Antibodies to Staphylococcus aureus serotype 8 capsular polysaccharide react with and protect against serotype 5 and 8 isolates

Abstract

Most Staphylococcus aureus isolates produce either a serotype 5 (CP5) or 8 (CP8) capsular polysaccharide, and the CP antigens are targets for vaccine development. Since CP5 and CP8 have similar trisaccharide repeating units, it is important to identify an epitope shared by both CP5 and CP8. To characterize cross-reactivity between CP5 and CP8, the immunogenicity of CP5 and CP8 conjugate vaccines in mice and rabbits was evaluated by serological assays. Immune sera were also tested for functional activity by in vitro opsonophagocytic-killing assays and a murine bacteremia model. Antibodies to the CP5-cross-reactive material 197 (CRM197) conjugate vaccine bound only to purified CP5. In contrast, antibodies to the CP8-CRM conjugate vaccine reacted with CP8 and (to a lesser extent) CP5. De-O-acetylation of CP5 increased its reactivity with CP8 antibodies. Moreover, CP8 antibodies bound to Pseudomonas aeruginosa O11 lipopolysaccharide, which has a trisaccharide repeating unit similar to that of the S. aureus CPs. CP8-CRM antibodies mediated in vitro opsonophagocytic killing of S. aureus expressing CP5 or CP8, whereas CP5-CRM antibodies were serotype specific. Passive immunization with antiserum to CP5-CRM or CP8-CRM protected mice against bacteremia induced by a serotype 5 S. aureus isolate, suggesting that CP8-CRM elicits antibodies cross-reactive to CP5. The identification of epitopes shared by CP5 and CP8 may inform the rational design of a vaccine to protect against infections caused by CP5- or CP8-producing strains of S. aureus.

FIG 1

S. aureus CP5 and CP8 conjugated to protein elicit capsular antibodies in mice. (A) CP5 and CP8 have similar trisaccharide repeating units comprised of ManNAc, l-FucNAc, and d-FucNAc. (B and C) Groups of ICR mice (n = 18) were immunized with 2.5 μg of CP5-CRM or CP8-CRM on days 0, 5, and 10. Each mouse serum was diluted 1:100 and tested by ELISA on plates coated with purified CP5 (B) or CP8 (C). The ELISA index was determined by dividing the absorbance reading of the test serum by the absorbance reading of a pool of high-titer immune mouse sera included on the same plate. The data shown are mean ELISA indices ± standard errors.

FIG 2

Rabbit antibodies to CP8-Epa and CP8-CRM cross-react with CP5. Purified rabbit IgG to CP5-Epa or CP8-Epa (A and B) or rabbit immune serum to CP5-CRM or CP8-CRM (C and D) was serially diluted and tested by ELISA on plates coated with purified CP5, de-O-acetylated-CP5, CP8, or de-O-acetylated CP8. The data represent means (± standard errors) of three independent experiments. OD, optical density.

FIG 3

Serologic reactivity of CP5- and CP8-specific antiserum to Epa, CP8-Epa, and CP5-Epa assessed by immunoblotting. Epa, CP8-Epa, and CP5-Epa were separated by SDS-PAGE and detected by serum antibodies produced by immunization with rabbit serum raised to killed encapsulated bacteria (CP5⁺ or CP8⁺) and absorbed with acapsular strains to render the sera CP5 or CP8 specific.

FIG 4

Opsonic activity of serum antibodies generated by immunization with CP5-CRM or CP8-CRM against S. aureus strains. The test mixture contained diluted sera, S. aureus, guinea pig serum as a complement source (concentration, 1%), and differentiated HL60 cells. (A and B) Pooled mouse serum (A) or rabbit serum (B) was tested at a 1:40 dilution. Controls lacking antibody, HL60 cells, and complement were included in each assay, as shown in panel A. (C and D) Serially diluted rabbit serum was analyzed against S. aureus serotype 5 strains Newman (C) and USA100 (D). The results are expressed as percent change in the number of CFU/ml at 2 h compared to the number of CFU/ml at time zero. The data shown are means (± standard errors) of two or more experiments. *, P < 0.05; **, P < 0.01.

FIG 5

Opsonic activity of rabbit sera to CP5-CRM or CP8-CRM against S. aureus Reynolds (CP5) and Reynolds (Δcap5H). The rabbit sera were serially diluted 3-fold, starting at a 1:40 initial dilution. The results are expressed as percent change in the number of CFU/ml after 2 h. The data shown are means (± standard errors) of three experiments. Differences were analyzed by the Student t test. *, P < 0.05; **P < 0.001.

FIG 6

Binding of rabbit antibodies to P. aeruginosa O11 LPS by ELISA. (A) Ninety-six-well plates were coated with P. aeruginosa O11 LPS (2 μg/ml), and antibody samples were serially diluted 5-fold. Rabbit antiserum to O11-Epa was used as a positive control, and preimmune serum and Shigella O1-Epa IgG were used as negative-control sera. The data represent means (± standard errors) of two independent experiments. (B) Repeating units of polysaccharides from different bacterial genera. The shared structure of the repeating units is underlined.

FIG 7

Rabbit antibodies to CP5-CRM or CP8-CRM reduce bacteremia induced by S. aureus Reynolds (CP5). Mice were passively immunized with 0.5 ml of preimmune serum or serum raised to CP5-CRM or CP8-CRM 24 h before challenge with 1 × 10⁷ CFU S. aureus Reynolds (CP5). The data represent the results of quantitative blood cultures on individual animals, and the horizontal lines and numbers represent median numbers of CFU/ml blood for each group (n = 15). Bacteremia levels from mice given immune serum were compared by the Mann-Whitney test to those from mice given preimmune serum.