependent signaling, in vitro pro-inflammatory mediator production induced by serotype 9 was much lower. In vivo, however, serotype 9 causes an exacerbated inflammatory response, which combined with persistent bacterial presence, is probably responsible for host death during the systemic infection. Though presence of the serotype 9 CPS masks surface components less efficiently than those of serotypes 2 and 14, the serotype 9 CPS remains critical for virulence as it is required for survival in blood and development of clinical disease, and this regardless of its unique composition and structure.

Fig 1. Absence of S. suis CPS is associated with increased surface hydrophobicity.

Surface hydrophobicity of the S. suis serotype 2 (S2), 9 (S9) or 14 (S14) wild-type and mutant strains was determined using n-hexadecane. Data represent the mean ± SEM (n = 3). * (p < 0.05) indicates a significant difference between wild-type and mutant strains.

Fig 2. Deletion of S. suis serotype 9 cpsG gene results in non-encapsulation.

Transmission electron microscopy following stabilization with ferritin of the serotype 9 wild-type (20 000x magnification) (A) and mutant strain (15 000x magnification) (B). Black bars = 1 μm.

Fig 3. Presence of CPS partially or significantly modulates S. suis serotype 9 adhesion to and invasion of respiratory epithelial cells, respectively.

Adhesion (A & B) and invasion (C & D) of the S. suis serotype 2 (S2), 9 (S9) or 14 (S14) wild-type and mutant strains to NPTr porcine tracheal epithelial cells after 2 h (A & C) or 4 h (B & D) of incubation. Data represent the mean ± SEM (n = 4). * (p < 0.05) indicates a significant difference between wild-type and mutant strains.

Fig 4. Presence of S. suis CPS confers anti-phagocytic properties towards macrophages regardless of serotype.

Internalization of the S. suis serotype 2 (S2), 9 (S9) or 14 (S14) wild-type and mutant strains by J774A.1 murine macrophages after 1 h (A) or 2 h (B) of incubation in the absence of mouse serum or following opsonization with 20% heat-inactivated or complete serum. Data represent the mean ± SEM (n = 4). * (p < 0.05) indicates a significant difference between wild-type and mutant strains.

Fig 5. Presence of CPS is required for S. suis whole blood bactericidal resistance regardless of serotype.

Capacity of the S. suis serotype 2 (S2), 9 (S9) or 14 (S14) wild-type and mutant strains to resist the bactericidal effect of murine (A) or porcine (B) whole blood after 2 h of incubation. Percentage of bacterial survival was calculated in comparison to bacteria in plasma alone. Data represent the mean ± SEM (n = 3). * (p < 0.05) indicates a significant difference between wild-type and mutant strains.

Fig 6. Presence of CPS modulates S. suis-induced pro-inflammatory mediator production by dendritic cells (DCs) regardless of serotype.

Pro-inflammatory mediator production by DCs following infection with the S. suis serotype 2 (S2), 9 (S9) or 14 (S14) wild-type and mutant strains after 16 h of incubation, as measured by ELISA. Production of TNF (A), IL-6 (B), CXCL1 (C), CCL2 (D), and CCL3 (E). Data represent the mean ± SEM (n = 4). C- denotes cells in medium alone. * (p < 0.05) indicates a significant difference between wild-type and mutant strains.

Fig 7. MyD88-dependent Toll-like receptor (TLR) signaling is required for S. suis-induced pro-inflammatory mediator production by dendritic cells (DCs) regardless of serotype.

Pro-inflammatory mediator production by wild-type, MyD88^-/-, TLR2^-/-, and TLR4^-/- DCs following infection with the S. suis serotype 2 (S2), 9 (S9) or 14 (S14) wild-type strains after 16 h of incubation, as measured by ELISA. Production of TNF (A), IL-6 (B), CXCL1 (C), CCL2 (D), and CCL3 (E). Data represent the mean ± SEM (n = 4). C- denotes cells in medium alone. * (p < 0.05) indicates a significant difference between wild-type and knockout cells.

Fig 8. CPS is required for S. suis serotype 9 virulence and persistence in blood in a mouse model of infection.

Survival (A) and blood bacterial burden 12 h (B) and 24 h post-infection (C) of CD-1 mice following intraperitoneal inoculation of the S. suis serotype 9 wild-type (S9) or mutant (S9ΔcpsG) strain. Data represent survival curves (A) or geometric mean (B & C) (n = 15). * (p < 0.05) indicates a significant difference between survival or blood bacterial burden of mice infected with wild-type and mutant strain.

Fig 9. Reduced plasma pro-inflammatory mediator levels in the absence of S. suis serotype 9 CPS during systemic infection.

Plasma levels of IL-6 (A), IL-12p70 (B), IFN-γ (C), CCL2 (D), CCL3 (E), CCL4 (F), CCL5 (G), and CXCL2 (H) in mice 12 h following following mock-infection or intraperitoneal inoculation of the S. suis serotype 9 wild-type (S9) or mutant (S9ΔcspG) strain. Data represent mean ± SEM (n = 6). C- denotes mock-infected mice. * (p < 0.05) indicates a significant difference between plasma levels of mice infected with the wild-type strain and mutant strain.