Distribution, genetic diversity, and variable expression of the gene encoding hyaluronate lyase within the Streptococcus suis population

Abstract

Although Streptococcus suis is an economically important pathogen of pigs and an occasional cause of zoonotic infections of humans knowledge of crucial virulence factors, and as a consequence targets for therapeutic or prophylactic intervention, remains limited. Here we describe a detailed study of the distribution, diversity, and in vitro expression of hyaluronate lyase, a protein implicated as a virulence factor of many mucosal pathogens. The gene encoding hyaluronate lyase, hyl, was present in all 309 bona fide S. suis isolates examined representing diverse serotypes, geographic sources, and clinical backgrounds. Examination of the genetic diversity of hyl by RFLP and sequence analysis indicated a pattern of diversity shared by many gram-positive surface proteins with a variable 5' region encoding the most distal cell surface-exposed regions of the protein and a much more conserved 3' region encoding domains more closely associated with the bacterial cell. Variation occurs by several mechanisms, including the accumulation of point mutations and deletion and insertion events, and there is clear evidence that genetic recombination has contributed to molecular variation in this gene. Despite the ubiquitous presence of hyl, the corresponding enzyme activity was detected in fewer than 30% of the 309 isolates. In several cases this lack of activity correlates with the presence of mutations (either sequence duplications or point mutations) within hyl that result in a truncated polypeptide. There is a striking absence of hyaluronate lyase activity in a large majority of isolates from classic S. suis invasive disease, indicating that this protein is probably not a crucial virulence factor, although activity is present in significantly higher numbers of isolates associated with pneumonia.

FIG. 1.

Schematic illustration of hyl diversity based on the virtually complete sequence from nine distinct isolates. The areas shaded at the extreme 5′ and 3′ parts of the gene represent the location of PCR primer binding sites and were therefore not sequenced in the present study. The dashed arrow represents the location of the putative signal peptide cleavage site.

FIG. 2.

Examples of duplication or deletion events generating diversity in the 5′ regions of hyl shown relative to the 1307 (serotype 7) reference sequence AJ308328. Repeated sequences that may be associated with these events are underlined and in boldface.

FIG. 3.

The upper panel shows a comparison of the 5′ sequences of a representative of each of the 21 RFLP types defined in the present study. Only sites variable (not absent) relative to the P1307 sequence are shown. Three arbitrary sequence groups have been identified based on the sequences of hyl1307 (blue), hyl20 (pink), and hyl21 (red). Below is a schematic illustration of the mosaic nature of a number of the hyl sequences, reflecting a history of recombination in the evolution of this locus. The lack of diversity in some regions makes unequivocal differentiation of sequences corresponding to blue and pink groups impossible, and where this is the case, the sequence is shown in black and illustrated as dappled on the schematic.