Streptococcus suis is a gram-positive coccus with a capsule. In previous studies, Streptococcus suis is divided into 35 serotypes (type 1–34, type 1/2), according to the different capsular polysaccharide (CPS) antigens (Fittipaldi et al., 2012). However, recent studies indicated that certain serotypes (type 20, 22, 26, and 32–34) are not associated with Streptococcus suis (Hill et al., 2005; Nishibori et al., 2013). Among them, Streptococcus suis serotype 2 (SS2) is the most common and virulent. SS2 primarily infects through wounds, causing acute septicemia, meningitis, arthritis, endocarditis, pneumonia, and other diseases in pigs (Zhu et al., 2006). It can also infect humans, leading to bacterial meningitis or toxic shock-like syndrome (Tang et al., 2006; Jiang et al., 2020).

With the advancement of sequencing technologies, an increasing number of species' genome sequences are being uploaded (Pareek et al., 2011). Particularly, the hybrid assembly of third-generation sequencing data and second-generation sequencing data is currently the mainstream approach for completing bacterial genome drafts (Loman and Pallen, 2015). Whole-genome sequences contribute to deepening our understanding of individual organisms (Orsini et al., 2016). Unfortunately, a single genome cannot reflect how genetic variations drive the pathogenic mechanisms within bacterial species (Tettelin et al., 2005; Maturana and Cárdenas, 2021). With the introduction of the pan-genome concept, the genetic variation trends and scope within bacterial populations can be described (Medini et al., 2005). The pan-genome consists of the core genome present in all strains and the accessory genome present in one or more strains (Vernikos, 2020). Moreover, with the increase in the number of genomes, core genomes and accessory genomes may exhibit differences in functionality, metabolic pathways, and resistance (Rasko et al., 2008; Mira et al., 2010). Currently, there have been reports of an open pan-genome composed of 19 SS2 strains, showing notable differences in pili and prophage regions (Guo et al., 2021). However, due to the limited number of samples, the specific differences in the core genome and accessory genome remain unknown.

The development of SS2 disease typically begins with the colonization of bacteria in the upper respiratory tract (Xia et al., 2019). During this process, various virulence factors of SS2 contribute to initial adhesion, immune evasion, and host invasion (Ji et al., 2016; Li et al., 2017). For example, CPS is the most representative virulence factor of Streptococcus suis (Zheng et al., 2013). It not only forms the basis for serotyping but also protects the bacteria from phagocytosis (Xia et al., 2019). Although several virulence factors have been proven to play a role in the early stages of infection, there are still quite a few virulence factors that remain unclear.

Currently, the most effective approach for treating Streptococcus suis involves antibiotics, as there are no commercially available vaccines (Palmieri et al., 2011). However, with the overuse of antibiotics in farming, the increasing prevalence of multidrug-resistant strains is limiting the options for effective antibiotics. Studies have reported a slow increase in resistance of Streptococcus suis to tetracyclines and macrolide/lincosamide antibiotics, which are widely used in the global animal sector (Dechêne-Tempier et al., 2023). Hence, it is necessary to explore relevant resistance genes to investigate the types of resistance within the SS2 population.

In this study, a strain of Streptococcus suis serotype 2 was isolated from a pig farm in Sichuan Province, China, and the complete genome sequence of the strain was obtained. Subsequently, we reported a whole-genome association study on the genomic sequences of 230 naturally isolated strains of SS2. This research unveiled significant distinctions between the core genome and accessory genome of SS2, such as in the transport and metabolism of essential substances, resistance to external conditions, and genetic reproduction. Additionally, our findings revealed virulence genes associated with the SS2 infection process and suggested its potential intrinsic antibiotic resistance.

Pan-genome analysis provides a comprehensive understanding of the overall lineage of SS2. It can be used to screen and identify various virulence and antibiotic resistance genes. In this study, the strain cnzyss2-311 was initially assembled into a genomic sketch using third-generation sequencing data, and second-generation sequencing data were used for error correction, resulting in the completion of the full genomic sequence. Subsequently, pan-genome analysis of SS2 revealed that only 1353 genes were shared among different individuals, constituting the core genome (Figure 3). The core genome serves as the essential framework supporting the rest of the genome, rather than the minimal set of genes required for bacterial survival (Medini et al., 2005; Tettelin et al., 2008). If we expand the definition of the core (including genes that are only partially missing in a small fraction of the genomes), the core genome consists of 1,458 genes, including both core and soft genes. These genes are present in at least 95% of the sampled genomes. The results suggested that SS2 has an open pan-genome, and the number of core genome genes does not significantly change with an increasing number of strains and consistent with previous research (Guo et al., 2021). In contrast, genes present in certain strains and those unique to individual strains constitute the accessory genome, comprising 4,337 genes. The accessory genome reflects the genetic diversity and unique genomic features present in specific strains (Kim et al., 2020). Pan-genome analysis provides insights into the genomic variability of SS2, highlighting both the conserved core genes and the flexible accessory genes that contribute to its adaptability and diversity across strains.

The core genome and accessory genome are extensively involved in various aspects of bacterial life activities, including metabolism and genetic variation (Figures 4–6). The core genome primarily focuses on the transport, metabolism, and translation processes of essential substances for life. Similarly, the accessory genome exhibits certain biological characteristics, especially in functions such as carbohydrate transport and metabolism, transcription, etc. These examples indicate that the primary function of the core genome is to control the normal morphology, reproduction, and execution of basic biological functions in bacteria. In addition, genes participating in fundamental biological processes have been discovered in the accessory genome. It is speculated that when the core genome is damaged, the accessory genome may have alternative functions to maintain certain biological processes in bacteria.

Similarly, the core genome and accessory genome exhibit differences in the expression of carbohydrate enzymes (Figure 8). For example, the annotation of the core genome revealed a high abundance of CE and GT. These are associated with the hydrolysis of carbohydrates and the formation of glycosidic bonds (Venegas et al., 2022). It plays a crucial role in bacteria's acquisition of external carbon sources, adhesion to host cells, and biofilm formation (Middleton et al., 2021; Na et al., 2021). In addition, the accessory genome exhibits significantly higher levels of GH compared to the core genome. GH is mainly involved in the hydrolysis of polysaccharide glycosidic bonds, enabling the degradation of complex host polysaccharides, facilitating bacterial acquisition of carbohydrate nutrients, and promoting the colonization of Streptococcus suis (Chen et al., 2020; Hamre and Sørlie, 2020; Redman et al., 2020).

In general, bacterial adhesion to the surface of objects to form a biofilm is crucial for further infecting the host (Fittipaldi et al., 2012). SS2 typically colonizes the upper respiratory tract, and crossing the mucosal barrier is a prerequisite for SS2 to cause invasive infections (Xia et al., 2019). The annotation results of virulence factors also suggest that, aside from undefined capsule-related genes, the most abundantly annotated virulence genes in the pan-genome belong to the adhesion category. Streptococcal lipoteichoic acid rotamase A (SlrA) has been shown to indirectly promote host cell adhesion and invasion, as well as prevent phagocytosis of Streptococcus pneumoniae, making it a potential therapeutic target for preventing bacterial colonization (Cron et al., 2009). Additionally, SS2 possesses a homologous lipoteichoic acid protein gene (slrA), but lacks choline-binding proteins (CBPs; Hermans et al., 2006). Interestingly, the virulence factor annotation results also show that CBPs are completely absent in the core genome of SS2. CBPs have been shown to play a crucial role in the growth, autolysis, and biofilm formation of S. pneumoniae and are essential for its pathogenicity (Galán-Bartual et al., 2015). Moreover, in Gram-positive bacteria, various surface proteins decorated with proteins are accomplished by sortase (Spirig et al., 2011). Although not essential for bacterial viability, sortase may be an important virulence factor, as it is involved in mediating bacterial adhesion to host tissues, host cell entry, evasion and inhibition of the immune response, and acquisition of essential nutrients by surface proteins (Marraffini et al., 2006). Our results show that streptococcal sortase A (srtA) and pilin-sorting enzyme C (srtC) are fixed encoded sortases in SS2. SrtA recognizes specific LPXTG motifs, anchoring a diverse array of functionally distinct proteins to the cell wall, playing a crucial role in adhesion to extracellular matrix (ECM) components (Spirig et al., 2011; Li et al., 2013; Alharthi et al., 2021). Considering the essential role of SrtA in the pathogenic mechanisms of Gram-positive bacteria, it is also considered an ideal target for potential drugs (Spirig et al., 2011). SrtC sorting enzyme not only anchors smaller substrates but also participates in pilus assembly, suggesting that the formation of pili is an inherent ability of SS2 to promote microbial adhesion and biofilm formation (LeMieux et al., 2008; Qian et al., 2018; Faulds-Pain et al., 2019). In addition, during the process of adhering to endothelial cells and epithelial cells in host tissues, SS2 mediates host cell invasion through its own ECM-binding protein, binding to host ECM components fibronectin (FN) and laminin (LN; Schwarz-Linek et al., 2003; Wahid et al., 2005; Ragunathan et al., 2009). Interestingly, highly conserved hexapeptide sequences (LPXTGE) have been identified in the C-terminus of known surface proteins in Gram-positive cocci and play an important role in cell adhesion, such as the aforementioned sortases, fibronectin-binding proteins (FnBPs), and laminin-binding proteins (LNBPs; Fischetti et al., 1990; Tenenbaum et al., 2007; Yamaguchi et al., 2013). However, there is also research indicating that surface proteins encoded by the pavA gene, which lacks a typical LPXTG anchoring motif, still play an important role in the adhesion and virulence of Streptococcus pneumoniae (Pracht et al., 2005). In conclusion, the core genome of SS2 contains diverse adhesion factors and adhesion proteins, and surface proteins with LPXTG anchoring motifs may not be the sole criterion driving the interaction between the bacterium and the host.

Interestingly, in the core genome, enolase (eno) was identified as an essential virulence factor for Streptococcus suis serotype 2 (Xia et al., 2019). It imparts the bacterium with the ability to directly produce toxicity to porcine brain microvascular endothelial cells (PBMECs), promoting cell apoptosis, inhibiting the expression of tight junctions, or activating the ERK and p38MAPK signaling pathways in porcine brain microvascular endothelial cells to secrete the pro-inflammatory factor IL-8, and increasing the permeability of the blood-brain barrier (Liu H. et al., 2021). Therefore, even in strains considered non-virulent or low-virulent, there is still a possibility of penetrating the blood-brain barrier, which may be one of the reasons for SS2 causing host meningitis. Moreover, classical virulence factors of Streptococcus suis, such as sly, muramidase-released protein (mrp), and extracellular factor (ef), are found exclusively in the accessory genome (Li et al., 2017). For example, suilysin (sly), as one of the cholesterol-dependent bacterial cytolysins, exhibits direct cytotoxicity (Vötsch et al., 2019). Sly induces cell membrane rupture, decreased cytoplasm density, and even exudation through perforation, leading to lesions and damage to epithelial cells and fibroblasts from different host sources (Liu M. et al., 2021). It is a recognized important virulence factor of SS2. This phenomenon suggests that these virulence factors could serve as criteria for assessing the virulence of SS2, as their presence varies among different strains, as previously reported.

In addition, only in the core genome, virulence genes related to anti-phagocytosis and phagosome capture, such as cps2J, cdsA, and ndk, were found. Cps is a recognized virulence factor, not only an important basis for the serotype classification of Streptococcus suis, but also an important factor hindering phagocytosis. Its presence requires the combined action of immunoglobulins and complement to promote phagocytosis (Zhao et al., 2015). In addition, researchers have found that the capsule switch in Streptococcus suis can be achieved through gradual evolution with a combination of minor mutation, deletion, and recombination in the cps locus (Zhu et al., 2020). Therefore, this may also be the reason why cps genes were found in both the core genome and accessory genome. Phosphatidylglycerol transferase, encoded by cdsA gene, is a major enzyme in the synthesis of cell membrane phospholipids (Adams et al., 2017; Sawasato et al., 2019). Although studies have found that it mediates resistance to daptomycin in streptococci and enterococci and may resist innate and exogenous antimicrobial peptide attacks, the specific anti-phagocytic mechanism remains unclear (Mishra et al., 2017; Tran et al., 2019). In addition, nucleoside diphosphate kinase (ndk) is a nucleotide metabolism enzyme that not only maintains the ribonucleotide and deoxyribonucleotide pools in cells but also participates in the regulation of virulence-related traits related to quorum sensing systems (QS), Type III Secretion Systems (T3SS), and virulence factor production systems in Pseudomonas aeruginosa (Neeld et al., 2014; Yu et al., 2016, 2017). Moreover, the ndk in Mycobacterium tuberculosis has macrophage toxicity and plays a crucial role in evading the host immune system's elimination (Chopra et al., 2003). In summary, the core genome serves as the fundamental assurance for the bacterium to invade the host, while the virulence factors present in the accessory genome complement the core genome advantageously, providing various pathways for bacterial adhesion to the host and evasion of the immune system. Moreover, various virulence factors existing in the core genome indicated that strains previously considered non-virulent may still contribute to infection. The diverse virulence factors present in the accessory genome can be viewed as an effective complement to the core genome, influencing the pathogenicity of the bacterium. The redundancy of various virulence factors in Streptococcus suis allows them to compensate for the loss of another factor. Finally, non-virulent strains also have the possibility of acquiring virulence genes and transforming into virulent strains (Griffith, 1928; Tribble et al., 2012).

It is generally believed that antimicrobial resistance can be categorized into intrinsic resistance and acquired resistance based on its underlying causes (Olaitan et al., 2014; Zhang and Feng, 2016). While acquired resistance is commonly considered the major factor leading to the development of resistant bacteria, intrinsic resistance may still pose challenges to the treatment of bacterial infections (Sirichoat et al., 2020). According to the annotation results from the CARD database, the core genome of SS2 contains two types of antibiotic resistance genes, namely patB and vanY (Figure 10 and Table 9). PatB is a natural resistance gene in Streptococcus pneumoniae, conferring resistance to fluoroquinolone antibiotics through the ABC family (El Garch et al., 2010). This result supported previous research, indicating that SS2 achieves efflux of fluoroquinolones through SatAB (an ABC transporter homologous to PatA and PatB; Escudero et al., 2011). Gram-positive bacteria like lactobacilli have natural resistance to glycopeptide antibiotics (such as vancomycin) controlled by the vanY gene (Finch et al., 2010; Alby and Miller, 2018). The results suggest that SS2 may possess natural resistance to fluoroquinolone and glycopeptide antibiotics. Remarkably, despite the single strain showing sensitivity to ofloxacin (S, fluoroquinolones) based on MIC results, it demonstrated resistance to levofloxacin (I, fluoroquinolones), trovafloxacin (S, fluoroquinolones), and vancomycin (S, Glycopeptides; Table 4). The distinct phenotypes of the strain toward these three fluoroquinolones suggest that the resistance mechanism mediated by patB may not uniformly affect antibiotics of the same class.

Additionally, bacteria can acquire resistance to a particular class of antibiotics through various mechanisms such as mutation, transformation, and integration of exogenous DNA (Magiorakos et al., 2012; Impey et al., 2020). The CARD annotation for the single strain identified only four classes of resistance genes, and the represented resistant drugs align well with the MIC results. When compared to the CARD annotation results for the individual strain cnzyss2-311, the pan-genome shows a more diverse overall resistance profile with a greater variety of resistance mechanisms (Figure 10). Notably, the tetracycline resistance gene tet was carried at high frequencies, consistent with previous reports (Haenni et al., 2018). Therefore, the accessory genome is a major contributor to bacterial acquisition of antibiotic resistance. Our results indicate a close correlation between antibiotic resistance phenotypes and the composition of CARD-related genes, highlighting the accessory genome as a primary source of antibiotic resistance. The acquisition and loss of resistance genes contribute to the challenges in clinical treatment. Antibiotic therapy is currently a major approach for managing Streptococcus suis infections in pigs, and this may lead to the widespread dissemination of antimicrobial resistance (AMR) genes, adding pressure to clinical treatment. Thus, it is essential to regulate antibiotic use in clinical practice. These annotation results indicate that SS2's main reservoir of resistance genes comes from the accessory genome. Bacteria can acquire exogenous resistance genes through various means, and the continuous acquisition or loss of resistance genes plays a crucial role in adapting to environmental pressures. This phenomenon is likely a direct cause of the current challenges in clinical treatment.

In this study, various databases and annotation tools were employed, as shown in Table 1. Specifically, eggnog-mapper was used for alignment with the COG database, providing annotation results including COG annotations, GO annotations, and KEGG annotations. Considering that the GO descriptions in the COG database might not be the latest version, we conducted a reannotation using the Interproscan database. Additionally, we excluded some descriptions unrelated to bacteria in the KEGG annotations, such as human cancers, to ensure the accuracy of KEGG annotations.

Our analysis has some limitations. Firstly, we are confined to publicly available genomes retrieved from NCBI, including a total of 229 isolates, and the genomes assembled in this study. Lastly, as our analysis is limited to assumed proteins in the database, further experiments are required to validate these presumed genes associated with virulence and drug resistance.

In summary, we have demonstrated the characteristic differences between the core genome and accessory genome of SS2. Additionally, we have identified potential virulence genes associated with SS2, laying the foundation for the exploration of new virulence factors. During the early stages of infection, multiple genes influence the adhesion of the bacterium and may contribute to the pathogenicity of SS2. The presence of diverse virulence factors in SS2 suggests redundancy, possibly serving as advantageous supplements to other virulence factors. Finally, our findings indicate that SS2 may exhibit natural resistance to fluoroquinolone and glycopeptide antibiotics.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary material.

YZ: Writing – original draft, Writing – review & editing. TT: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – review & editing. XY: Data curation, Project administration, Resources, Supervision, Writing – review & editing. YL: Data curation, Investigation, Project administration, Writing – review & editing. ZY: Conceptualization, Data curation, Investigation, Visualization, Formal analysis, Funding acquisition, Methodology, Project administration, Resources, Software, Supervision, Validation, Writing – review & editing. MR: Data curation, Formal analysis, Supervision, Software, Validation, Visualization, Writing – review & editing. GZ: Conceptualization, Software, Visualization, Methodology, Supervision, Validation, Writing – review & editing. YY: Project administration, Funding acquisition, Visualization, Writing – review & editing. AL: Formal analysis, Resources, Supervision, Validation, Visualization, Writing – review & editing. YW: Conceptualization, Funding acquisition, Validation, Writing – original draft, Writing – review & editing.