Amoeba Host Model for Evaluation of Streptococcus suis Virulence

Abstract

The Gram-positive bacterium Streptococcus suis is a major swine pathogen worldwide that causes meningitis, septicemia, and endocarditis. In this study, we demonstrate that the amoeba Dictyostelium discoideum can be a relevant alternative system to study the virulence of S. suis.

TEXT

Streptococcus suis is a major swine pathogen mainly associated with meningitis, pneumonia, and endocarditis (14). This Gram-positive bacterium can also affect humans in close contact with sick or carrier pigs or with their derived products (13, 17). Thirty-five serotypes of S. suis have been described, but serotype 2 is most commonly associated with diseases in pigs and humans (20, 21). Many potential virulence factors have been described in S. suis (2). Among them, the capsule allows the bacterium to escape killing by both macrophages and neutrophils (7, 8, 11, 19), while the subtilisin-like protease (SspA) degrades host proteins and induces secretion of inflammatory mediators (3, 5, 6, 15).

Dictyostelium discoideum is a soil-living amoeba that uses especially bacteria as a food source (1, 9). In the last decade, D. discoideum has been used as an alternative host model to study various aspects of host-pathogen interactions and to characterize bacterial virulence mechanisms, mainly of Gram-negative pathogenic bacteria (9). More specifically, the virulence plaque assay using this amoeba offers a simple readout of bacterial virulence based on the capacity of D. discoideum to multiply and form plaques on lawns of nonpathogenic bacteria but not on pathogenic bacteria when appropriate conditions are used (12). The use of laboratory animals to study the virulence of S. suis implies many logistical and ethical limitations that are mainly absent when using a nonmammalian model such as D. discoideum. In the present study, we investigated whether the amoeba D. discoideum can be used as a relevant host model to study the virulence of the Gram-positive bacterium S. suis. For this purpose, we first adapted the model of D. discoideum for S. suis and then compared the behaviors of the amoeba using wild-type strains and mutants defective for virulence factors known to be essential for the pathogenicity of S. suis in an animal model.

The reference strain S. suis P1/7 (serotype 2) and its two SspA (subtilism-like protease)-deficient mutants (M3G and G6G) were used in this study. Mutants M3G and G6G were obtained from a mutant library constructed by insertion of a Tn917 transposon into S. suis P1/7 (3, 18). The reference strain S. suis S735 (serotype 2) and its two capsule-deficient mutants (BD101 and 2A) were also used in this study. Mutant BD101 was constructed by deletion of the aro promoter (11), while mutant 2A was constructed by insertion of the Tn916 transposon into S. suis S735 (8). The decreased virulence of these four mutants was previously demonstrated in animal models (4, 8, 11). A nontypeable strain of S. suis (1078212), which has been shown to be naturally devoid of capsule (4), was also included in this study. All bacteria were grown at 37°C in Todd-Hewitt broth (THB; BBL Microbiology Systems, Cockeysville, MA).

In a first attempt, the virulence plaque assay using D. discoideum was performed with the two virulent wild-type strains of S. suis (S735 and P1/7), the capsule-deficient mutant BD101, and the SspA-deficient mutant M3G. D. discoideum DH1-10 was grown in a petri dish at 21°C in HL5 liquid medium (14.3 g/liter peptone [Oxoid, Nepean, Ontario, Canada], 7.15 g/liter yeast extract [EMD, Gibbstown, NJ], 18 g/liter maltose, 3.6 mM Na₂HPO₄, 3.6 mM KH₂PO₄) containing 15 μg/ml tetracycline (10). Amoebae were harvested by centrifugation (5 min at 1,500 × g), washed, and resuspended in tetracycline-free HL5 medium at a concentration of 100 cells per 5 μl. Thereafter, bacteria were harvested by scraping the surface of two petri plates with confluent lawns of growth (overnight at 37°C) and suspended in 2 ml of HL5 medium without antibiotic. The bacterial suspensions (100 μl; optical density at 660 nm [OD₆₆₀] of ≈5) were then applied to wells of 24-well plates containing HL5 agar medium (2 ml/well). Once dried, bacterial lawns were spotted with 5 μl containing 100 cells of D. discoideum. Plates were incubated at 23°C for 2 days and then examined visually for plaque formation. As shown in Fig. 1, D. discoideum was able to grow and to form plaques only on lawns of S. suis mutants. This result supported the idea that wild-type strains of S. suis have virulent behavior against D. discoideum, as was also observed in animal models.

Fig. 1.

Assessment of S. suis virulence using the D. discoideum model. Lawns of S. suis S735 and its capsule-deficient mutant, BD101, as well as S. suis P1/7 and its SspA-deficient mutant, M3G, were prepared. Drops (5 μl) of D. discoideum containing 100 cells were then applied. The ability of D. discoideum cells to create plaques (white areas) in the bacterial lawns (black) was recorded visually after 2 days of incubation at 23°C.

In a second step, the virulence of the wild-type strain S735 and its two capsule-deficient mutants (2A and BD101), the wild-type strain P1/7 and its two SspA-deficient mutants (M3G and G6G), as well as the nontypeable isolate (1078212) was characterized quantitatively by using serial dilutions of D. discoideum from 3,000 to 10 cells per 5 μl. As reported in Table 1, S. suis S735 did not allow the growth of amoebae even if 3,000 cells were deposited on the bacterial lawn. On the other hand, D. discoideum at a concentration in the range of 10 to 300 cells/well was sufficient to form clear plaques on lawns of mutants BD101 and 2A. The ability of D. discoideum to produce plaques on lawns of mutants that do not express capsular material suggests that these mutants are avirulent. These results are in agreement with previous studies that used an animal model to demonstrate that the capsule is an important virulence factor for S. suis (8, 11). On lawns of the two SspA-deficient mutants (M3G and G6G), D. discoideum was able to form clear plaques even when only 10 or 30 amoebae were applied onto lawns (Table 1). In contrast, the parental strain of these mutants, S. suis P1/7, was much more resistant since in three assays out of four, 3,000 amoebae were not enough to form plaques (Table 1). This supports our previous report indicating the critical role of the SspA protease in the virulence of S. suis in a mouse model (3).

Table 1.

Susceptibility of S. suis to D. discoideum predation

Straina	Capsule	SspA	Virulence assay result inb:
			Expt 1	Expt 2	Expt 3	Expt 4
S735	+	+	>3,000	>3,000	>3,000	>3,000
BD101	−	+	30	10	100	30
2A	−	+	30	30	300	300
P1/7	+	+	300	>3,000	>3,000	>3,000
M3G	+	−	30	10	30	30
G6G	+	−	30	10	30	30
1078212	−	+	10	10	1,000	30

^aStrains are grouped by the wild-type parent strains S735 and P1/7 and their mutant derivatives. Strain 1078212 is a nontypeable strain shown to be naturally devoid of capsule (4).

^bThe ability of D. discoideum to grow on a bacterial lawn was assessed by plating different amounts of amoebae. The numbers shown represent the minimal numbers of D. discoideum cells required to induce plaque formation on the bacterial lawns. Data from four independent experiments are presented.

The capsule of S. suis may be necessary to allow the resistance of bacteria to phagocytosis, while SspA may be important in another step during the disease process. This study suggests that the capsule and the SspA subtilisin-like protease are both required to ensure the virulence of S. suis in the D. discoideum model. However, this has to be confirmed using mutants derived from the same parental strain. Transmission electron microscopy coupled with ruthenium red staining confirmed that S. suis S735 expresses a dense capsule, while the mutants 2A and BD101 showed markedly less capsular material on their surface (Fig. 2). Interestingly, the nontypeable isolate 1078212, which displayed attenuated virulence in the D. discoideum model (Table 1), did not show any capsular material (Fig. 2). As shown in Fig. 2, the parent strain P1/7 and the two SspA mutants possessed a capsule, although some variation in thickness may be observed. In addition, Table 1 reports the SspA activity of all S. suis strains, measured using the chromogenic substrate succinyl-Ala-Ala-Pro-Phe-pNa (6). With the exception of mutants M3G and G6G, the other strains of S. suis possessed cell-associated SspA activity. Interestingly, Li et al. demonstrated the upregulation of the gene encoding the SspA subtilisin-like protease during the course of infection (16).

Fig. 2.

Visualization of S. suis capsule by transmission electron microscopy following ruthenium red staining. Magnification, 80,000×.

The quantitative virulence assay developed for S. suis in the present study can give an idea of the relative virulence of each strain tested (wild type or mutant). The loss of virulence of the analyzed mutants against D. discoideum is analogous to that previously observed with animal models, which supports the usefulness of this system as a novel tool for the analysis of virulence determinants of S. suis, especially by its technical and ethical advantages. Thus, the D. discoideum virulence assay by its simplicity may help to identify new virulence genes of S. suis. It should be mentioned that in this study, S. suis is grown at 37°C prior to be used in the assay with D. discoideum at room temperature. Because of these specific assay conditions, it is likely that not all of the virulence factors potentially expressed by S. suis may be identified. To the best of our knowledge, the present work is the first comprehensive study on the use of D. discoideum as an alternative model to assess the virulence of many reference strains and associated mutants of a pathogenic Gram-positive bacterial species using the virulence plaque assay.