Candidate proteomic biomarkers for three genogroups of the swine pathogen Streptococcus suis serotype 2

Abstract

Background: Streptococcus suis, more specifically serotype 2, is a major swine pathogen and an emerging zoonotic agent that causes severe infections such as meningitis, endocarditis, and septicemia. In this study, surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI) was used to investigate the protein expression profiles of 45 strains of S. suis serotype 2 that had previously been clustered by multilocus sequence typing (MLST) into three sequence types (ST1, ST25, and ST28) (n = 15 for each ST).

Results: The SELDI data were analyzed using the univariate Mann-Whitney and Kruskal-Wallis tests and multivariate statistical methods (heatmap/hierarchical clustering). The heatmap identified 136 cell proteins, and hierarchical clustering provided a 100% correct classification of all fifteen ST1 and ST25 strains and thirteen of the fifteen ST28 strains (87% correct). The univariate statistical analyses of the SELDI protein expression profiles identified nine significant proteins that discriminated the strains of the three STs of S. suis. Of these proteins, two were overexpressed in ST1 (5958 Da and 10249 Da), four in ST25 (5989 Da, 6646 Da, 7421 Da, and 9825 Da), and three in ST28 (4516 Da, 7833 Da, and 9342 Da). Two of the proteins associated with the ST28 strains (p4516 and p9342) were purified and were identified as a putative ABC transporter and a nucleoid-DNA-binding protein, respectively.

Conclusions: SELDI analysis of 45 strains of S. suis allowed to identify nine statistically significant proteins that can be specifically correlated with either ST1, ST25 or ST28. The possible involvement of the overexpressed proteins in the pathology of S. suis infections will require further investigation.

Figure 1

Heatmap/hierarchical clustering of 136 proteins of 45  S. suis strains discriminated by MLST into three sequence type (ST) groups: ST1 (n = 15), ST25 (n = 15), and ST28 (n = 15). The clusters were obtained by combining the average intensity values of all samples tested in duplicate on CM10 and Q10 ProteinChip arrays (acquisition protocol 1). The consecutive numbers of each strain used in the SELDI expression difference mapping (EDM) are indicated above the image (S1-S15 for the ST1 group (in red); S16-S30 for the ST25 group (in blue), and S31-S45 for the ST28 group (in green) followed by the respective sequence types and original names of the S. suis strains (in brackets). The protein masses detected on the CM10 (red) and Q10 (blue) ProteinChip arrays are indicated on the right side of the image. Specific EDM conditions: First pass: peak S/N ≥ 5, valley depth S/N ≥ 3, minimal peak threshold – 20% for all spectra; second pass: peak S/N ≥ 2, valley depth S/N ≥ 2; third pass: adding estimated (missing) peaks to complete the clusters, clustered mass window width – 0.1%, autocentroid marks on peaks, M/Z range of analysis (z = 1): 3000–20000 Da.

Figure 2

Candidate biomarkers of S. suis selected by univariate analysis of two or three groups of strains divided by sequence type.