The bias of experimental design, including strain background, in the determination of critical Streptococcus suis serotype 2 virulence factors

Abstract

Streptococcus suis serotype 2 is an important porcine bacterial pathogen and emerging zoonotic agent mainly responsible for sudden death, septic shock, and meningitis. However, serotype 2 strains are genotypically and phenotypically heterogeneous. Though a multitude of virulence factors have been described for S. suis serotype 2, the lack of a clear definition regarding which ones are truly "critical" has created inconsistencies that have only recently been highlighted. Herein, the involvement of two factors previously described as being critical for S. suis serotype 2 virulence, whether the dipeptidyl peptidase IV and autolysin, were evaluated with regards to different ascribed functions using prototype strains belonging to important sequence types. Results demonstrate a lack of reproducibility with previously published data. In fact, the role of the dipeptidyl peptidase IV and autolysin as critical virulence factors could not be confirmed. Though certain in vitro functions may be ascribed to these factors, their roles are not unique for S. suis, probably due to compensation by other factors. As such, variations and discrepancies in experimental design, including in vitro assays, cell lines, and animal models, are an important source of differences between results. Moreover, the use of different sequence types in this study demonstrates that the role attributed to a virulence factor may vary according to the S. suis serotype 2 strain background. Consequently, it is necessary to establish standard experimental designs according to the experiment and purpose in order to facilitate comparison between laboratories. Alongside, studies should include strains of diverse origins in order to prevent erroneous and biased conclusions that could affect future studies.

Fig 1. The S. suis serotype 2 dipeptidyl peptidase IV and autolysin are not involved in adhesion to fibronectin, regardless of the sequence type (ST) of the strain used.

Adhesion of different wild-type strains and dipeptidyl peptidase IV (DPPIV)- or autolysin (Atl)-deficient mutants to human plasma fibronectin (10 μg/mL), as determined by ELISA after 2 h of incubation. The optical density (OD) was measured at 450 nm and values corrected using the appropriate controls. Results are expressed as mean ± SEM obtained from three independent experiments.

Fig 2. The S. suis serotype 2 dipeptidyl peptidase IV is not involved in adhesion to porcine tracheal epithelial cells, regardless of the sequence type (ST) of the strain used, unlike the autolysin.

Adhesion of different wild-type strains and dipeptidyl peptidase IV (DPPIV)- or autolysin (Atl)-deficient mutants to porcine epithelial cells was evaluated after 2 h of incubation with bacteria (MOI = 10). Results are expressed as mean ± SEM obtained from three independent experiments and represent the percentage of adhered inoculum. # indicates a significant difference (p < 0.01) between the wild-type ST1 strain P1/7 and ST25 strain 89–1591; *** (p < 0.001) between the wild-type strain and its Atl-deficient mutant.

Fig 3. The dipeptidyl peptidase IV is not involved in S. suis serotype 2 biofilm formation, while implication of the autolysin is dependent on the sequence type (ST) of the strain used.

Biofilm formation of different wild-type strains and dipeptidyl peptidase IV (DPPIV)- or autolysin (Atl)-deficient mutants in the presence of 2 mg/mL of porcine fibrinogen was evaluated after 24 h of incubation. The optical density (OD) was measured at 575 nm and values corrected using the appropriate controls. Results are expressed as mean ± SEM obtained from three independent experiments. # indicates a significant difference (p < 0.05) between the wild-type ST1 strain P1/7 and ST7 strain SC84 or ST25 strain 89–1591; a (p < 0.01) between the wild-type ST7 strain SC84 and ST25 strain 89–1591; *** (p < 0.001) between the wild-type strain and its Atl-deficient mutant.

Fig 4. The S. suis serotype 2 dipeptidyl peptidase IV and autolysin are not implicated in host virulence in a C57BL/6 mouse model of infection, regardless of the sequence type (ST) of the strain, with the exception of a minor role for the autolysin of the ST25 strain.

Survival of C57BL/6 mice infected with 5 x 10⁷ CFU of different wild-type and dipeptidyl peptidase IV (DPPIV)- or autolysin (Atl)-deficient mutants by intraperitoneal inoculation. (A) P1/7 (ST1) and its mutants, (B) SC84 (ST7) and its mutants, and (C) 89–1591 (ST25) and its mutants. * indicates a significant difference (p < 0.05) between the wild-type ST25 strain 89–1591 and its Atl-deficient mutant.

Fig 5. The autolysin of the ST25 strain, but not the dipeptidyl peptidase IV, hinders bacterial survival in the blood.

Blood bacterial burden of surviving C57BL/6 mice 24 h following intraperitoneal inoculation of 5 x 10⁷ CFU of the ST25 strain 89–1591 and its dipeptidyl peptidase IV (DPPIV)- or autolysin (Atl)-deficient mutants. Results are expressed as geometric mean. ** indicates a significant difference (p < 0.01) between the wild-type ST25 strain 89–1591 and its Atl-deficient mutant.