||Streptococcus suis is an important pig pathogen which is mainly associated with meningitis, arthritis and septicaemia. Control of the disease is hampered by the lack of effective vaccines and the lack of reliable diagnostic tests with high specificity and sensitivity. The development of these tools is complicated by the number of existing serotypes, by the fact that we still lack knowledge of the factors responsible for virulence and protection, and by the fact that strains may vary in virulence. Therefore, research focused on the identification of virulence-associated markers that discriminate between virulent and less virulent or avirulent isolates, has gained considerable interest in recent years. The aim of the investigations described in this thesis, was to test whether these virulence-associated markers could be used in diagnostic assays for the detection of S. suis infections and/or for use in vaccines to protect against the disease. In previous work, muramidase-released protein (MRP) and extracellular-factor protein EF were identified as markers of virulence in serotype 1 and 2 strains. In other serotypes the production of MRP and EF, and their potential importance for bacterial virulence has not been investigated. Therefore, we determined the serotypes as well as MRP and EF phenotypes for a collection of S. suis strains isolated from diseased pigs in seven European countries. Overall, S. suis serotype 2 appeared to be most prevalent (32Œ followed by serotype 9 (20€and serotype 1 (12Ž EF-positive strains, were found in serotype 1 (66Œ 2 (71€and 14 (85€strains. Variants of MRP (MRP* or MRPs) were found in nearly all serotypes. A high percentage (81€of the serotype 9 strains belonged to the MRP*EF- phenotype. For the detection of pigs carrying virulent serotype 2 strains, serotype 1, 1/2, 7, 9 and 14 strains, Multiplex PCR tests have been developed. In Multiplex PCR 1, three DNA targets, based on the S. suis serotype 1 (and 14), 7 and 9 specific capsular polysaccharide (cps) genes, were amplified. In Multiplex PCR II, two other targets, based on the serotype 2 (and 1/2) specific cps gene and the epf gene encoding the EF-protein, were amplified. The evaluation of these PCRs for use on tonsillar specimens of diseased pigs demonstrated that the assays were highly specific and sensitive. For the development of protective vaccines, the efficacy of a MRP and EF vaccine applied in pigs was tested. Pigs were vaccinated twice and challenged intravenously with virulent S. suis serotype 2 strains. At challenge, pigs vaccinated with MRP and EF were protected against infection and disease. Pigs vaccinated with either MRP or EF were less well protected. Apparently the combination of both proteins is necessary to obtain full protection. The protective efficacy of an avirulent, non-encapsulated isogenic mutant of S. suis serotype 2 was determined in pigs, and compared with the efficacy of the capsulated wild-type strain. Vaccinations were with formalin-killed cells of the wild-type (WT-BAC), formalin-killed cells of the non-encapsulated mutant (CM-BAC) or with the live non-encapsulated mutant (CM-LIVE) strain. Pigs were challenged intravenously with the homologous, wild-type S. suis serotype 2 strain. The results demonstrated that, as expected, the formalin-killed cells of WT-BAC induced complete protection in pigs against mortality and morbidity after challenge. The formalin-killed cells of CM-BAC induced complete protection against mortality, but only partial protection against morbidity. The CM-LIVE vaccine induced only partial protection, both against mortality and morbidity. These findings indicate that CPS and other bacterial components of WT-BAC are probably essential for full protection against homologous challenge.