Mechanisms of protective immunity to Staphylococcus aureus infection in humans remain elusive. While the importance of cellular immunity has been shown in mice, T cell responses in humans have not been characterised. Using a murine model of recurrent S. aureus peritonitis, we demonstrated that prior exposure to S. aureus enhanced IFNγ responses upon subsequent infection, while adoptive transfer of S. aureus antigen-specific Th1 cells was protective in naïve mice. Translating these findings, we found that S. aureus antigen-specific Th1 cells were also significantly expanded during human S. aureus bloodstream infection (BSI). These Th1 cells were CD45RO+, indicative of a memory phenotype. Thus, exposure to S. aureus induces memory Th1 cells in mice and humans, identifying Th1 cells as potential S. aureus vaccine targets. Consequently, we developed a model vaccine comprising staphylococcal clumping factor A, which we demonstrate to be an effective human T cell antigen, combined with the Th1-driving adjuvant CpG. This novel Th1-inducing vaccine conferred significant protection during S. aureus infection in mice. This study notably advances our understanding of S. aureus cellular immunity, and demonstrates for the first time that a correlate of S. aureus protective immunity identified in mice may be relevant in humans.

Host genetic factors appeared to be the predominant determinant for S. aureus persistent nasal carriage in humans.

Attempts to develop an efficient anti-staphylococcal vaccine in humans have so far been unsuccessful. Therefore, more knowledge of the antigens that are expressed by Staphylococcus aureus in human blood and induce an immune response in patients is required. In this study we further characterize the serial levels of IgG and IgA antibodies against 56 staphylococcal antigens in multiple serum samples of 21 patients with a S. aureus bacteremia, compare peak IgG levels between patients and 30 non-infected controls, and analyze the expression of 3626 genes by two genetically distinct isolates in human blood. The serum antibody levels were measured using a bead-based flow cytometry technique (xMAP®, Luminex corporation). Gene expression levels were analyzed using a microarray (BµG@s microarray). The initial levels and time taken to reach peak IgG and IgA antibody levels were heterogeneous in bacteremia patients. The antigen SA0688 was associated with the highest median initial-to-peak antibody fold-increase for IgG (5.05-fold) and the second highest increase for IgA (2.07-fold). Peak IgG levels against 27 antigens, including the antigen SA0688, were significantly elevated in bacteremia patients versus controls (P≤0.05). Expression of diverse genes, including SA0688, was ubiquitously high in both isolates at all time points during incubation in blood. However, only a limited number of genes were specifically up- or downregulated in both isolates when cultured in blood, compared to the start of incubation in blood or during incubation in BHI broth. In conclusion, most staphylococcal antigens tested in this study, including many known virulence factors, do not induce uniform increases in the antibody levels in bacteremia patients. In addition, the expression of these antigens by S. aureus is not significantly altered by incubation in human blood over time. One immunogenic and ubiquitously expressed antigen is the putative iron-regulated ABC transporter SA0688.

Expanding knowledge on the humoral immune response in Staphylococcus aureus-infected patients is a mandatory step in the development of vaccines and immunotherapies. Here, we present novel insights into the antibody responses following S. aureus bacteremia. Fifteen bacteremic patients were followed extensively from diagnosis onwards (median 29 days, range 9–74). S. aureus strains (median 3, range 1–6) and serial serum samples (median 16, range 6–27) were collected. Strains were genotyped by pulsed-field gel electrophoresis (PFGE) and genes encoding 19 staphylococcal proteins were detected by polymerase chain reaction (PCR). The levels of IgG, IgA, and IgM directed to these proteins were determined using bead-based flow cytometry. All strains isolated from individual patients were PFGE-identical. The genes encoding clumping factor (Clf) A, ClfB, and iron-responsive surface-determinant (Isd) A were detected in all isolates. Antigen-specific IgG levels increased more frequently than IgA or IgM levels. In individual patients, different proteins induced an immune response and the dynamics clearly differed. Anti-ClfB, anti-IsdH, and anti-fibronectin-binding protein A IgG levels increased in 7 of 13 adult patients (p < 0.05). The anti-IsdA IgG level increased in 12 patients (initial to peak level: 1.13–10.72 fold; p < 0.01). Peak level was reached 7–37 days after diagnosis. In a bacteremic 5-day-old newborn, antistaphylococcal IgG levels declined from diagnosis onwards. In conclusion, each bacteremic patient develops a unique immune response directed to different staphylococcal proteins. Therefore, vaccines should be based on multiple components. IsdA is immunogenic and, therefore, produced in nearly all bacteremic patients. This suggests that IsdA might be a useful component of a multivalent staphylococcal vaccine.