Human exposure to S. aureus triggers the development of immunity, imprinting an adaptive immune response that is lacking in most current preclinical models. In a recent study, Tsai et al. explored the impact of prior infection on the efficacy of vaccination with the iron-regulated surface determinant B (IsdB) in mice (14). IsdB is a conserved surface protein in S. aureus that is involved in iron acquisition (15) and pathogenesis (16). As with many antigens, IsdB elicits robust and protective responses in naive mice (14, 17). However, in a randomized placebo-controlled human trial in patients undergoing cardiothoracic surgery, IsdB vaccination failed to prevent S. aureus bacteremia and/or deep wound infection (18). Tsai et al. demonstrated that while vaccination of naive animals elicited the expected protective response, vaccination of previously infected mice failed to protect due to the elaboration of antibodies that are impaired in opsonophagocytosis and opsonophagocytic killing (OPK), processes essential for pathogen clearance (14). Interestingly, the reduced OPK activity was linked with the α2,3 sialylation of the anti-IsdB antibodies, a modification that blocks engagement of the Fcγ receptors on phagocytes (19).

In this issue of the JCI, Tsai et al. reports elevated levels of IL-10 in animals previously infected with S. aureus that were subsequently vaccinated with IsdB compared with those without prior infection (20). IL-10 is a multifactorial cytokine that has been linked to increased mortality in patients with S. aureus bacteremia (21). Animals infected with S. aureus had elevated levels of B10 lymphocytes, a type of B cell responsible for the production of IL-10. Mechanistically, IL-10 induced the expression and production of St3gal4 and St6gal2, sialyltransferases that mediated the α2,3 sialylation on the anti-IsdB antibodies. Importantly, protection could be restored in mice that were transferred anti-IsdB antibodies treated with an α2,3 neuraminidase, which removed antibody sialylation and restored OPK. Thus, restoration of antibody-mediated OPK activity was sufficient to overcome vaccine failure in mice. Of note, IL-10 also induced α2,3 sialylation of antibodies elicited upon IsdA and MntC vaccination. Thus, prior S. aureus infection induces IL-10, which limits the vaccine efficacy of at least three different surface antigens (Figure 1).

Imprinting due to prior S. aureus infection blocks vaccine efficacy. In the naive host, vaccination with IsdB elicits opsonic antibodies that promote phagocyte-mediated eradication of S. aureus. In contrast, IsdB vaccination of mice previously infected with S. aureus triggers a recall response accompanied with copious levels of IL-10, which results in the sialylation of the anti-IsdB antibodies, rendering these antibodies ineffective at promoting OPK.

To provide a link between the mouse studies and the human scenario, Tsai et al. (20) examined the glycosylation state of antibodies from healthy individuals and patients with cystic fibrosis (CF), who are known to harbor higher amounts of IL-10 (22). In support of the murine data, antibodies from patients with CF were hyper-α2,3 sialylated compared with antibodies from healthy controls. Moreover, removal of α2,3 sialylation from naturally occurring human anti-IsdB antibodies enhanced OPK of S. aureus. From these findings, we can wonder if the observed linkage between IL-10 and mortality in patients with S. aureus bacteremia (21) could be due to hyper-α2,3 sialylation of natural anti–S. aureus antibodies rendering them ineffective.