The involvement of MsmK in pathogenesis of the Streptococcus suis serotype 2

Abstract Streptococcus suis serotype 2 (SS2) is an important swine and human pathogen that causes global economic and public health problems. Virulent S. suis strains successfully maintain high bacterial concentrations in host blood and rapidly adapt to challenging environments within hosts. Successful survival in hosts is a major factor influencing the pathogenesis of SS2. We have previously identified that SS2 colonization in mouse brain is possibly affected by the ATPase, MsmK of carbohydrate ATP‐binding cassette (ABC) transporters because of carbohydrate utilization. In this study, the chain length of the msmK deletion mutant was longer than that of the wild type, and the former was significantly more susceptible than the latter when theses strains were exposed to mouse blood both in vivo and in vitro. The hemolytic activity of the mutant strain was decreased. Although the adhesion of the mutant to HEp‐2 cell lines was enhanced, the deletion of msmK impaired the abilities of SS2 to resist phagocytosis and survive severe stress conditions. MsmK contributed to the survival and adaptation of SS2 in host bloodstream. Therefore, MsmK was identified as a multifunctional component that not only contributed to carbohydrate utilization but also participated in SS2 pathogenesis.

mammalian brain contains plenty of glycogen as the sole carbohydrate reserve, which could sustain the growth of SS2 under infection conditions (Brown, 2004;Ferrando et al., 2014). By contrast, glucose concentrations ranging from 4.4 mmol/L to 6.6 mmol/L are present in the bloodstream, where SS2 undergoes metabolism favorable for its optimal growth in a glucose-containing medium (Ferrando et al., 2014).
S. suis can sense the nutrient supply and subsequently modulate the expression of its virulence factors. Capsular polysaccharide (CPS) expression increases when bacteria are grown in pig blood rather than in cerebrospinal fluid . CPS expression is also directly associated with carbohydrate metabolism in gram-positive bacteria (Willenborg et al., 2011). Nineteen predicted or confirmed virulence factors, including suilysin, are expressed at high levels when SS2 is supplied with starch or pullulan instead of glucose as a sole carbon source (Ferrando et al., 2014). Thus, carbohydrate availability affects the pathogenetic performance of SS2.
We proposed that MsmK, an ATPase of ABC transporters that transport the degradation products of glycogen for SS2, possibly contributes to the colonization of SS2 in mouse brain because of glycogen utilization (Tan, Gao, Liu, Zhang, & Yang, 2015). We further identified the additional role played by MsmK in the bloodstream because different nutritional conditions exist in host blood and brain. Our results revealed that MsmK is involved in the resistance of SS2 to eradication in host blood and adaptation in host. Our study increased the number of carbohydrate metabolism components known to play specific roles in pathogenesis. Our study also provided further insights into the relationship between carbohydrate contents and SS2 pathogenesis.

| Scanning electron microscopy
Scanning electron microscopy (SEM) assays were performed in accordance with previously described methods (Shi et al., 2014). SC-19, ΔmsmK, and CΔmsmK were grown at 600 nm optical density (OD 600 ) of 0.8 (mid-log phase) and spotted onto polylysine coverslips (WHB, Shanghai, China). The bacteria were fixed with 2.5% glutaraldehyde at 4°C overnight. The subsequent dehydration steps with ethanol were conducted as follows: 30% for 15 min, 50% for 15 min, 70% for 15 min, 90% for 15 min, 100% for 15 min, and 100% for 15 min. The dried samples were covered with a 10 nM thick gold layer and observed with a JSM-6390LV SEM (NTC, Tokyo, Japan).

| Transmission electron microscopy
Transmission electron microscopy (TEM) assays were performed in accordance with previously described methods (Gao et al., 2016). SC-19, ΔmsmK, and CΔmsmK were harvested at OD 600 of 0.8 and fixed with 2.5% glutaraldehyde overnight. The samples were then treated with 2% osmium tetroxide for 2 hr and dehydrated in a serial dilution of ethanol. The dehydrated cells were embedded in epoxy resin and their morphological characteristics were observed using an H-7650 TEM (Hitachi, Tokyo, Japan). Twenty bacterial cells from each strain were randomly chosen from the TEM micrographs to measure the thickness of capsule using Image J. The cells were then statistically analyzed with GraphPad Prism 6.

| Bacterial survival in mouse blood after infection
Survival assay was performed in accordance with previously described methods (Marion, Aten, Woodiga, & King, 2011;Tan et al., 2015).
Four-to six-week-old specific-pathogen-free (SPF) female Kun-Ming mice were inoculated intraperitoneally with 3 × 10 7 mid-log-phase cells at a 1:1 mixture of SC-19 and ΔmsmK. Five control mice were inoculated with normal saline. Five mice from each group were killed through carbon dioxide asphyxiation at 12 hr, 1 day, 3 days, or 5 days postinfection. Blood samples were collected, serially diluted, vortexed, and plated on TSA plates. TSA with streptomycin was prepared for the wild-type (WT) SC-19, whereas TSA with streptomycin and erythromycin was selective for ΔmsmK. Colonies were counted and presented as colony-forming unit (CFU)/ml. Data were drafted and analyzed with GraphPad Prism 6.

| In vitro bacterial survival in the presence of mouse whole blood
Assays were performed in accordance with previously described methods (de Buhr, Neumann, Jerjomiceva, von Kockritz-Blickwede, & Baums, 2014). SC-19, ΔmsmK, and CΔmsmK were cultured in the early stationary phase (OD 600 of 1.2). The bacteria were suspended in normal saline at OD 600 of 0.2. Subsequently, 1 ml of whole blood or 1 ml of TSB with 10% NBS was mixed with 100 μl of SS2 cells and incubated for 2 hr at 37°C. The incubated mixtures were harvested at 0 and 2 hr, serially diluted, vortexed, and plated onto TSA plates to determine the bacterial survival factors. Survival factors were calculated as the ratio of the data at 2 hr to the data at 0 hr. S. suis strains grown in broth were regarded as positive control. The data at 0 hr were considered as 1. The assays were performed in triplicate and repeated thrice.

| Hemolytic activity detection
Hemolytic activity (HA) was detected in accordance with previously described methods (Ferrando et al., 2014). The supernatant was collected from 1 ml of cultures in the early stationary phase through centrifugation at 12,000g for 1 min. Each 100 μl of the supernatant was incubated for 2 hr at 37°C with 100 μl of 2% washed sheep erythrocytes. Unlysed red blood cells were allowed to pellet through centrifugation. Afterward, 100 μl of the supernatant was transferred to a sterile new microplate. Absorption was determined at 550 nm in a microplate reader (Biotek, Vermont, USA). Sterile culture media were used as negative controls. The experiments were conducted in triplicate to detect HA and repeated at least thrice.

| Adherence and invasion assays
Adherence and invasion assays were conducted in accordance with previously described methods (Li, Wan, Tao, Chen, & Zhou, 2013).
HEp-2 cells (1.6 × 10 5 cells per well) were infected with log-phase SS2 strains (OD 600 of 0.8) to reach a multiplicity of infection (MOI) of 100:1 (bacteria:cells) and incubated at 37°C for 2 hr. HEp-2 cells were then lysed in 1 ml of sterile distilled water to count the adherent and intracellular bacteria. The appropriate diluted lysates were vortexed and plated on TSA plates. HEp-2 cells were also incubated with ampicillin (100 μg/ml) for 2 hr before lysis. Plating was performed to count the intracellular bacteria alone. The experiments were performed in triplicate wells. All experiments were performed thrice.

| Phagocytosis assay
Phagocytosis assay was performed in accordance with previously described methods (Li et al., 2013). RAW264.7 cells were infected with log-phase SS2 strains (OD 600 of 0.8) to reach a MOI of 10:1. Penicillin (100 μg/ml) was added for 2 hr of incubation to kill the extracellular bacteria before lysis and plating were performed. Viable intracellular bacteria were determined by plating serial dilutions on TSA plates. The experiments were performed in triplicate wells and repeated thrice.

| Osmotic stress assay
The adaptability of the SS2 strains to osmotic stress was evaluated by monitoring bacterial growth in TSB containing 10% NBS with 0.4 mol/L NaCl in accordance with previously described methods . The overnight cultures of SC-19, ΔmsmK, and CΔmsmK were diluted in a fresh medium with or without NaCl to obtain OD 600 of 0.2. The samples were incubated at 37°C for 8 hr. Bacterial growth was monitored at 1 hr interval by determining OD 600 . The assays were performed in triplicate and repeated thrice.

| Oxidative stress assay
The SS2 strains were challenged with H 2 O 2 to evaluate oxidative stress tolerance in accordance with previously described methods . The sensitivity to H 2 O 2 of bacteria was examined by exposing culture aliquots to 0, 20, and 30 mmol/L H 2 O 2 for 20 min at 37°C. The results were presented as the percentage of survival. The assays were performed in triplicate and repeated thrice.

| RNA isolation, reverse transcription, and quantitative real-time PCR (qRT-PCR)
SC-19 and ΔmsmK were grown in an early stationary phase in a TSB medium to detect the expression levels of sly. SC-19 and ΔmsmK were grown in an early stationary phase in a TSB medium containing 10% NBS to analyze adherence. SC-19 and ΔmsmK were grown for 4.5 hr in a TSB medium containing 10% NBS with/without 0.4 mol/L NaCl for osmotic stress analysis. Total RNA was isolated using an SV total RNA isolation system (Promega, Shanghai, China) according to the manufacturer's instructions. cDNA was synthesized using reverse transcriptase mixture (Vazyme, Nanjing, China) according to the manufacturer's instructions. Parallel samples were processed without the addition of reverse transcriptase as a negative control.

| Biofilm formation assay
Biofilm formation assay was conducted in accordance with previously described methods (Grenier, Grignon, & Gottschalk, 2009). The overnight cultures were diluted at OD 600 of 0.1 and 200 μl of each diluted culture was added to 96-well microplates. The bacteria were incubated for 12 hr and 24 hr at 37°C. Afterward, the medium and free-floating bacteria were removed and stained with 100 μl of 0.04% crystal violet for 10 min. The wells were washed to remove the unbound crystal violet dye and dried for 2 hr at 37°C. After 100 μl of 95% ethanol was added to each well, the plates were shaken for 10 min. Absorbance was recorded at 550 nm. The wells with a sterile medium were used as negative controls. The experiments were run in triplicate and repeated at least thrice.

| Murine infection experiments
Fifty 4-to 6-week-old SPF female Kun-Ming mice were randomly classified into five groups with 10 mice per group. The log phase cultures of SC-19 and ΔmsmK were diluted in normal saline to achieve the following concentrations: 1.0 × 10 9 CFU/0.5 ml and 3.0 × 10 9 CFU/0.5 ml.
Afterward, 0.5 ml of inoculum was intraperitoneally introduced. Ten control mice were treated with normal saline. The infected mice were observed for 1 week and deaths were recorded.

| Microbiological characterization of the mutant
Morphological examination by Gram staining showed that MsmK mutation lengthened the bacterial chain (Figure 1a). The same phenotype was observed in subsequent SEM-based assays (Figure 1b).
The chain length of SS2 was partially restored after complementation.

| MsmK contributes to SS2 survival in mouse blood both in vivo and in vitro
Previous studies proposed that the mutation of msmK affects SS2 survival in infected mouse brains mainly because of the mutant loss of the ability to utilize glycogen in vivo (Tan et al., 2015). Nutritional and immunological conditions are different between the blood and the brain.
The finding indicated that different survival performances of SS2 can be expected in the bloodstream. In our study, in vivo bacterial survival assays in the blood were conducted. In Figure 2a, ΔmsmK manifested significant defects in the blood survival after 24 hr of infection. The maximum growth of ΔmsmK was obtained at 12 hr after infection. By contrast, the peak timepoint of SC-19 was 24 hr. To confirm the susceptibility of the mutant to eradication in the blood, we carried out in vitro assays. Compared with the survival factor of SC-19, the survival factor of ΔmsmK in mouse whole blood decreased significantly ( Figure 2b). The results suggested that ΔmsmK was significantly more susceptible than SC-19 when they were exposed to mouse blood. Therefore, MsmK contributes to SS2 survival in mouse blood.

| Reduced HA of the mutant
We determined the erythrocyte HA of the supernatants of SS2 strains. In Figure 3, the HA of ΔmsmK was significantly lower than that of SC-19 and CΔmsmK. The expression level of sly of ΔmsmK was also lower (−3.46-fold change) than that of SC-19. This finding indicated that the absence of MsmK could influence the HA of SS2. Survival factors of SC-19, ΔmsmK, and CΔmsmK in culture medium and mouse whole blood. Colony-forming unit (CFU) was determined at 0 and 2 hr of incubation on a rotator at 37°C. CFU at 0 hr was designated as 1. Survival factor was calculated as the ratio of the data at 2 hr to the data at 0 hr. Strains grown in culture medium were regarded as positive control. Data are presented as means ± SEM

| msmK absence promotes adhesion in vitro
We investigated whether ΔmsmK changes its abilities to adhere and invade host cells. The adherence efficiencies of SC-19 and its derivatives to the HEp-2 cells were calculated to determine whether the absence of msmK affects the cellular adhesion of SS2. The binding rate of ΔmsmK to the HEp-2 cells was 1.5-fold higher than that of SC-19 ( Figure 4a). By contrast, the msmK mutant did not significantly differ from the WT strain in the invasion assay (Figure 4b).
Transcriptional analysis revealed that sadP, an adhesion that mediates the binding of SS2 to galactosyl-α1-4-galactose-containing host receptors (Kouki et al., 2011), is upregulated to a higher extent in ΔmsmK than that in SC-19 (Figure 4c). The expression levels of other known adhesion factors, such as gapdH, fbpS, aupA, and dltA, also increased ( Figure 4c). These data suggested that the mutation of msmK promotes adhesion in vitro.

| Susceptibility to macrophages
ΔmsmK was more likely phagocytosed by phagocytic cells than the WT (Figure 4d). This result indicated that msmK inactivation significantly decreased the resistance of SS2 to phagocytosis. F I G U R E 3 Hemolysis assay of SS2 strains. The hemolysis production was quantified by analyzing the supernatants collected from SS2 strains grown in culture medium. Data points are the mean ± SEM from three independent experiments performed in quintuplicate

| MsmK deficiency affects the osmotic and oxidative tolerance of SS2
We determined the osmotic and oxidative tolerance of SC-19,

ΔmsmK and CΔmsmK under stress conditions in vitro. Compared
with that of SC-19 and CΔmsmK, the growth of ΔmsmK in 0.4 mol/L NaCl was remarkably inhibited (Figure 5a). A series of genes related to osmotic defence was selected for qRT-PCR assays (Table 1). In Figure 5b, the gene expression levels were higher in both strains grown in 0.4 mol/L NaCl than those of the strains grown without NaCl. The fold changes of ΔmsmK were significantly less than those of SC-19. As a result, the tolerance of ΔmsmK to osmotic stress decreased. These data revealed that MsmK contributed to the resistance of SS2 to osmotic stress.
In contrast to the high sensitivity to NaCl, the sensitivity to 20 mmol/L H 2 O 2 of the mutant was similar to that of the WT strain ( Figure 5c). The survival rate of ΔmsmK decreased significantly at 30 mmol/L H 2 O 2 (Figure 5c). This result revealed that MsmK contributed to the resistance of SS2 to oxidative stress to some extent.

| MsmK deficiency influences biofilm formation
The formation of biofilms by microorganisms is a mechanism that allows them to become persistent colonizers, resist clearance by the host immune system, enhance resistance to antibiotics, and exchange genetic materials (Grenier et al., 2009). In this study, the ability of ΔmsmK to form biofilms was weaker than that of SC-19 after these strains were incubated for 12 or 24 hr (Figure 5d). Therefore, msmK also plays a role in the biofilm formation of SS2.

| Absence of MsmK impairs SS2 infection in mice
The half lethal dose (LD 50 ) of SC-19 in mice is 1.5 × 10 9 CFU (Li, Hu, Liu, Chen, & Zhou, 2011). Mice were intraperitoneally inoculated in parallel with a low dose (0.7 × LD 50 ) of SC-19/ΔmsmK and a high dose (2 × LD 50 ) of SC-19/ΔmsmK to evaluate the effect of MsmK on SS2 infection. In Figure 6, the survival rate of the mutant was higher than that of the WT at the same dosage. This result indicated that the mutant became less virulent than SC-19.

| DISCUSSION
Bacterial carbohydrate metabolism components have been associated with environmental survival, colonization, host-pathogen interaction, and virulence, such as those observed in S. suis (Ferrando et al., 2010(Ferrando et al., , 2014, Streptococcus pneumoniae (Marion, Burnaugh, Woodiga, & King, 2011;Marion, Aten, et al., 2011), Streptococcus mutans (Klein et al., 2010), and group A streptococcus (Shelburne et al., 2009). Among the known ones are pullulanases and the lipid-anchored solute-binding proteins (Abbott et al., 2010). In this study, MsmK, an ATPase of carbohydrate ABC transporters, is a potential pathogenic factor involved in blood killing and host adaptation of SS2. survival of the msmK mutant in host blood was also significantly lower than that of SC-19 in the following days. Figure 2b shows that the survival of the mutant in whole blood decreased significantly in vitro.
These results indicated that MsmK was involved in the survival of this pathogen in host blood. CPS is essential for SS2 survival in blood because of its strong anti-phagocytic properties (Doran et al., 2016).
We observed the capsules of SC-19 and ΔmsmK with TEM and found that the thickness of capsules of the WT and the mutant were similar conditions. Compared with that of SC-19 and CΔmsmK, the growth of ΔmsmK in 0.4 mol/L NaCl was remarkably inhibited (Figure 5a). qRT-PCR further revealed that the expression levels of the related genes were upregulated more remarkably in SC-19 than in ΔmsmK after the strains were exposed to NaCl (Figure 5b). The downregulation of stress response genes may be attributed to the decreased stress tolerance of ΔmsmK. The resistance to high H 2 O 2 levels of the mutant was also weak (Figure 5c). The decreased tolerance of ΔmsmK to environmental stresses may be an important factor causing the defective survival of the mutant in the bloodstream in vivo ( Figure 1a) and brain (Tan et al., 2015) in late infection stages, because mutant bacteria may be less likely to adapt to edematous and pathological environments postinfection.
F I G U R E 6 Survival rates of the mice challenged with SC-19 or ΔmsmK. Each mouse was intraperitoneally inoculated with 3.0 × 10 9 CFU (a) and 1.0 × 10 9 CFU (b) of SC-19 and ΔmsmK, respectively. The mice inoculated with normal saline served as controls. Ten mice were used in each group. CFU, colony-forming unit S. suis is considered a major swine pathogen increasingly isolated from a wide range of mammalian species, including humans, and birds. S. suis is also a normal inhabitant of the intestines of various ruminants (Gottschalk et al., 2010). In summary, MsmK contributed to the resistance to blood killing, phagocytosis, and severe stress, enhanced HA, and promoted the ability of SS2 to infect its hosts (Figures 2-6). This study not only explained the increased sensitivity of ΔmsmK to host blood but also re-