The key virulence-associated genes of Streptococcus suis type 2 are upregulated and differentially expressed in vivo


  • Editor: Tim Mitchell

Correspondence: Weicheng Bei, State Key Laboratory of Agricultural Microbiology; Laboratory of Animal Infectious Diseases, College of Animal Science & Veterinary Medicine, Huazhong Agricultural University, Wuhan, Hubei, China, 430070. Tel.:+86 27 8728 8629; fax:+86 27 8728 1795; e-mail:


Streptococcus suis causes disease in both pigs and humans. Many virulence genes, including mrp, ef, sly, gapdh, fbps, and hyl, are shown to contribute to S. suis pathogenesis. In this study, the differential expression of the key S. suis serotype 2 virulence-associated genes was monitored 24 and 48 h after infection of HeLa cells. Gene expression was further investigated in bacteria harvested from the blood, lungs, brains, and synovia of swine 24, 48, and 72 h postintravenous infection with S. suis serotype 2. Bacterial RNA was extracted and gene expression was monitored using real-time reverse transcriptase-PCR at various time points. The results showed that after 24 h postinfection in vivo, mrp, gapdh, fbps, and hyl were highly expressed in all organs. At 48 h postinfection, ef and sly were highly expressed in the lungs and brain. At 72 h postinfection, all genes had a high level of expression in all organs. This study provides direct evidence that S. suis serotype 2 key virulence-associated genes are differentially expressed in a time-specific or an organ-specific manner in vivo. These findings also show the dynamic distribution of gene expression. Characterization of the differential expression patterns of this subset of key virulence-associated genes facilitates further study into the importance of these genes in S. suis serotype 2 pathogenesis in pigs.


Streptococcus suis is the causative agent of several diseases in pigs, including meningitis, septicemia, arthritis, pneumonia, and even acute death, and is responsible for high economic losses in the pig industry each year. Thirty-five serotypes of S. suis have been identified so far based on discrepancies in capsular antigens (King et al., 2001). Streptococcus suis serotype 2 (S. suis 2 or SS2) is the most frequently isolated serotype worldwide, and comprises pathogenic, weakly pathogenic, and nonpathogenic strains (Gottschalk et al., 1999). In addition to causing disease in pigs, S. suis 2 can also cause severe enzootic infections in humans that result in septicemia, meningitis, and endocarditis (Arends & Zanen, 1998).

Approximately 200 cases of S. suis 2 infection have been reported in humans since the first case of meningitis was recorded in Denmark in 1968 (China CDC, Between July 2005 and August 2005, an outbreak of S. suis 2 infection occurred in Sichuan, China, involving 200 cases, of whom 38 died (Yu et al., 2006; Zhu et al., 2006).

In recent years, research on S. suis 2 has concentrated on its potential virulence factors and pathogenic mechanisms. Many proteins associated with virulence have been characterized, including muramidase-released protein (MRP) (Smith et al., 1992), extracellular factor (EF) (Vecht et al., 1991), suilysin (Jacobs et al., 1994), glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Brassard et al., 2004), fibronectin-binding proteins (FBPS) (de Greeff et al., 2002), and hyaluronate lyase (HYL) (King et al., 2004). MRP is often present in the virulent strain of this pathogen and is absent in the avirulent strain (Smith et al., 1992; Wisselink et al., 2001). EF is also considered an important virulence factor, and the EF deletion strain has lower virulence than the wild type (Vecht et al., 1991). Suilysin belongs to a family of thiol-activated toxins that lose activity during oxidation but regain full activity upon deoxidation (Boulnois et al., 1991; Jacobs et al., 1995). Many S. suis strains show hemolysis activity on blood agar plates, and hemolysin may also be involved in S. suis pathogenesis.

Recent studies demonstrate that bacterial adherence to host cells is important for pathogenicity. Some virulence factors associated with bacterial adherence may play a role in the early stages of infection. Streptococcus suis 2 has the ability to adhere to many different epithelial and endothelial cell lines (Lalonde et al., 2000; Benga et al., 2005). Specific S. suis 2 adhesins have been characterized, including gapdh and fbps. Many pathogenic streptococci contain gapdh, a gene shown to play an important role in S. suis 2 adhesion to host cells (Brassard et al., 2004). FBPS is an adherence protein that binds to fibronectin in host cells (de Greeff et al., 2002). Fibronectin is a large glycoprotein composed of two polypeptide chains with a combined molecular weight of 450 kDa. It is found in a soluble form in the bodily fluids and in an immobilized form in both the extracellular matrix and on the surface of host cells. Some bacteria bind fibronectin in order to colonize or invade host cells.

The hyaluronate lyase of S. suis 2 is a secreted protein that degrades hyaluronic acid (HA) into an unsaturated disaccharide. HA is a high-molecular-weight polysaccharide consisting of repeating disaccharide units and is a fundamental component of the extracellular matrix and capsular material (Laurent & Fraser, 1992). Many gram-positive bacteria capable of handling HA are able to cause infection at the mucosal and skin surfaces of humans and animals. It is proposed that a decrease in viscosity due to HA depolymerization increases the permeability of connective tissues, which increases the ability of bacteria to spread and contributes to subsequent virulence. Human tissues known to contain HA include the blood, brain, liver, umbilical tissue, and skin. Members of several streptococcal species are known to produce a cell surface-associated hyaluronate lyase (Marciel et al., 1997; Hynes & Walton, 2000), and research indicates that this lyase may be important in pathogenesis.

While many studies have been conducted to define the S. suis 2 mechanism of pathogenesis, the detailed role of key virulence-associated genes in S. suis 2 pathogens remains unknown. Mutagenesis studies illustrate that mrp, ef, fbps, sly, gapdh, and hyl are specifically induced in host tissues and contribute to virulence. However, it is not currently known whether these genes are coordinately regulated, or whether each gene is specifically up- or downregulated at different time points and in distinct organs in vivo. There is little direct molecular evidence confirming in vivo expression. In this study, in vivo expression of important S. suis 2 virulence factors is shown. Because HeLa cells are known to interact with the S. suis 2, differential expression of these virulence-associated genes was first monitored after infection in this cell. mRNA transcript levels of each gene were then measured using relative quantitative reverse transcriptase (RT)-PCR analysis of total bacterial RNA isolated from the blood, lungs, brains, and synovia of swine at various times postinfection. This is the first comparison of the relative kinetics of in vivo expressed S. suis 2 virulence-associated factors.

Materials and methods

Bacterial strains and growth conditions

Streptococcus suis 2 strain SC19 was isolated from diseased pigs in the Sichuan region of China. Bacteria were grown on blood agar plates [tryptic soy agar (TSA) containing 10% bovine blood TSA, Difco, France] for 18 h at 37 °C, and inoculated into serum broth [tryptic soy broth (TSB) containing 10% bovine blood TSB, Difco, France] for 3 h at 37 °C. For infections of swine, the serum broth culture was adjusted to c. 1 × 108 CFU mL−1. For isolation of in vitro-derived total RNA, the serum broth culture was centrifuged at 15 500 g for 5 min to pellet the bacteria.

Cell culture and S. suis 2 strain SC19 infection

To investigate the differential expression of the virulence-associated genes of S. suis after infection, HeLa cells were grown in RPMI 1640 containing 20% fetal bovine serum. Cells were grown into monolayers on six-well plates and each well was infected with 2 mL containing 1 × 106 CFU S. suis bacteria. At 24 and 48 h postinfection, the bacteria were harvested for RNA isolation.

Harvesting S. suis 2 bacteria for RNA isolation and infection of swine

Three groups of nine 4-week-old swine were intravenously infected with c. 5 × 107 CFU of S. suis 2 strain SC19. After 24, 48 or 72 h, each group was sacrificed, blood was heparinized and placed on ice for further use, and lung samples were perfused with sterile phosphate-buffered saline (PBS: 136.89 mM NaCl, 2.68 mM KCl, 8.1 mM Na2HPO4, 2.79 M KH2PO4, pH 7.2) to remove blood-borne bacteria. The brains were washed with sterile PBS (pH 7.2), and the lungs and brains were homogenized. One hundred milligrams of each sample homogenate and 1.5 mL of the blood and synovia samples were centrifuged at 825 g for 5 min at 4 °C, as described previously (Ogunniyi et al., 2002; LeMessurier et al., 2006). The sample supernatants were centrifuged at 15 500 g for 5 min at 4 °C to pellet the bacterial cells. Supernatants were then discarded and RNA was extracted from the bacterial pellet.

RNA isolation from bacteria

Total bacterial RNA was extracted from in vitro- and in vivo-harvested tissues as described previously (Ogunniyi et al., 2002; LeMessurier et al., 2006). The RNA was treated with 0.5 U μL−1 RQ1 RNAse-free DNAse (Promega M610A) for 30 min at 37 °C. The RNA preparation was then treated with RQ1 DNAse stop buffer (Promega M198A) to inactivate the DNAse. The purified RNA preparation was confirmed by one-step RT-PCR with or without reverse transcriptase, using 16S rRNA gene-specific primers, and the products were electrophoresed on a 1% TAE-agarose gel.

RNA quantitation

Sly, mrp, ef, gapdh, fbps, and hyl mRNA in the cell infection was determined using semiquantitative RT-PCR. mRNA levels from in vivo infections were measured using one-step relative quantitative RT-PCR. Primers specific for the 16S rRNA gene were used as an internal control. The specific primers used for the various RT-PCR assays are listed in Table 1. The SYBR Green method was used. All reactions were performed in triplicate. Each reaction tube contained 25 μL 2 × SYBR Green real-time PCR Master Mix (TOYOBO QPK-201), 0.5 U μL−1 RNAse Inhibitor, 0.3 U μL−1 ReverTra Ace, 0.4 μM of gene-specific F and R primers, and 5 μL template, made up to a final volume of 50 μL with distilled water. The RT-PCR cycling conditions were as follows: 42 °C for 30 min and 95 °C for 5 min, followed by 40 cycles of 95 °C for 15 s, 60 °C for 30 s, and 72 °C for 45 s. For relative quantitative RT-PCR, an ABI 7500 real-time PCR system was used.

Table 1.   Oligonucleotide primers
PrimerSequence (5→3)


Infection of pigs

The body temperature of each swine was measured preinfection and at each day postinfection with S. suis 2 strain SC19. The body temperatures were c. 37.5 °C before infection and remained between 41 and 42 °C after infection. The clinical signs of S. suis 2 infection at 20 h postinfection included depression, anorexia, lassitude, recumbency, tremors, opisthotonus, ataxia, dyspnea, arthritis, lameness, erythema, and stiffness. The results indicated that S. suis 2 strain SC19-infected swine had typical symptoms of infection.

Differential expression of S. suis virulence genes at different time points after infection of HeLa cells

Expression of the six key virulence-associated genes, mrp, ef, sly, gapdh, fbps, and hyl in S. suis 2 SC19 was measured by conducting RT-PCR on S. suis 2 SC19-infected HeLa cells (Fig. 1). All six genes were upregulated at different time points, with expression levels increasing as time increased. The mRNA levels of fbps, gapdh, hyl, and mrp were high by 24 h, while the mRNA levels of ef and sly did not reach a high level until 48 h.

Figure 1.

 Semiquantitative analysis of ef, fbps, gapdh, hyl, mrp, and sly mRNA expression in Streptococcus suis-infected HeLa cells. RNA from infected cells was analyzed at 24 and 48 h postinfection.

Differential expression of S. suis 2 virulence genes in vivo

The mRNAs from various virulence factors were extracted from S. suis 2 grown in vitro, and isolated from the blood, lungs, brains, and synovia of swine at 24, 48, or 72 h postinfection. Gene expression of sly, mrp, ef, gapdh, fbps, and hyl were compared using relative quantitative RT-PCR in the same organs at different time points, or in different organs at the same time points. In all organs, GAPDH, FBPS, HYL, and MRP were higher than their in vitro levels, while suilysin and EF levels were similar to that observed in vitro at 24 h (Fig. 2a). At 48 h postinfection, all genes examined were higher than their in vitro levels in the lungs and brains (Fig. 2b). Gene expression levels were higher than their in vitro levels in all tissues at 72 h postinfection (Fig. 2c).

Figure 2.

 mRNA levels of the Streptococcus suis type 2 virulence genes, mrp, ef, sly, gapdh, fbps, in various organs at 24 h (a), 48 h (b), and 72 h (c) postinfection, relative to in vitro levels. Real-time RT-PCR data for each gene relative to that obtained for the 16S rRNA gene control. Data points represent the means±SE of RT-PCR reactions performed on RNA extracts from bacteria harvested from different organs.

mRNA levels of the six virulence-associated genes were compared at different time points and in different organs both in vivo and in vitro. The results demonstrated that gapdh, hyl, and mrp reached a high level in the blood at 24 h postinfection, and were the highest at 72 h postinfection. Expression of ef, sly, and fbps was similar to that observed in vitro at 24 and 48 h, but was remarkably higher at 72 h (Fig. 3a). All genes were more highly expressed in the lung than they were in vitro at 24 h postinfection, with the exception of sly, which was the highest at 48 h postinfection, and declined by 72 h postinfection. Sly gene levels began to increase at 48 h postinfection (Fig. 3b). Gene expression in the brain was similar to that observed in the lung; however, ef was higher at 48 h postinfection, and no significant differences in sly expression were observed (Fig. 3c). In this study, all infected pigs had symptoms of arthrocele, and mRNA levels of these genes were measured in the synovia. Fbps, gapdh, and mrp reached high levels at 24 h postinfection, declined by 48 h, and reached their highest levels at 72 h postinfection. Ef and sly gene levels were high at 72 h postinfection (Fig. 3d). In this study, all the gene expression changes were determined using Student's t-test.

Figure 3.

Streptococcus suis type 2 virulence gene mRNA concentrations in the blood (a), lung (b), brain (c), and synovia (d) at different time points postinfection, relative to in vitro concentrations. Real-time RT-PCR data for each gene relative to those obtained for the 16S rRNA gene control. Data points represent the means±SE of RT-PCR reactions performed on RNA extracts from bacteria harvested from in vivo and in vitro cultures.


During the last few years, S. suis 2 infections have become a major problem in all countries with an intensive pig industry. Streptococcus suis is frequently isolated from pigs with meningitis, arthritis, septicemia, and bronchopneumonia. The organism can also be isolated from the tonsils of healthy carrier pigs. Occasionally, S. suis causes meningitis with various complications in humans (Arends & Zanen, 1998; Heidt et al., 2005). In recent years, research has concentrated on the potential virulence genes and the pathogenic mechanism of this organism. Many genes are shown to be correlated with virulence, including MRP, EF, suilysin, GAPDH, FBPS, and HYL. However, no studies have extensively analyzed differences in the in vivo expression of these genes, and their contribution to the survival and growth of S. suis 2 at various time points and in different organs remains poorly understood.

In this study, relative quantitative RT-PCR was used on total bacterial RNA isolated from the blood, lungs, brains and synovia of swine at various times following an intravenous challenge. The mRNA levels of six virulence-associated genes (mrp, ef, sly, gapdh, fbps, and hyl) from S. suis 2 were measured and compared. The results demonstrated that gene upregulation differed from organ to organ.

To investigate how bacteria interact with the host, it is important to understand how the bacterial outer surface proteins contribute to evasion of host defense, adhesion, and invasion. The GAPDH of S. suis 2 is a surface expression protein with the ability to bind albumin. This protein appears to play an important role in bacterial adhesion because expression-defective mutants or wild-type bacteria preincubated with recombinant GAPDH protein show a reduction in adhesion to porcine tracheal rings (Brassard et al., 2004). The highly conserved GAPDH protein is implicated in many surface-specific activities, serving as a receptor for the plasmin receptor, and contributing to host cytoskeletal protein binding and signal transduction between bacteria and host cells (Gase et al., 1996; Pancholi & Fischetti, 1997; Marie-Claude Jobin et al., 2004). GAPDH is a key glycolytic enzyme, which has been found to be associated with membranes and subcellular cytoskeletal structures. Thus, high initial levels of GAPDH expression may be critical for bacterial growth, proliferation, adhesion, and pathogenesis. In this study, gapdh was significantly upregulated in all examined organs at 24 h postinfection. At 48 and 72 h postinfection, it was still expressed at high levels in the lungs and brains, although these levels were lower than those observed at 24 h (Fig. 2). This finding is also consistent with a function for the GAPDH protein in S. suis 2 pathogenesis.

Bacterial attachment to the host cell surface is considered a prerequisite for subsequent colonization or invasion of the host. Attachment is mediated through specific interactions between adhesins and host cell receptors. For group A streptococci, fibronectin is one receptor that has been well characterized. Many bacteria have a number of fibronectin-binding proteins (fbps) that are important virulence factors (Joh et al., 1999). Streptococcus suis 2 also has an fbps protein that is shown to be immunogenic in pigs, and may function in bacterial adherence to host cells (de Greeff et al., 2002). In this study, fbps expression was upregulated in all organs at 24 h postinfection (Fig. 2a). This demonstrated that fbps acts as an adhesin and that high initial expression levels were critical for virulence.

MRP and EF were identified as two proteins associated with the virulence of S. suis 2 (Vecht et al., 1991). The mrp and ef genes were predominantly expressed in strains isolated from diseased pigs. Based on the presence or the absence of MRP and EF, S. suis 2 strains were grouped into three phenotypic categories including MRP+EF+, MRP+EF−, and MRP−EF−. MRP+EF− strains were less pathogenic than MRP+EF+ strains, suggesting that EF plays a more important role in the pathogenesis of S. suis 2 infections (Vecht et al., 1991). MRP is a peptidoglycan-associated protein that is similar to the M protein, an important virulence factor in group A streptococci. Some findings indicate that mrp is an S. suis 2 adhesin, because the anti-MRP serum can degrade S. suis 2 adherence. In this study, mrp gene expression levels were dramatically upregulated both in vitro and in vivo. It is believed that mrp upregulation in the initial stage of infection may play a role in adherence. The ef gene functions differently, however. In most organs, it was not notably upregulated until 48 or 72 h postinfection (Fig. 2). EF was upregulated later than most adhesins, suggesting that it may play a different role during infection.

Suilysin, a protein belonging to the thiol-activated toxin family, is an important S. suis 2 virulence factor. A number of bacteria contain this toxin, and its function is not well known. In some reports, sly enhances epithelial invasion and cell lysis, and while the sly+ strain is cytotoxic, the sly− strain only has adherence properties (Shichun et al., 2003). Some studies suggest that the sly+ S. suis strain is an adherence strain and does not play a role in direct cellular invasion as part of a complicated multistep process that leads to disease in pigs. It remains unknown whether sly is an early- or advanced- stage functional protein. In the present study, sly mRNA levels were upregulated at 48 and 72 h postinfection, suggesting that sly has an effect on the advanced stages of S. suis 2 infection.

The hyl gene expression pattern was an interesting finding of this study. This gene was the least abundant mRNA species in vitro, but was significantly upregulated after 24 h postinfection in vivo. The hyaluronate lyase of S. suis is a secreted protein that degrades HA into unsaturated disaccharides; however, the role of this protein in pathogenesis is not well known. A number of pathogenic gram-positive bacteria that produce hyl are able to initiate infections at skin and mucosal surfaces (Marciel et al., 1997; Hynes & Walton, 2000; King et al., 2004). Some research suggests that hyl may not be an important virulence factor because only some isolated S. suis bacteria contain this gene. In this study, hyl mRNA levels were very low in vitro, but reached a significant level in vivo. It is possible that some host factors are required for hyl gene expression.

In conclusion, it is believed that this is the first study in which S. suis 2 virulence gene mRNA was directly quantified from bacteria in the blood, lung, brain, and synovia of infected pigs at different times postinfection. The difference in expression levels of these virulence genes in vivo and in vitro may help to explain their putative roles in S. suis 2 pathogenesis.


This research is was supported by a grant from The National Basic Research Program (also called 973 Program) (No.2006CB504404).