Streptococcus suis is an important Gram-positive encapsulated bacterium that mainly causes disease in weaned piglets (Gottschalk and Segura, 2019) and is also capable of zoonotic infections (Gottschalk et al., 2010). Pig farms around the globe are considered positive for S. suis as it is a natural inhabitant of the upper respiratory tract of swine (Higgins et al., 2013). In the presence of virulent strains, S. suis infection in pigs can lead to serious life-altering conditions such as septicemia leading to sudden death, meningitis, arthritis and endocarditis (Gottschalk and Segura, 2019). Prompt recognition of early clinical signs related to S. suis infection followed by treatment with appropriate antibiotics maximizes the survival of affected pigs (Gottschalk, 2020). However, prudent usage of antimicrobials is recommended as multiple strains of S. suis presenting antimicrobial resistances have been detected worldwide (Varela et al., 2013; Hadjirin et al., 2021; Uruen et al., 2022) and such strains represent health risks as possible antibiotic resistance reservoirs able to spread resistance genes to other pathogenic streptococci (Palmieri et al., 2011; Uruen et al., 2022). In this light, vaccines have become highly desirable to prevent S. suis infections on the farm and reduce antimicrobial usage. Multiple efforts have been made to develop efficacious vaccines against S. suis, however, there is currently no effective universal vaccine commercially available (Segura, 2015).

An important issue encountered in the development of a universal vaccine is the high genotypic and phenotypic diversity of S. suis isolates. Indeed, there are 29 distinct serotypes of S. suis that are defined by the immunogenicity of the capsular polysaccharide (CPS) that surrounds the bacteria (Okura et al., 2016). Additionally, significant differences within strains of a same serotype of S. suis have been uncovered by multilocus sequence typing (MLST) and finally, the distribution of serotypes and sequence types (ST) also highly differs between regions (Goyette-Desjardins et al., 2014) which makes the development of a universal cross-protective vaccine complicated. Most studies aimed to understand the immune response, protective antigens and vaccine development have focused on serotype 2 (Segura, 2015). Regarding the other serotypes of S. suis, studies remain scarce (Rieckmann et al., 2020). In the absence of efficacious commercial vaccines to prevent diseases caused by S. suis, the industry has resorted to the use of autogenous vaccines made by licensed laboratories. Autogenous vaccines are formulated with killed bacteria (bacterins) made from one or many S. suis strains isolated from the farm and are then only administered to the animals of that same farm with the objective to provide protection against the local circulating strains. The few controlled experimental studies on S. suis bacterins have shown contradictory results (Segura, 2015; Corsaut et al., 2020; Rieckmann et al., 2020), and there is no consistent evidence that bacterins providing protection against one serotype could provide protection against other strains and serotypes (Baums et al., 2009).

As there is still a lot of uncertainty regarding the efficacy of S. suis vaccines, more knowledge is needed to achieve better understanding of what influences cross-protection. One bacterial component that could influence such cross-protection would be the CPS that protects S. suis. Indeed, the CPS of S. suis serotype 2 is a known critical virulence factor implicated in many functions such as the impairment of phagocytosis and cytokine release by innate immune cells by masking bacterial surface components, such as proteins (Lecours et al., 2011). The biochemical structure of the CPS can highly differ between S. suis serotypes (Goyette-Desjardins et al., 2020) and differential exposure of cell wall components (mainly proteins) could influence host cell interactions and virulence (Okura et al., 2021). However, it remains unknown if these differences in the CPS between S. suis serotypes can influence the cross-protective potential of antibodies directed against sub-capsular antigens. In this study, serotype-switched mutants of S. suis were used to compare the effect of CPS structures of serotypes 2, 3, 4, 7, 8, 9 and 14 on antibody production, antibody binding and antibody-induced bacterial elimination. All the mutants possess the same sub-capsular antigens (those of the serotype 2 background) but express different CPS at their surfaces. By using this strategy, only the effect of the CPS on the antibody response against the same sub-capsular antigens was evaluated.

The biochemical composition of the CPS that surrounds S. suis is known to be widely different between serotypes (Goyette-Desjardins et al., 2020). Our leading hypothesis was that these different CPS structures could differently influence the establishment and role of the antibody response against sub-capsular antigens (cell wall components). Strains expressing the CPS of serotypes 3 and 4 had markedly impaired survival in murine and porcine whole blood (Okura et al., 2021) and showed increased susceptibility to host defenses as they were more significantly eliminated than the parent SS2 strain during primary and secondary infections in mice. Albeit less pronounced, reduced blood bacterial burdens were also observed after primary infection with strains SS2to7 and SS2to9 in our mouse infection model (sub-lethal infection dose). However, the difference in bacteremia levels between these mutants and the parent SS2 strain was lost after a secondary infection and not noticeable in a high-dose mouse infection model (Okura et al., 2021). Altogether, these observations suggest that CPS structures of serotype 7 and 9 provides a similar protection against host defenses than that of CPS serotype 2. Interestingly, mutant SS2to14 behaves similarly than the parent strain SS2 after primary infection, as also reported in the high-dose mouse infection model (Okura et al., 2021). However, the capacity of this mutant to multiply in blood after a secondary infection was significantly impaired. This prompted us to evaluate the adaptive humoral response generated during infection, which has been known to be of importance for S. suis elimination (Holt et al., 1989; Blouin et al., 1994). Production of antibodies following a primary infection with S. suis was not affected by CPS structure, except for IgG production in mice infected with the CPS 3 mutant. We hypothesized that the quick elimination of the SS2to3 mutant in vivo (compared to SS2 and other serotype-switched mutants) might have reduced the quantity of bacterial antigens available to activate B cells and drive optimal IgG production. However, no clear correlation was found between IgG titers and bacterial blood burden (data not shown), although it is possible that the relatively low number of subjects used (n = 8-9) limited this analysis. Antibody production profiles were slightly different after the secondary infection where mice infected with CPS 3, 4 and 14 mutants had a significant reduction in anti-SS2 IgM and IgG production compared to the SS2 strain, which could again be attributed to less antigens available.

Protection from antibodies might be influenced by CPS structure as the CPS 14 mutant was significantly more eliminated in the presence of antibodies than the parent SS2 strain during a secondary infection. This observation came off as a surprise since mice infected with CPS 14 shared the same levels of antibodies against the bacteria than other strains after a primary infection (day 14). Enhanced elimination could be due to anti-CPS antibodies; however, anti-CPS titers are practically absent in animals infected with serotype 14 strains (Calzas et al., 2017). This observation led to the hypothesis that the different CPS structures might influence the ability of antibodies to bind to cell wall components exposed at the bacterial surface, which could then enable elimination by immune cells.

Flow cytometry experiments revealed that CPS 3, 4 and 14 mutants bound more IgG antibodies than the SS2 strain. This finding goes in hand with previous evidence that CPS 3, 4 and 14 mutants were more eliminated in vivo and would suggest that these CPS structures, and notably that of serotype 14, are more permeable to antibodies. On the other hand, CPS 7 and 9 conferred more protection against IgG binding than the parental CPS 2, which might explain the capacity of these strains to reach bacteremia levels similar to SS2 after the secondary infection. Taken together, these observations suggest that the different CPS structures of S. suis have different antigen masking properties, which could affect survival against the humoral response.

A controlled in vitro OPA using sera from mice infected with parent SS2 strain confirmed that the CPS structure could influence protection against antibodies directed against sub-capsular antigens. Since the serum used in the OPA was generated against SS2, any reaction with the serotype-switched mutants (expressing a different CPS) would be mainly directed against proteins. CPS mutants 3 and 14 were susceptible to antibody-mediated killing while the CPS 9 mutant showed resistance, indicating that CPS structures influence antigen availability and antibody detection which in turn influences bacterial elimination. It is worth mentioning that SS2 strain was significantly more eliminated than any of the tested strains in our OPA assay. This is because the sera used for all OPA trials was collected from a group of mice infected twice with the homologous SS2 strain and was positive for anti-CPS2 IgMs which are potent antibodies known to be able to induce high levels of opsonophagocytosis killing (Goyette-Desjardins et al., 2019). However, in the absence of specific anti-CPS IgMs, antibody-dependent phagocytosis (directed against sub-capsular antigens) of the mutant strains is affected by the CPS expressed at the bacterial surface.

The evidence gathered in this study indicates that the structural differences of S. suis CPS does influence the development and the role of the antibody response. Of the tested serotypes, the CPS of serotypes 2, 7, 8 and 9 provided better protection while the CPS of serotypes 3, 4 and 14 provided the weakest defense. It could be hypothesized that a more efficacious CPS structure would permit strains to persist in the host and cause disease more easily. This could help explain the overwhelming presence of serotype 2 strains in both pig and human cases world-wide (Goyette-Desjardins et al., 2014), and the rising incidence of serotype 9 strains, which are particularly virulent in European farms (Segura et al., 2020). However, serotype 3 strains are highly prevalent in the USA (Goyette-Desjardins et al., 2014) despite being equipped with a CPS that does not offer optimal protection and is also considered ‘‘highly’’ immunogenic (Goyette-Desjardins et al., 2020). Taking this into account, the ability of the CPS structure to influence antibody development and role is thus not solely responsible for virulence. Indeed, strain dependent differences can be observed within a same serotype (Auger et al., 2018) suggesting that other bacterial components might also play an important role in virulence.

There are currently over 60 published studies on S. suis vaccine candidates catalogued by PubMed, with some studies reporting efficacy against specific strains of S. suis. However, multiple serotypes of S. suis are present in herds (Reams et al., 1996) and mixed infections caused by different serotypes have been reported (Johannson, 2006). In this light, it would be ideal to develop a ‘‘universal’’ cell-surface protein vaccine that offers cross-protection against multiple strains and serotypes of S. suis. However, the results obtained in this study suggest that obtaining cross-protection might be more difficult than originally thought. Indeed, an “universal’’ vaccine made with highly conserved antigens shared between most serotypes and strains of S. suis could elicit strong antibody responses against said antigens but still fail to provide protection (Dumesnil et al., 2019; Weisse et al., 2021), probably due to the fact that CPS structures provide different levels of antigen masking and protection. As an indicator, autovaccines for S. suis serotype 9 have been found to be less efficacious than ones developed against the serotype 2 (Segura et al., 2020), which could be due to the higher level of bacterial protection conferred by the serotype 9 CPS. This study is the first experimental report that directly compares the influence of the CPS structure from multiple serotypes of S. suis on antibody production, antibody binding and antibody-induced bacterial killing. The results obtained indicate that CPS structure plays an important role in antigen availability and bacterial elimination, a finding of importance for vaccine development. Indeed, when anti-CPS specific antibodies are absent, the IgG anti-protein response would have a variable role depending on the capsular serotype of the strain. This study identifies the CPS of serotypes 2, 7, 8 and 9 as a potential high-risk since those CPS confer higher levels of protection against antibodies. Overall, this study highlights the need to closely monitor cross-protection when designing S. suis vaccines.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by Animal Welfare Committee of the University of Montreal.

DD and MS conceived the study. DD and ML performed all animal infections and sampling. DD analyzed the animal data. DD performed the ELISA and OPA experiments. ML performed the FACS experiments and DD analyzed the FACS data. DD performed statistical analyses. DD prepared tables and figures. DD and MS interpreted the data. DD and MS wrote the paper. MO and DT provided the mutants. All authors read and approved the final manuscript.