Impact of Phosphorylcholine Expression on the Adherence and Invasion of Streptococcus pyogenes to Epithelial Cells

Phosphorylcholine (PC) is a structural component of various pathogens and is involved in bacterial adhesion via the platelet-activating factor receptor (PAF-R). In this study, we investigated how PC expression affects cell adhesion and invasion of Streptococcus pyogenes (S. pyogenes). Eight clinical strains of S. pyogenes were cultured, and PC expression was measured using fluorescence-activated cell sorting. Bacterial adherence and invasion were examined using Detroit 562 cells. An anti-PC-specific monoclonal antibody (TEPC-15) was used to inhibit bacterial PC, and a PAF-R antagonist (ABT-491) was used to inhibit cellular PAF-R. The emm gene was amplified by the polymerase chain reaction with the standard primers. The level of PC expressed on the S. pyogenes surfaces differed in each strain and differed even in the same emm genotype. Adherence assay experiments showed a significant negative correlation between TEPC-15 and ABT-491 inhibitory effects and PC expression in S. pyogenes. Similarly, intracellular invasion assay experiments showed a significant negative correlation between TEPC-15 and ABT-491 inhibitory effects and PC expression in S. pyogenes. This study suggests that S. pyogenes is involved in cell adhesion and invasion by PC.


Introduction
Recurrent tonsillitis is a common disease encountered by otolaryngologists [1]. Viruses are the most common causes of recurrent tonsillitis, followed by bacteria, including Streptococcus pyogenes (S. pyogenes), Haemophilus influenzae, and Streptococcus pneumoniae [2]. S. pyogenes is a species of Gram-positive bacteria that causes respiratory tract infections with mild to modest disease (tonsillitis and pharyngitis) and invasive and potentially life-threatening diseases (cellulitis, necrotizing fasciitis, and toxic shock syndromes) [3,4]. Mortality from lethal S. pyogenes remains high in developed and developing countries [5]. In addition, there are reports of S. pyogenes outbreaks in communities and hospitals [6]. Early diagnosis is required, and a recent study reported that marker genes (spyCEP, ssa, sic, sdaB, speG) can be used as a rapid diagnostic tool for S. pyogenes [7].
One of the first basic stages of S. pyogenes etiology is epithelial adhesion and colonization [8]. The production of reactive oxygen species by pathogen-infected cells is often associated with high levels of cell death, and S. pyogenes induce apoptosis in infected human epithelial cells [4]. Moreover, S. pyogenes induce the transcription of the interleukin (IL)-1a, IL-1b, IL-6, and IL-8 genes and the release of prostaglandin E2 [9].
The surface of S. pyogenes cells incorporates a number of proteins adhesins (e.g., pili, Sfb1 / PrtF1, and M protein) that allow S. pyogenes to colonize distinct tissue sites [10]. The prominent M protein is important for the attachment of S. pyogenes to keratinocytes in skin infections [4]. M protein, a cell surface protein that is a major pathogenic determinant of S. pyogenes, is encoded by the emm gene [11]. The emm genotype of S. pyogenes has more than 170 genotypes and 750 subtypes, and the genotype distribution varies by country or region [11,12].
In contrast, phosphorylcholine (PC) involvement in S. pyogenes' cell adhesion is unknown. PC is a structural component of many bacteria, such as Streptococcus pneumoniae and non-typeable Haemophilus influenzae [13,14]. Bacterial PC binds to the platelet-activating factor receptor (PAF-R) on the cell surface and plays an important role in Streptococcus pneumoniae and non-typeable Haemophilus influenzae [13,14]. However, there is no report on the relationship between the PC expression of S. pyogenes and PAF-R or emm gene.
In this study, we investigated the effect of bacterial surface PC expression on the adhesion and invasion in epithelial cells in S. pyogenes. We also investigated the relationship between PC expression and the emm gene.

Bacteria Culture
From March 2019 to December 2020, S. pyogenes were collected from the oropharynx of patients with recurrent tonsillitis (age 24-42 years) at Kagoshima University Hospital. Bacteria were stored in skim milk containing glycerol at −80 • C and incubated overnight in a 5% CO 2 incubator at 37 • C in sheep blood agar (Nissui Pharmaceutical Co., Ltd., Tokyo, Japan) the day before use. After washing with 0.5% bovine serum albumin-phosphate buffered saline (PBS), we measured the optical density at 580 nm and determined the colonyforming units (CFU). The concentration of S. pyogenes was adjusted to 1.0 × 10 8 CFU/mL, as previously described [8].

Cell Culture
Detroit 562 cells (CCL-138; ATCC, Manassas, VA, USA), which are human pharyngeal carcinoma epithelial cells, were cultured in minimal essential medium (Nacalai Tesque Inc., Kyoto, Japan) in a 5% CO 2 incubator at 37 • C, as previously described [9]. The cells were harvested using trypsin (final concentration, 0.02%) and ethylenediaminetetraacetic acid (final concentration, 0.02%; Nacalai Tesque). Thereafter, 2 × 10 4 viable cells were placed per well on a 96-well BD Falcon tissue culture plate and cultured in a 5% CO 2 incubator at 37 • C for 48 h. It was confirmed that the cells were sufficiently cultured in the culture plate.

emm Genotypes of S. pyogenes
The emm gene encoding the M protein was performed according to the recommended protocol by the Center for Disease Control and Prevention (http://www.cdc.gov/ncidod/ biotech/strep/protocols.html, accessed on 16 April 2020, with minor modifications, as previously described [15].

Adherence Assay
Bacteria were incubated on a sheep blood agar plate overnight at 37 • C in a 5% CO 2 incubator. Bacteria (1.0 × 10 5 CFU/mL) were administered to a 96-well plate and adhered to at 37 • C for 2 h in a 5% CO 2 incubator. Each well was washed 10 times with 200 µL PBS to remove bacteria that did not adhere to the cells. Then, it was treated with 100 µL saponin at 37 • C for 15 min in a 5% CO 2 incubator, and 100 µL of solution from each well was seeded on sheep blood agar plates. After incubation for 12 h, it was counted as a control. To investigate the effect of PC-specific IgA on bacterial adhesion, the number of colonies was counted. Bacteria were treated with TEPC-15 (1 µg/mL) at 37 • C for 1 h in a 5% CO 2 incubator to investigate the effect of PC on bacterial adhesion. Furthermore, PAF-R expressed on the surface of Detroit 562 cells was blocked with a 10 µg/mL PAF-R antagonist (ABT-491; Cayman Chemical, Ann Arbor, MI, USA) at 37 • C for 1 h in a 5% CO 2 incubator.

Intracellular Invasion Assay
Bacteria were incubated on a sheep blood agar plate overnight at 37 • C in a 5% CO 2 incubator. Bacteria (1.0 × 10 8 CFU/mL) were administered to a 96-well plate and invaded at 37 • C for 6 h in a 5% CO 2 incubator. Then, gentamicin (200 µg/mL) was added to each well at 37 • C in a 5% CO 2 incubator for 1 h. Each well was washed 10 times with 200 µL of PBS. It was then treated with 100 µL of saponin at 37 • C for 15 min in a 5% CO 2 incubator, and 100 µL of the solution from each well was seeded on sheep blood agar plates. After incubating for 12 h, the number of colonies was counted. In addition, as in the adherence assay, bacteria were treated with TEPC-15, and cells were treated with ABT-491 to investigate their effects.

Statistical Analyses
Statistical analyses were conducted using SPSS for Mac software (version 22.0; IBM Corp., Armonk, NY, USA). The data were statistically analyzed using the unpaired Student's t-test and Pearson's correlation coefficient (cross-reaction data). The values are presented as means ± standard deviations. The level of significance was set at p ≤ 0.05.

PC Expression and emm Genotype
In S. pyogenes, PC was expressed in all strains, but the intensity differed in each strain (Table 1). In this study, we detected four strains of genotype emm75 and one strain of each genotype emm89, emm28, emm12, and emm11 (Table 1). Furthermore, PC expression tended to be high in emm75 strains, and PC expression was different even in the same emm75 strains (Table 1). PC expression was different among the bacterial strains. In addition, PC expression was different, even with the same emm genotype.

Inhibitory Effects of TEPC-15 on Bacterial Adherence
As shown in Figure 1A, the number of adhered bacteria (strain No. 7) treated with TEPC-15 at 50, 10, and 5 µg/mL significantly decreased (p < 0.05), but not with TEPC-15 at 100 and 1 µg/mL. Based on these results, the inhibitory effects of 10 µg/mL of TEPC-15 or mouse IgA antibody as a control were used on the adherence and intracellular protocol. 100 and 1 μg/mL. Based on these results, the inhibitory effects of 10 μg/mL of TEPC-15 or mouse IgA antibody as a control were used on the adherence and intracellular protocol.

Inhibitory Effects of ABT-491 Dose on Bacterial Adherence
As shown in Figure 1B, the number of adhered bacteria (strain No. 7) treated with ABT-491 at 50, 25, and 10 μg/mL significantly decreased (p < 0.05), but not with ABT-491 at 5 and 1 μg/mL. Based on these results, the inhibitory effects of 10 μg/mL of ABT-491 were used on the adherence and intracellular protocol.

Effects of TEPC-15 and ABT-491 on Bacterial Adherence
There was a significant positive correlation between the PC expression of each bacteria and the number of bacteria adhering to the cells (r = 0.76, p < 0.05) (Figure 2A). Furthermore, the number of adhered emm75 was significantly higher than that of other emm genotypes (p < 0.05) ( Figure 2B). There was a significant negative correlation between the PC expression of each bacteria and the number of bacteria adhering to the cells after pretreatment with TEPC-15 (r = −0.76, p = 0.03) ( Figure 3A). Moreover, a negative correlation was found with ABT-491 (r = −0.73, p = 0.04) ( Figure 3B).

Inhibitory Effects of ABT-491 Dose on Bacterial Adherence
As shown in Figure 1B, the number of adhered bacteria (strain No. 7) treated with ABT-491 at 50, 25, and 10 µg/mL significantly decreased (p < 0.05), but not with ABT-491 at 5 and 1 µg/mL. Based on these results, the inhibitory effects of 10 µg/mL of ABT-491 were used on the adherence and intracellular protocol.

Effects of TEPC-15 and ABT-491 on Bacterial Adherence
There was a significant positive correlation between the PC expression of each bacteria and the number of bacteria adhering to the cells (r = 0.76, p < 0.05) (Figure 2A). Furthermore, the number of adhered emm75 was significantly higher than that of other emm genotypes (p < 0.05) ( Figure 2B). There was a significant negative correlation between the PC expression of each bacteria and the number of bacteria adhering to the cells after pretreatment with TEPC-15 (r = −0.76, p = 0.03) ( Figure 3A). Moreover, a negative correlation was found with ABT-491 (r = −0.73, p = 0.04) ( Figure 3B). 100 and 1 μg/mL. Based on these results, the inhibitory effects of 10 μg/mL of TEPC-15 or mouse IgA antibody as a control were used on the adherence and intracellular protocol.

Inhibitory Effects of ABT-491 Dose on Bacterial Adherence
As shown in Figure 1B, the number of adhered bacteria (strain No. 7) treated with ABT-491 at 50, 25, and 10 μg/mL significantly decreased (p < 0.05), but not with ABT-491 at 5 and 1 μg/mL. Based on these results, the inhibitory effects of 10 μg/mL of ABT-491 were used on the adherence and intracellular protocol.

Effects of TEPC-15 and ABT-491 on Bacterial Adherence
There was a significant positive correlation between the PC expression of each bacteria and the number of bacteria adhering to the cells (r = 0.76, p < 0.05) (Figure 2A). Furthermore, the number of adhered emm75 was significantly higher than that of other emm genotypes (p < 0.05) ( Figure 2B). There was a significant negative correlation between the PC expression of each bacteria and the number of bacteria adhering to the cells after pretreatment with TEPC-15 (r = −0.76, p = 0.03) ( Figure 3A). Moreover, a negative correlation was found with ABT-491 (r = −0.73, p = 0.04) ( Figure 3B).

Effects of TEPC-15 and ABT-491 on Bacterial Invasion
There was a significant positive correlation between the PC expression of each bacteria and the number of bacteria invading the cells (r = 0.94, p < 0.05) ( Figure 4A). Furthermore, when comparing the number of invaded emm75 with other emm types, the number of invaded emm75 was significantly higher (p < 0.05) ( Figure 4B). There was a significant negative correlation between the PC expression of each bacteria and the number of bacteria adhering to the cells after pretreatment with TEPC-15 (r = −0.83, p = 0.01) ( Figure 5A). Moreover, a negative correlation was found with ABT-491(r = −0.90, p < 0.05) ( Figure 5B).

Effects of TEPC-15 and ABT-491 on Bacterial Invasion
There was a significant positive correlation between the PC expression of each bacteria and the number of bacteria invading the cells (r = 0.94, p < 0.05) ( Figure 4A). Furthermore, when comparing the number of invaded emm75 with other emm types, the number of invaded emm75 was significantly higher (p < 0.05) ( Figure 4B). There was a significant negative correlation between the PC expression of each bacteria and the number of bacteria adhering to the cells after pretreatment with TEPC-15 (r = −0.83, p = 0.01) ( Figure 5A). Moreover, a negative correlation was found with ABT-491(r = −0.90, p < 0.05) ( Figure 5B).

Effects of TEPC-15 and ABT-491 on Bacterial Invasion
There was a significant positive correlation between the PC expression of each bacteria and the number of bacteria invading the cells (r = 0.94, p < 0.05) ( Figure 4A). Furthermore, when comparing the number of invaded emm75 with other emm types, the number of invaded emm75 was significantly higher (p < 0.05) ( Figure 4B). There was a significant negative correlation between the PC expression of each bacteria and the number of bacteria adhering to the cells after pretreatment with TEPC-15 (r = −0.83, p = 0.01) ( Figure 5A). Moreover, a negative correlation was found with ABT-491(r = −0.90, p < 0.05) ( Figure 5B).

Discussion
This study showed that PC expression was found in S. pyogenes, and PC expression was different even for the same emm genotype. High bacterial PC expression also in-

Discussion
This study showed that PC expression was found in S. pyogenes, and PC expression was different even for the same emm genotype. High bacterial PC expression also increased the number of bacteria adhered to the cells. In addition, treatment with TEPC-15 and ABT-491 further suppressed adhesion and invasion of bacteria with high PC expression.
The existence of the PC was first discovered in Streptococcus pneumoniae by Tomasz in 1967 [16]. PC are small hapten moieties that bind to teichoic acid in the cell walls of several Gram-positive bacteria, including Streptococcus pneumoniae [17]. In addition, the phase variation of PC expression provides a mechanism for Haemophilus influenzae to display a variety of phenotypes, allowing rapid adaptation to different host environments [18]. This study shows that S. pyogenes also expressed PC, which was different for each strain, as assessed by the fluorescence intensity of flow cytometry. Furthermore, even with the same emm genotype, PC expression differed.
The M protein, a cell surface protein that is the major virulence of S. pyogenes, is encoded by the emm gene [19]. Classic M protein serological typing was largely replaced by sequence typing of the 5 end of the emm gene in the late 1990s [20]. Kuhn et al. reported in a cohort study of 248 children in Canada that the most common emm genotype in patients with recurrent pharyngitis was emm12 (24.2%), followed by emm3 (18.2%), emm1 (15.2%), and emm4 (12.1%) [21]. In addition, emm4, emm6, and emm75 strains invaded Detroit 562 cells significantly more than other genotypes [22]. Because of the small number of samples, it is impossible to compare. In this study, emm75 strains were 50% (4/8), while emm89, emm28, emm12, and emm11 strains were 12.5% each (1/8), and emm75 had a higher adhesiveness than the other emm genotypes.
High PC expression is thought to be highly pathogenic of Streptococcus pneumoniae and Haemophilus influenzae [23]. For instance, Streptococcus pneumoniae has been reported to cause more invasive infections in bacteria with a higher PC expression [24]. Andersson et al. reported that Streptococcus pneumoniae isolated from patients with acute otitis media adhered easily to cells, suggesting that the adhesive force is the virulence factor of Streptococcus pneumoniae [25]. Furthermore, strains isolated from the blood of patients with severe S. pyogenes infection have been reported to adhere more easily to cells [26]. In this study, PC expression and cell adhesion showed a positive correlation. These findings suggest that in S. pyogenes, higher PC expression makes it easier to adhere to cells and is more virulent.
This study showed that there was a significant negative correlation between TEPC-15 inhibitory effects and PC expression in S. pyogenes. Antibody responses to PC have been demonstrated in mouse experiments, and cells involved in the regulation of anti-PC antibody production have also been identified [27,28]. Kurono et al. reported that cell adhesion of Streptococcus pneumoniae and non-typeable Haemophilus influenzae was significantly reduced by treatment with a secretory IgA antibody [29]. In addition, the administration of PC to the nasal cavity of mice induced PC-specific IgA in mucosal secretions reacted with Streptococcus pneumoniae and non-typeable Haemophilus influenzae and increased clearance from the nasal cavity [30]. These findings supported the inhibitory effect of TEPC-15 on the adherence of these bacteria.
Furthermore, intracellular invasion assay experiments showed a significant negative correlation between TEPC-15 inhibitory effects and PC expression. In addition, PC expression and cell invasion showed a positive correlation. S. pyogenes invade host cells by internalization via phagosomes or endosomes [31,32]. Additionally, S. pyogenes form mature and antibiotic-resistant biofilms on physiologically relevant epithelial cells [33]. Ogawa et al. reported that more than half of the emm75 strains invaded cells and escaped the action of antibacterial agents even when penicillin was administered at 10 times the minimum inhibitory concentration [22]. In this study, emm75 strains had high PC expression and a large number of intracellular invasions. These findings suggested that PC expression is an important factor for bacteria that invades the cells.
PAF-R was inhibited on the cell surface to confirm the relationship between PC and PAF-R in adhesion to cells in S. pyogenes. Iuchi et al. reported that PAF-R expression in Detroit 562 cells was confirmed by FACS [23]. Cundell et al. reported that in Streptococcus pneumoniae, only highly pathogenic bacteria bound to PAF-R when adhering to cells [13]. Moreover, the expression of PAF-R is enhanced by viral antigens and contributes to the development of recurrent and persistent upper respiratory tract infections [18]. In this study, the inhibition of PAF-R with ABT-491 significantly inhibited cell adhesion and infiltration of bacteria with high PC expression rather than those with low PC expression. These results indicate that cellular PAF-R and bacterial PC are significantly involved in adhesion in S. pyogenes with high PC expression.
Even in bacteria with high PC expression, inhibition with TEPC-15 or ABT-491 did not suppress all adhesion to cells. This observation suggests that S. pyogenes involves PC and other adhesion factors. In fact, on the surface of S. pyogenes, there are various proteins, such as pili, SfbX and Lsp, which promote cell adhesion and colonization [10]. Furthermore, the administration of 100 µg/mL of TEPC-15 did not suppress cell adhesion. Weiser et al. reported three different levels of reactivity differences with TEPC-15 because of phase variation in the PC structure [34]. In addition, coating the PC epitope with TEPC-15 may lead to persistent infection [35]. Administering 100 µg/mL of TEPC-15 may influence avoid suppression of the adhesion by these factors.
Our study had some limitations. First, the number of bacteria used was small. However, by investigating the PC and emm genotypes together, we could examine from two aspects. Second, this study used cell lines derived from pharyngeal cancer. As these cells are of human origin, the results of this study were likely to be similar to those obtained using pharyngeal cells.

Conclusions
This study demonstrates that PC was expressed in S. pyogenes, and even if they had the same emm genotype, their remarks were different. It was found that PC and PAF-R are involved in cell adhesion and intracellular invasion of S. pyogenes.