(b) Description of Prior Art Streptococcus suis, and particularly serotype 2, is responsible for various pig's infections as meningitis, pneumonia, arthritis, endocarditis, abortion and septicemia (Higgins et al. 1992. Can. Vet. J. 33: 27-30).

Economic losses caused by S. suis are very important in porcine industry, either because of mortality or lower growing rates.

Furthermore, S. suis can be involved as a zoonotic agent and many meningitis and endocarditis cases have been diagnosed in humans having contacts with contaminated pigs or pork meat (Trottier et al.

1991. Rev. Infect. Dis. 13: 1251-1252). Serotype 2 is the major serotype of S. suis. Others serotypes may also be involved in different pathologies.

Different bacterial structures or bacterial products as capsule, fimbriae, extracellular or cell wall associated proteins and hemolysin have been considered as virulence factors. S. suis serotype 2 capsule is composed of five sugars and sialic acid. Even if previous studies suggest that the capsule may act as an anti-phagocytic factor (Charland et al., 1998. Microbiology, 144: 325-332, Miami, Brazeau et al. 1996.

Microbiology 142: 1231-1237), some S. suis serotype 2 strains well encapsuled seems to be totally non-virulent and possess the same acid sialic concentration (Charland et a/., 1996. FEMS Immunol. Med.

Microbiol. 14: 195-203). Previous studies, performed firstly with spontaneous mutants (Gottschalk et al. 1992. Vet. Microbiol. 30: 59-71) and, recently, with isogenic mutants obtained from a unique insertion of Tn916 transposon (Charland et a/., 1998. Microbiology, 144: 325-332),

show that the capsule is essential to strain virulence. It seems that the capsule is a necessary virulence factor, but other structures or components play an important role in virulence. Some proteins have been suggested as being virulence factors. Proteins MRP ("Muraminidase Released Protein") and EF ("Extracellular Factor") are some of those proteins (Vecht et al. 1991. Infect. Immun. 60: 550-556).

It has also been observed that only few of Canadian S. suis strains obtained from sick pigs have these proteins (1 % of the strains) ( Gottschalk et al., 1997, Can. J. Vet. Res. ). Moreover, mutants lacking the MRP and EF proteins obtained from a highly virulent MRP+EF+ strain have conserved their virulence (Smith et al. 1996. Infect. Immun.

64: 4409-4412). A hemolysin produced by S. suis serotype 2, the suilysin, has been recently described, In recent studies, a difference between European and Canadian strains has been noted; 58 to 90% of European strains produce the suilysin and only 1% of Canadian strains produces it (Gottschalk et al., 1997, Can. J. Vet. Res. ). The suilysin production does not seem to be a virulence marker for North American strains.

Controlling this infection is generally difficult. In addition to the means for reducing stress factors, the S. suis infections control is based on antibiotherapy, antibioprophylaxy and vaccination. Strategic antibioprophylaxy, administered in high risk periods, is only good for delaying clinical symptoms (Martineau et al. 1992. Recueils des conferences. Rencontres internationales de production porcine.

Bretagne, France. pp 66-76.). Early weaning, which is quite useful for eliminating many porcine pathogens does not give satisfactory results in the control of S. suis infections (Martineau et al. 1992. Recueils des conferences. Rencontres internationales de production porcine.

Bretagne, France. pp 66-76.).

In the case of vaccination, results obtained up to now are disappointing. Many different approaches have been tested: a) Bacterins (whole killed bacteria): There is no clear scientific evidence that vaccines using whole killed bacteria can control the infection. Some experimental results show that bacterins are not very

efficient for protection and/or antibodies induction (Blouin et al. 1994.

Can. J. Vet. Res. 58 : 49-54; Ripley R. 1983. Proc. Pig Vet. Soc.

8: 25-39). Is has been observed that repeated inoculations with formolized bacteria can induce protection (Ripley R. 1983. Proc. Pig Vet. Soc. 8: 25-39; Holt et al. 1990. Res. Vet. Sci. 48: 23-27). It has also been observed that bacterial inactivation with heat cannot induce an efficient protection, which lead to believe that protein antigenic determinants are necessary for the induction of a protection. b) Capsular polysaccharide : It is well known that anticapsular antibodies are important for host defense in several bacterial species as Haemophilus influenza, Streptococcus pneumoniae, group B Streptococcus and Neisseria meningitidis. In the case of S. suis, polysaccharides injections alone has not been efficient for inducing an immunity (Elliott et al. 1980. J. Hyg. Camb. 85: 275-285). It is possible that coupling polysaccharides to a carrier protein increase anti-capsular immune response. However, anti-capsular antibodies do not seem to have a correlation with protection (del Campo Sepulveda et al., 1996.

Vet. Microbiol. 52: 113-125) and a monoclonal antibody directed against a capsular epitope only offered partial protection (Charland et al. 1997.

Microbiology, 143: 3607-3614). Finally, our preliminary results and Holt et coll.'s work (Holt et al. 1988. Res. Vet. Sci. 45: 349-352) indicate that a slightly encapsulated strain can induce an adequate protection. c) Extracellular or surface proteins: It has been shown that IgG and IgM directed against structures that are on bacteria's surface (mostly proteins) are responsible for immunoserum's protective activity (Holt et al. 1989. J. Comp. Path. 100: 435-442). Several proteins have been identified as being capable for protective antibodies generation in rabbit (Holt et al. 1989. J. Comp. Path. 100: 435-442; Vecht et a/. 1991.

Infect. Immun. 60: 550-556). It is important to note that some of those proteins are found either in virulent strains and in non-virulent strains, which can explain partially the protection obtained using non-virulent strain immunizations. Studies effected in Netherlands show that a sub- unit vaccine constituted of the hemolysin induces an adequate protection in mouse and pig models of infection (Jacobs et al. 1996.

Vet. Rec. 139: 225-228). However, it was shown that most Canadian

strains do not produce that toxin (Gottschalk et al., 1997, Can. J. Vet.

Res). d) Live vaccine (non-virulent strain): It has been shown that a non-virulent strain can protect against a challenge with a virulent strain (Foster et al. 1994. Vet. Res. Commun. 18: 155-163; Kebede et al. 1990.

Vet. Microbiol. 22: 249-257). The strains used were heat-sensitive mutant strains (serotype 2), streptomycin-dependent mutant strains (serotype 1/2) or natural non-virulent strains (serotype 2). Assays executed with a heat-sensitive strain gave encouraging results but those strains have a highly reversion potential. Streptomycin- dependent strains had provided only a partial protection against S. suis serotype 2. Finally, several natural non-virulent strains present a good protection (Kobisch et a/. 1995. Journees Rech. Porcine en France. 27: 97-102; Quessy et al. 1994. Can. J. Vet. Res. 58: 299-301); however, reversion risk after in vivo passage and human lack of virulence for those strains are still to be verified. The protection obtained by non- virulent strains may be caused by cellular immunity. Due to the importance of phagocytes in the resistance to S. suis infection, cellular immunity is believed to play an important role. Moreover, we have observed that convalescent (and protected) animals do not have a high antibodies level against the bacteria. This cellular protection might also be directed against different serotypes.

It would be highly desirable to be provided with a well characterized non-virulent S. suis serotype 2 mutant strain to be used as a live vaccine. This vaccine could be used to produce a humoral response against surface proteins and also stimulate cellular immunity.

SUMMARY OF THE INVENTION In accordance with the present invention there is provided an isolated or recombinant nucleic acid encoding a aro gene cluster of Streptococcus suis or a gene fragment derived thereof.

The nucleic acid in accordance with a preferred embodiment of the present invention, wherein the nucleic acid hybridize to a nucleic acid encoding a gene derived from a Streptococcus suis of any serotype or untypable strain.

The nucleic acid sequence in accordance with a preferred embodiment of the present invention, wherein the sequence is set forth in Fig. 7 (SEQ ID NO : 1).

In accordance with the present invention, there is provided a nucleic acid probe or primer derived from the nucleic acid of the present invention, the probe or primer allowing species, detection and/or identification of Streptococcus suis.

The probe or primer in accordance with a preferred embodiment of the present invention, further comprising at least one reporter molecule.

In accordance with the present invention, there is provided a diagnostic test comprising the probe or primer of the present invention.

In accordance with the present invention, there is provided a protein or functional fragment thereof encoded by the nucleic acid of the present invention.

In accordance with the present invention, there is provided a mutant virulent Streptococcus strain having a mutation of a aro gene cluster.

The strain in accordance with a preferred embodiment of the present invention, wherein the strain is free of capsular polysaccharide and/or sialic acid as a consequence of the mutation of aro gene cluster.

The strain in accordance with a preferred embodiment of the present invention, wherein the Streptococcus is Streptococcus suis.

The strain in accordance with another embodiment of the present invention, wherein the Streptococcus suis is of serotype 2.

The strain in accordance with a preferred embodiment of the present invention, wherein the strain is recombinant.

The strain in accordance with a preferred embodiment of the present invention, comprising at least a part of the aro gene cluster.

In accordance the present invention, there is provided a vaccine to immunize a mammalian host against a Streptococcal disease comprising an amount of a strain of the present invention sufficient to

elicit an immune response in association with a pharmaceutical acceptable carrier.

The vaccine in accordance with a preferred embodiment of the present invention, wherein the mammalian host is selected from the group consisting of human, porcine, bovine, caprine, ovine and equine.

The vaccine in accordance with another embodiment of the present invention, wherein the mammalian host is porcine.

The vaccine in accordance with another embodiment of the present invention, wherein the strain is capable of expressing a Streptococcus antigenic determinant.

The vaccine in accordance with a further embodiment of the present invention, wherein the strain is capable of expressing a non- Streptococcus antigenic determinant.

The vaccine in accordance with an embodiment of the present invention, wherein the non-Streptococcus antigenic determinant is derived from a pathogen.

The vaccine in accordance with another embodiment of the present invention, further comprising at least a second Streptococcus suis immunogen and/or strain.

The vaccine in accordance with an embodiment of the present invention, further comprising at least one adjuvant or any other vehicle adapted to increase or modify immune response of the host.

The vaccine in accordance with an embodiment of the present invention, wherein the immunogen is derived from a pig pathogenic virus or microorganism.

The vaccine in accordance with an embodiment of the present invention, wherein the immunogen is selected from the group consisting of Actinobacillus pleuropneumoniae, Pseudorabies virus, Porcine Influenza virus, Porcine Parvovirus, Transmissible Gastroenteritis virus, rotavirus, Escherichia coli, Erysipelothrix rhusiopathiae, Pasteurella multocida, Bordetella bronchiseptica.

In accordance with the present invention, there is provided the use of the strain of claim 9 or 10 to raise polyclonal antibodies against the mutant in a host.

In accordance with the present invention, there is provided a passive immunization composition to protect a host against a Streptococcal disease comprising an immunoprotective amount of antibody obtained in the present invention in association with a pharmaceutical acceptable carrier.

In accordance with the present invention, there is provided a method for controlling or eradicating a Streptococcal disease in a population comprising administering the vaccine of the present invention to subjects of the population.

The method in accordance with a preferred embodiment of the present invention, wherein the vaccine is administered intramuscularly, intravenously, intranasally, subcutaneoulsy or orally.

In accordance with the present invention, there is provided a method for controlling or eradicating a Streptococcal disease by the use of inhibitors and/or analogues of the aromatic biosynthesis pathway to prevent the expression of capsular polysaccharides by pathogenic streptococci.

In accordance with the present invention, there is provided the use of the composition of the present invention for controlling or eradicating a streptococcal disease in a population by administering said composition to subjects of said population.

In accordance with the present invention, there is provided the use of inhibitors and/or analogues of the aromatic biosynthesis pathway to prevent the expression of capsular polysaccharides by pathogenic streptococci.

For the purpose of the present invention the following terms are defined below.

The term"population"is intended to mean a group of organisms of the same species occupying a particular geographic region.

All references herein are hereby incorporated by reference.

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 illustrates the organization of the aro locus of S. suis ; Fig. 2 illustrates the enlargement of the transposition insertion site and preparation of the construction for the allelic exchange; Fig. 3 illustrates the detection of the capsule of several S. suis mutants using the Z3 monoclonal antibody against capsular sialic acid; Fig. 4 illustrates the PCR control of allelic exchange mutant strain J119 ; Fig. 5 shows electromicroscopy photos of the wild type strain (a), mutant B524 (b) and mutant J119 ; Fig. 6 is a Northern Blot picture illustrating the detection of the aro transcript in the Wild type (WT), J119 and B524 mutant strains; and Figs. 7A-D illustrate a nucleic acid sequence encoding a aro gene cluster of Streptococcus suis.

DETAILED DESCRIPTION OF THE INVENTION In accordance with the present invention, there is provided a mutant virulent S. suis serotype 2 strain to be used as a live vaccine.

This vaccine could be used to produce a humoral response against surface proteins and also stimulate cellular immunity.

Material and methods Bacterial strains and media S. suis reference strain S735 (ATCC 43765) and field strain 31533 (*) isolated from deceased pigs were used. Acapsular mutant S. suis 2A has been previously isolated in our laboratory (Charland et al., Microbiology 144 (pt 2): 325-332). E. coli strains DH5a and JM109 used for molecular biology has been provided by Dr J. Harel, and grown on LB medium, containing 40 ug/ml kanamycin (Boehringer Mannheim) when required. E. coli strain AB2829 (gin42 (AS), A-, aroA354) has been kindly provided by the E coli Genetic Stock Center (CGSC:

Webpage http : //cgsc. biology. yale. edu). The S. suis strains were grown on Todd-Hewitt medium supplemented when required with 400 pg/ml of kanamycin.

DNA and molecular biology materials Dr. Craig Rubens (Seattle, USA) has provided pCIV2 and pVE6007. The E. coli expression vector pMT020 came from our collection. Restriction enzymes, T4 DNA ligase, Taq DNA polymerase and calf intestinal alcaline phosphatase were purchased from Pharmacia. The PCRTM purification kit and the Gel extraction kit were purchased from Quiagen and used as specified by the manufacturer. Some PCR fragments were cloned using the TA Cloning Kitm (Invitrogen), as described by the manufacturer. Probes labeling and Northern blots were performed using the DIG Non-radioactive hybridization system (Boehringer Mannheim). Routine DNA techniques were performed as described by Sambrock et al,. 1989. Custom made sequencing primers were purchased from Life Technologies and BioCorp. Sequencing was performed by the University of Maine DNA Sequencing Facility, and the BLAST software was used to determine nucleotidics and protein sequences homologies in the GenBank/EMBL databases.

DNA constructions To prepare the pBEA756 vector for S. suis mutagenesis, the temperature sensitive Gram-positive origin of replication was amplified from the pVE6007 plasmid using the TS1 F and TS1 R primers. The 1.5 kb amplified fragment was then digested with Kpn 1, and ligated in the Kpn 1 linearized pCIV2. To obtain single insertion mutants, PCR- amplified fragments were Eco R1 digested and ligated in the corresponding site of pBEA756. The deletion section for allelic exchange was prepared by amplifying two arms of about 450 and 550 pb that flanks the deleted region from the S. suis 735 genomic DNA; these two arms were then digested and ligated by the Bgl II site (see Fig. 2), and the corresponding 1000 bp fragment was gel purified and re-amplified. The resulting fragment was then Eco R1 digested and ligated into pBEA756, to obtain pBEA 860.

Isolation of S. suis genomic DNA S. suis was grown overnight on blood agar and harvested with a bent Pasteur pipette on 1 ml of TE buffer into Eppendorf tubes. They were then centrifuged 5 min (full speed), re-suspended in 100 pLI TE, and incubated 30 min at 37°C with 20 lli of 100 mg/ml lysozyme, then next re-incubated with 40 ul of a 20 mg/ml pronase solution. The cells were then lysed in 500 pLI of GES buffer (60% w/v guanidium thiocyanate, 40 % v/v of pH 8.0 250mM EDTA, 5% v/v of a 10% laurylsarcoside), then 250 Ll of 7.5mM ammonium acetate pH 7.7 was added and the mixture was incubated 15 min on ice. The genomic DNA was then extracted two times with 500 1ll of phenol/chloroform/isoamylic alcohol (25: 24: 1), then the remaining aqueous phase was precipitated with 0.54 vol of chilled isopropanol, for 30 min at-20°C. The DNA pellet was then washed twice with 1 ml of 75% ethanol, then re-suspended in 50 tl of TE buffer or water.

PCR identification and cloning of the insertion site S. suis genomic DNA (strain S735 or 2A) was digested with the appropriate restriction enzyme, as determined by Southern blot experiments. After removal of the enzyme by phenol-chloroform, it was ligated in the pKS (+) vector previously digested with the same enzyme and treated with calf intestinal alcaline phosphatase. A PCR was then performed on the ligation mixture using a primer designed for a known section of the bacterial DNA, starting with the Tn916 sequence, with either the T3 or T7 primers for the pKS vector. The amplified fragments were then used directly for sequencing, or first cloned in the pCR 2.1 vector using the TA-cloning kit (Invitrogen) as described by the manufacturer.

Southern blots DNA was separated on a 0,8% agarose gel, depurinated in 0.25N HCI, denatured in 0.25N NaOH, 1.5M NaCI and neutralized in 0.5M Tris, 1.5M NaCI pH7.4 (15 min each). The DNA was then fixed for 15 min in 20X SSC and blotted onto a nylon membrane using the

Schleicher & Schuell TurboBlotterm for 2 hours. The membrane was dried for 30 min and UV irradiated for 3 min to cross-link the DNA to the membrane. The hybridization was performed using the DIG system (Boehringer Mannheim), as specified by the manufacturers. Briefly, the membrane was pre-hybridated in the standard hybridization buffer (5X SSC, 0. 1% N-lauroylsarcosine, 0.02% SDS, 1% blocking reagent) for 1 h30 at 68°C, then the buffer was replaced by the one containing the DIG labeled probe (5 ng/ml) previously denaturated in boiling water for 10 min. Hybridization was done overnight at 68°C in a VWR 2710 hybridization oven. The membrane was then washed 2 x 5 min in 2X SSC, 0. 1% SDS at 20°C, then 2 x 15 min in 0. 1X SSC, 0. 1% SDS at 68°C, blocked 1 h in maleate buffer containing 1 % blocking reagent and, 30 min in the same buffer with the DIG-Alcaline phosphatase conjugate (1/10000). The membrane was next washed twice for 15 min in maleate buffer, soaked in the CSPD chemiluminescent substrate (Boehringer Mannheim) in the detection buffer (0. 1M Tris, 0. 1 M NaCI, 50mM MgCiz, pH 9.5) and the hybridized probe was visualized by exposing the membrane to an Amersham HyperfilmTM MP autoradiography film. To label the DNA fragments used as probes, 0.5 to 3 zig of heat-denaturated DNA is cooled quickly on ice, then 0.1 vol of hexanucletides, 0.1 vol of DIG/dNTP mix and 0.05 vol of Klenow enzyme were added in a total volume of 20 11, incubated for 4 to 6 hours at 37°C, then precipitated with LiCI and ethanol, washed in 70% ethanol, and finally re-suspended in 50 zizi TE.

Trizol RNA extraction RNA has been extracted using the Trizol reagent (Life Technologies). Briefly, a pellet of about 109 cells was re-suspended in 100 pl of 25 g/mi lysozyme for 20 min at 37°C to digest the cell wall.

The cells were then lysed by repeated pipetting in 1 ml of Trizol reagent.

After the complete dissolution of the cells, 0.2 ml of chloroform was added, then the mixture was centrifuged to separate the phases. An upper, clear, aqueous phase containing RNA was transferred to a new tube and the RNA was precipitated by adding 0.5 ml of isopropanol, mixing, and settling for 10 min at room temperature, then centrifuged.

The RNA pellet was then washed with 1 mi of 75% ethanol,

resuspended in 20 jJ of DEPC-treated water and stored at-80°C until use.

Northern blots The RNA samples (5 to 10, ut) were denatured for 10 min at 65°C, then separated 90 min at 100V in a 1,1% agarose formaldehyde gel. The gel was then rinsed in DEPC water, equilibrated twice for 15 min in 20X SSC, then blotted onto a nylon membrane using the Schleicher & Schuell TurboBlotter for 2 hours. The RNA was then cross-linked by irradiating for 3 min on the UV light, and pre-hybridized for 90 min at 50°C in the Hi-SDS buffer (7% p/v SDS, 50% v/v formamide, 5X SSC, 2% p/v casein, 0. 1 % p/v N-lauroylsarcosine, 50mM sodium phosphate, pH7.0). This buffer was then replaced by the one containing the probe to be hybridized overnight. The membrane was next washed twice for 5 min in 2X SSC, 0. 1% SDS at 20°C, and twice for 15 min in 0. 1X SSC, 0. 1% SDS at 68°C. It was next blocked for 1h in the blocking reagent 1%, then incubated for 30 min with the DIG-Alcaline phosphatase conjugate (1/10000). The membrane was next washed twice for 15 min in maleate buffer, soaked in the CSPD chemiluminescent substrate (Boehringer Mannheim) in the detection buffer (0. 1M Tris, 0. 1M NaCI, pH 9.5) and the hybridized probe was visualized by exposing the membrane to the Amersham HyperfilmTM MP film. To label the DNA fragments used as probes, 0.5 to 3 g of heat- denaturated DNA is cooled quickly on ice, then 0.1 vol of hexanucleotides, 0.1 vol of DIG/dNTP mix and 0.05 vol of Klenow enzyme were added in a total volume of 20 pI, incubated for 4 to 6 hours at 37°C, then precipitated with LiCI and ethanol, washed in 70% ethanol, then re-suspended in 50 lli TE.

Transformation of E. coli An overnight culture of the appropriate E. coli strain was diluted 1/50 in 100 ml of LB broth, and grown at 37°C in a rotary shaker (250 rpm) up to an OD600 of 0.5 to 0. 8. The culture was then transferred to 50 mi conical screw cap tubes (Falcon), chilled 15 min on ice, and centrifuged at 1250 x g for 15 min. The pellet was then re-suspended in 33 mi of 100mM KCI, 30mM potassium acetate, 60mM Cal2, 15%

glycerol, pH 5.8 and kept on ice for 1 h. The cells were then centrifuged at 1800 x g for 15 min and the pellet re-suspended in 4 ml of 10mM MOPS, 10mM KCI, 75mM CaCI2, 15% glycerol, pH 6.8. After another 15 min of incubation on ice, the suspension was distributed in 200 lli aliquots in sterile Eppendorf tubes and frozen at-80°C. For transformation, the cells were mixed with about 50 ng of DNA, kept on ice for 30 min, heated 45 sec at 42°C in a water bath and cooled on ice for 2 min. They were next incubated 1h at 37°C with 0. 8 ml LB broth, and plated on the appropriate selective medium.

Electroporation of S. suis Electroporation of S. suis strain S735 was performed as described by Smith et a/., (Smith et al. 1995. Microbiology.

141 : 181-188). Briefly, overnight culture of S. suis in Todd-Hewitt broth was diluted 1/50 in Todd-Hewitt broth supplemented with 40mM L- threonin and allowed to grow until OD600 reach 0.4 to 0.5. The cells were then washed two times in 0.1 vol of ice-cold twice-distilled water, two times in the same volume of ice-cold 0.3M sucrose, once in ice-cold 0.3M sucrose plus 15% (v/v) glycerol, then re-suspended in 0.005 vol of ice-cold 0.3M sucrose plus 15% (v/v) glycerol, fractionated in 50 ul and stored at-80°C until ready to use.

Thawed cells were then mixed with 5 ug of plasmid DNA and transferred into a pre-chilled electroporation recipient (electrodes gap: 0.2 mm). Electroporation conditions were 2.5kV, 200D and 25 uF, on the Bio-Rad Gene Pulser apparatus. Immediately after the electrical discharge, the cells were re-suspended in 1 ml of Todd-Hewitt broth containing 0.3M sucrose and incubated 2 h at 37°C before being plated on the appropriate selective medium.

Integration of plasmid and selection of recombinants strains After the electroporation of S. suis with the appropriate plasmid, the cells plated on appropriate selective media (Todd-Hewitt agar containing 400 pg/ml kanamycin) were incubated at permissive. temperature for the Gram-positive origin of replication (28°C). One of the resulting colony containing the unintegrated plasmid had then been multiplied in 2 ml selective broth at permissive temperature, then serially

diluted and plated on selective agar, and incubated at restrictive temperature (37°C). The resulting colonies were verified by PCR to insure the correct integration of the plasmid. While the single insertion mutants has been saved at this point, the clones used for the production of the allelic exchange mutant has been passed 6 times in the non- selective medium (Todd-Hewitt broth without kanamycin) at permissive temperature to allow the excision of the plasmid. The resulting culture has then been serially diluted, plated on the non-selective medium (and some with kanamycin as a control) and incubated at non-permissive temperature. About 10% of the colonies were still retaining the KanR phenotype. To eliminate them, the colonies were double-replicated in TH and TH+ kanamycin agar, and the kanamycin sensitive clones that had lost the plasmid were screened for the acapsular phenotype.

Identification of acapsular mutants (Dot blotting) The clones to be tested has been double-striked on TH agars, and allowed to grow overnight at 37°C. A circular nitrocellulose membrane was then applied to one of the plates and lifted back with the bacterial streaks on it. These membranes were then blocked in Tris buffered saline (TBS) containing 2% casein for 1 h with agitation, then incubated 2 h in the Z3 Mab (anti-Streptococcus suis serotype 2, sialic acid capsular material) undiluted hybridoma culture supernatant. The membrane was then washed in TBS, and incubated 1 h with the anti- mouse/horseradish peroxydase conjugate (Jackson Immunoresearch), diluted 1/4000 in TBS. The membranes were then washed again in TBS before the addition of the 4-chloronaphtol/hydrogen peroxyde based revelator for 15 min, washed in water and dried.

Sialic acid dosage The sialic acid concentration was determined by the thiobarbituric acid assay (Warren, 1963, Methods Enzymol 6 : 463-464).

Serotyping, coagglutination and biochemical identification Capsular reaction test, coagglutination and biochemical identification of wild type (WT) and mutants strains were carried out as

previously described (Higgins and Gottschalk, 1990, J. Vet. Diagn Invest 2: 249-252.

Transmission electron microscopy The visualization of capsule after immunostabilisation was carried out by electron microscopy as described previously by Jacques et a/., 1990 J. Bacteriol 172 : 2833-2838). Briefly the strains to study were mixed with specific serotype 2 antiserum (or negative serum as control) for 1 h at 4°C, then adjusted to an OD540 of 1.8. The cells were then re-suspended in cacodylate buffer (0. 1 M cacodylate, 5% v/v glutaraldehyde, 0.15% w/v ruthenium red, pH7.0) for 2h at 20°C. The cells were then immobilized in 4% (w/v) agar, washed 5X in cacodylate buffer and postfixed in 2% (v/v) osmium tetraoxyde for 2h. They were then washed as above and dehydrated in a graded series of acetone containing 0.05% ruthenium red, washed in propylene oxide and embedded in Spurr low-viscosity resin. The thin sections were then postained with uranyl acetate and lead citrate and examined with a Philips 201 electron microscope at an accelerating voltage of 60kV.

Results Sequencing of the 2A mutant insertion site Eight open reading frames of a reasonable size have been detected from a sequence of 8.3 kb (Fig. 1). The transposon has been inserted in the open reading frame ORF1 ; however, that open reading frame being very short, it present no valuable homology with a bacterial protein. These results are shown in Table 1: The transposon is however inserted at the beginning of the messenger RNA of the ORF 2 and aroA (each one on a DNA strand) between transcription initiation site and starting codon (Fig. 2). Since ORF2 only provides a small homology with ArcRABC operon, which is involved in anaerobic metabolism control (strain 2A grows well in anaerobic conditions), there are only few probabilities that the negative capsule phenotype comes from this ORF. Only ORF aroA and followers are organized as real operons. This ORF presents a significant homology with AroA protein of various bacterial gram positive

and gram negative species. The 3-phosphoshikimate-1- carboxyvinyltransferase (EPSP synthase) is used in aromatics amino acids synthesis and catalyses 3-phosphoshikimate addition of phosphoenolpyruvate (PEP). This reaction is similar to one of the essential steps of the N-acetyl neuraminic acid (NANA), an essential element of the capsule. The AroA protein produced by the ORF may then catalyze N-actetyl-D-mannosamine's PEP addition to form NANA.

The equivalent enzyme in E. coli is the NeuB protein. However, this protein is not very similar to AroA, although it shares some homology within some domains of the AroA ORF. AroA also possesses a certain homology with E. colts MurA and S. pyogenes MurZ proteins which generates the reaction of addition of PEP to N-Acetyl-D-glucosamine (see Table 1).

This ORF is followed by the reading frames homologous to AroK and PheA proteins. This particular arrangement of an operon dedicated to aromatic amino acids synthesis is almost identical to the one found in Lactococcus lactis and Streptococcus pneumoniae, two species phylogenetically very close of Streptococcus suis. Other open reading frames 10 and 12 follows, and present a good homology with numerous regulating proteins, and ORF 13 correspond to a RNA methyltransferase.

Effects of the insertion site A detailed examination of the insertion site of transposon in the 2A strain (see Fig. 2) shows that it is inserted in the leading sequence of messenger RNAs, the mRNA portion comprised between initiation and transcription sites and the beginning of the aroA and ORF 2 reading frames, and this for both DNA strands at the time. This transposition mutant is then very particular, having an acapsular phenotype obtained by transcription interruption instead of by reading frame interruption.

Single insertion mutants In order to determine which one of ORF 2 or aroA is responsible for negative capsular phenotype, three mutants have been prepared using simple inactivation where reading frames ORF 2, aroA and ORF10 are interrupted by plasmidic sequence insertion. To do so, an

internal section of each of the reading frames has been amplified using PCR with following primers: AroA; B6F2 and B6R primers, ORF 2; B2F and B2R primers, and ORF 10, B10F and B10R primers (see table 2).

The'fragments obtained, being of respectively 536,741 and 631 bp, have been digested with Eco R1 and been inserted in the pBEA756 vector to form respectively the plasmids pBEA1604, pBEA1013 and pBEA1002. Those plasmids had first been produced largely in E. coli, extensively washed and resuspended in water at 1g/) J, then, inserted in S. suis using electroporation.

Table 1<BR> Homology search summary ORF Length of AA Function % similarity Gene Organism homology Orf1 Orf2 140 ? ? ? 46 ? S. pneumoniae. aroA 420 3-phosphoshikimate 1-86 aroA S. pneumoniae carboxyvinyl transferase 82 aroA L. lactis aroK 155 Shikimate kinase 61 aroK L. lactis pheA 249 Prephenate dehydratase 70 pheA L. lactis OrM0 90 Requlator ? 47 cpsX S. aga/actiae OrF12 95 Regulator ? 40 cps2A S. pneumoniae Orf13 260 RNA methyltransferase 59 yefA B. subtilis 52 Ao B. subtilis Table 2<BR> PCR primers used in the present invention Primer Tm % GC Length Sequence (5'-3') BA7 47 39 23 ATTGAGATCTTGTCTTGTCAACC (SEQ ID NO : 12) BA8G 47 39 23 ACTTAGATCTTATATCCCGTTCG (SEQ ID NO : 13) BA9 52 47 21 ACTCGAATTCTACGATGACCG (SEQ ID NO : 14) BA11 49 28 25 TTTGGAATTCATTACCTAAAGTATC (SEQ ID NO : 15) B2F 49 40 25 TCACGAATTCTATATATCCCGTTCG (SEQ ID N0 : 16) B2R 51 26 TGAAGAATTCAGGAAGGACTGATAGC (SEQ ID NO : 17) B6F2 42 31 22 CTATGAATTCAGTAACTATTCC (SEQ ID NO : 18) B6R 48 45 22 CTTTGAATTCCGACGTTTCAAC (SEQ ID NO : 19) B10F 49 40 25 AAATGAATTCGAAAGTCGCGAGTTG (SEQ ID NO : 20) B10R 51 44 25 TCCTGAATTCCCAAAGTATAGGCAG (SEQ ID NO : 21) TS1F 54 52 21 GTTAGGTACCGAAAGCCGATG (SEQ ID NO : 22) TS1R 51 39 23 TTGTGGTACCTAATTCAACTTCC (SEQ ID NO : 23) ARA3 50 47. 23 GACTAGATCTGTCAGCAATAGGC (SEQ) DNO : 24) ARA5B 45 33 24 ATTGAGATCTCACAAGTAATTGTC (SEQ ID NO : 25) Restriction sites within primers are in bold<BR> After a control by PCR of the plasmid insertion site, integration mutant capsular expression has been verified using Dot Blot and sialic acid<BR> dosage. The results obtained had shown that only the B524 mutant in which the aroA gene is inactive shows a significant reduction of the<BR> capsular quantity produced. The inactivation of ORFs 2 and 10 (mutants D2 and A10) does not show a significant reduction of the quantity of<BR> capsular material produced (see Table 3 and Fig. 3). Transcription interruption of ORF 2 by the transposon has no effect on the capsule (D2<BR> strain). The ORF 10 has also been tested to verify the potential effect of these regulators on the capsule expression (A10 strain), with no<BR> visible effects.

Preparation of the allelic exchange mutant: The construction prepared for the allelic exchange is reproducing the effect produced by the transposon on the operon aro by deleting a DNA section of 420 bp containing the promoter and the beginning of the aroA gene. To do so, two arms respectively of 445 bp (BA9 and BA7 primers) and 526 bp (primers BA8G and BA11) flanking the region to be deleted have been amplified using PCR. The two primers immediately flanking the deletion site (BA7 and BA8G) comprised the restriction site Bgl II, and the two arms have been digested by this enzyme, incubated with DNA ligase and the resulting mixture separated on agarose gel.

The 967 bp fragment corresponding to the two arms ligation has been removed from the gel and purified, then re-amplified using primers BA9 and BA11. Those primers containing the restriction site Eco R1, the 967 bp amplified fragment has been digested and inserted in the pBEA756 vector linearized by the same enzyme. The resulting plasmid has also been produced in large quantities in E. coli and introduced in S. suis using electroporation as described previously.

After electroporation, bacteria were incubated at a temperature of 28°C (permissive temperature) with 400 pg/ml of kanamycin for plasmid recipient selection. One colony was therefore grown on Todd Hewitt broth containing kanamycin then spread on Todd Hewitt agar plate, still containing kanamycin, and was incubated at 37°C (restrictive temperature) in order to force plasmid integration into the genome.

After verifying integration in some colonies (all were occurring on the aroA side) by PCR technique, one of them has been subjected to 6 passages in a non-selective medium and at permissive temperature in order to provoque the excision and lost of the plasmid. This culture has then been spread on a non-selective Todd Hewitt plate (some Todd Hewitt plates + kanamycin being used as control) and incubated at restrictive temperature. At this point, about 90% of the colonies have lost the plasmid and became KanS. The colonies were then cultured in duplicate in a medium with or without kanamycin for resistant strain elimination and the sensitives colonies which have lost the plasmid have been tested for the acapsular phenotype by Dot Blotting using a

monoclonal antibody directed against the sialic acid epitope contained in the capsule and an alcaline phosphatase conjugate. Around 1 on 120 Kans colonies were presenting the acapsular phenotype, the other ones being revertants where the second crossing-over had occurred on the same side of the first one.

Control of the allelic exchange mutant The mutant strain J119 was selected as allelic exchange mutant has been tested by PCR technique for the deletion and the loss of the plasmid (see Fig. 4). The PCR control executed on genomic DNA extracted from J 119 mutant strain using primers BA9 and BA11 show a reduction of the length of the amplified fragment of 420 bp compares to the wild strain. This fragment is shortened from 1.4 kb (wild type, Fig. 4, lane 2) to 0.96 kb (mutant J119, lane 3), which correspond to the length of the deleted region and corresponds also to the length of the construction inserted in the pBEA860 plasmid (lane 5). Lane 4, which is wild type equivalent, is a revertant (J124) having the plasmid excision done on the same side of the deletion than of the integration. A controlled amplification on a plasmid vector section (primers TS1F and TS1 R, replication origin ts) shows the absence of the vector in the wild strain. A last amplification using a primer outside the concerned region (BA1 and BA11) shows also a difference of 420 bp between the amplified fragments of 1.62 kb (WT, lane 11) and 1.2 kb (J119, lane 12).

Deletion integrity as also been verified by sequencing the flanking region of the deletion. Moreover, the acapsular phenotype has also been verified by Dot Blot and the mutant J119 does not show any reaction with the Z3 anti-sialic acid antibody (See Table 3 and Fig. 3).

Sialic acid dosage, coagglutination with a polyclonal antiserum and serotyping using capsular antigen (See Table 3), and electron microscopy (see Fig. 5) have all confirmed the acapsular phenotype of J119 strain.

Table 3<BR> Characteristics of mutants obtained in this work Strain Capsule Sialic acid Coagglutination Serotype Id. S. suis expression 753 ++++ 4. 830. 97 ++++ 2 + 2A 0. 120. 03 NT + J119 0. 06+0. 03 NT B524 ++ 2. 010. 41 + 2 + D2 ++++ 4. 18+1. 02 ++++ ND ND A10 ++++ 3. 910. 81 +++ ND ND 31533 ++++ ND ++++ 2 B212 ND NT + B218 ND NT + NT: not-typable ; ND : not done

RNA expression In Fig. 6 is a Northern Blot showing the expression of mRNAs from the aro locus in the wild-type, allelic exchange mutant, and insertionnal inactivation of AroA mutant, probed with aroA. Black arrow indicates the normal 3.95 kb mRNA which spans from the aro promoter to the end of ORF 10, and disappears in both mutants. White arrow indicates the 1.85 kb truncated mRNA of the B524 insertionnal inactivation. J119 allelic exchange mutant did not express aroA mRNA.

Northern hybridization with the ORF 2 probe had failed to show any expression of this reading frame, thus providing another evidence that this gene is not implicated in the capsule synthesis (results not shown).

Transcription from the aro promoter in the wild-type results in a 3.95 kb mRNA, which ends downstream of the Orf 10, when probed with aroA.

The B524 mutant express the expected 1. 85 kb truncated mRNA being interrupted in the aroA reading frame. J119 allelic exchange mutant did not express any aro mRNA from this locus, as a result of the deletion of its promoter. Use of the Orf 10 probe show a wild type mRNA identical as the one detected with aroA probe, which support the fact that this aro mRNA include the Orf 10 regulatory reading frame (results not shown).

As the Orf 10 mutant did not show any alteration of the capsular expression, its gene product is likely to act in the regulation of the aromatic amino acids synthesis pathways.

Complementation of E. coli aroA gene In order to verify the functionality of this aroA gene, the gene was cloned and expressed in an E. coli aroA strain. However, this S. suis DNA region is very unstable when passed in E. coli, so the clones expressing AroA on selective agar plates have been selected directly at the transformation. To do so, the complete reading frame of the protein corresponding to the wild strain 735 has been amplified using the primers ARA3 and ARA5B (the fragment obtained is 1.5 kb in length) and inserted directionally into the inducible expression vector pMT020 by the sites 6grand Xba 1 comprised in these primers. This vector contains a resistance gene to chloramphenicol and the replication origin

of the plasmid pACYC184 (low to medium copy number), plus the Lac repressor and promoter beside which is inserted the reading frame of the desired gene. The ligation mixture has then been transformed in the E. coli AB2829 strain (gin42 (AS), A-, aroA354), and have been spread on selective agar plates CDM-aro containing IPTG as the promoter Lac's inducer. These CDM-aro selectives plates contain chloramphenicol for plasmid recipient selection but didn't contains aromatics amino acids. Only CmR bacteria expressing the AroA gene product should grow on this medium.

Few colonies have been obtained on the selective media CDM- aro; three of them have been chosen for plasmid extraction. The transformation negative control without DNA did not produce any colony on the selective media (so the clones obtained were probably not spontaneous mutants) and a second transformation control using the pBEA1002 plasmid and the kanamycin resistance only produced a moderate amount of colonies on LB-Kanamycin agar plates. The small amount of colonies obtained for the aroA gene complementation can probably be explained by the poor AB2829 strain transformation efficiency with the protocol used in comparison with the JM109 strain usually used. Plasmid extracted from resistant colonies digestion with Bgl II and Xba 1 allowed to retrieve the 1.5 kb (ORF aroA) and 4,3 kb (pMT020) fragments along with others fragments probably arising from recombination events. These colonies are also resistant to chloramphenicol, the selection marker carried by the pMT vector. Even if the aroA DNA undergoes heavy recombination in E. coli, this technique allowed us to complement the aroA mutation of the AB2829 strain with the aroA reading frame of S. suis. The mutated protein AroA of S. suis is then an AroA functional enzyme, capable to complement that of the E. coli.

Complementation of the S. suis aromatic and capsular functions In order to determine if this S. suis AroA protein is used for the aromatic amino acids synthesis, a synthetic media allowing S. suis growth was first developed. This medium has been developed from the CDM media developed for others streptococci. Many recipe variations

have been necessary but the inventors were finally successful in making 735 and J119 strains growth in the medium described in Table 4A, by using a relatively large inoculum. It is also preferable to seed the CDM medium with colonies grown on agar plates instead of broth to broth passage. As shown in Table 4B, the J119 strain did not grow in the CDM medium without aromatic amino acids, and amino-, hydroxy- and dihydroxy-benzoate. Not only the AroA mutated protein is really functional, but the J119 strain is also a double attenuated mutant, as it need aromatic amino acids to grow.

Preparation of others acapsular mutants from a different strain The same plasmid pBEA860 had been used to obtain another acapsular mutant, starting from the highly virulent S. suis 31533 strain.

Introduction by electroporation in the 31533 strain, integration, and excision of the plasmid has been done using exactly the same protocols developed for strain 735. Final screening for acapsular phenotype has yielded 2 allelic exchange mutants over 120 clones tested, that had been called B212 and B218. Verification of the genotype by PCR, acapsular phenotype using the Z3 anti-sialic acid antibody, serotyping and coagglutination had all given the same results than the J119 mutant (see Table 3). These mutants are then functionally equivalent.

Table 4A Composition of the CDM medium Compound Qty/liter KH2PO4 1000 mg K2HP04 200 mg Na2HP04 7.35 g NaH2P04 H20 3.195 g NaHC03 2.5 mg CaCl2#6H2O 10 mg (NH4) 2S04 0.6 g MgS04-7H20 700 mg NaCI 10 mg FeS04*7H20 10 mg Fe (N03) 3*9H20 1.0 mg MnS04 10 mg Riboflavin (B2) 2 mg Biotine (H) 0.2 mg Folic acid 0.8 mg Pantothenate (B5, Na salt) 2 mg p-aminobenzoic acid 0.2 mg Thiamine-HCI 1 mg Cyanocobalamine (B12) 0.1 mg (3-Nicotinamide adenine 2.5 mg dinucleotide Nicotinamide (Niacinamide) 1.0 mg Pyridoxal 1.0 mg Pyridoxamine 1 mg dihydrochloride Uracile 30.0 mg Thymine 4.0 mg Spermidine 5.0 mg Deoxyguanosine 8.0 mg Adenylic acid (AMP) 16.0 mg Cytidylic acid (CMP) 50.0 mg Adenine 35.0 mg Guanine-HCI 27.0 mg Compound Qty/liter (continued) Tween 80 1.0 g D-glucose 10 g Asparagine 200 mg Glutamin 300 mg L-glutamic acid 300 mg L-lysine 125 mg L-aspartate 100 mg L-isoleucine 100 mg L-leucine 100 mg L-methionine 100 mg L-serine 100 mg L-threonine 200 mg L-valine 100 mg DL-alanine 200 mg L-arginine 200 mg L-cystine 200 mg L-cysteine 500 mg L-histidine 200 mg glycine 200 mg L-hydroxyproline 200 mg L-proline 200 mg L-phenylalanine 100 mg L-tryptophane 200 mg L-tyrosine 200 mg p-hydroxybenzoic acid 0.14 mg 2,3-dihydroxybenzoic 0. 16 mg acid <BR> <BR> <BR> <BR> ddH20 To 1000ml Table 4B Growth of mutants in the CDM and CDM-Aro media Species/strain CDM CDM-Aro E. coli JM109 ++++ ++++ E. coli AB2829 ++++ 1- S. suis 735 +++ +++ S. suis J 119 +++ S. suis 2A +++

Protection tests in pigs The protective effect of the J119 acapsulated S. suis strain in healthy pigs was evaluated. Seven four-week old piglets were vaccinated with the J119 strain (vaccinated, group 1); another group of 7 piglets were not vaccinated but received the challenge (non- vaccinated, group 2) and a third group of 6 piglets did not receive any treatment (control, group 3). The animals received two dose of the vaccine (at 4 and 6 weeks old) by intramuscular injection with a 2 ml bacterial suspension of 1x109 CFU/ml concentration. At vaccination, fever, clinical symptoms, S. suis presence in blood, food intake and weight gains have been monitored. The challenge was carried out 14 days after the second vaccine dose by intravenous injection (109 CFU/animal) in groups 1 and 2 with the S735 strain. Animals were controlled with the same parameters as described previously.

Moreover, pathologic lesions studies and bacteriological isolation from different organs have been performed. The antibodies response in all animals has been verified with an indirect ELISA test developed in house, using whole acapsular bacteria as antigen. As positive and negative controls, a serum coming from an immunized pig with S735 strain and a serum coming from a healthy (EOPS) pig have been used, respectively.

Vaccination After the first J119 vaccine dose, a slight hyperthermia (40.2 and 40. 7°C) has been observed on animals of group 1 (See Table 5). This hyperthermia has also been observed in four animals after the second

immunization. The body temperature returned to normal for all vaccinated piglets and has been analogous to that of the control group's piglets. J119 vaccination had no other symptom in animals of group 1.

In addition, the J119 strain was not isolated in high quantities from the bloodstream of those animals. It was only found on days 2 (for 3 pigs) and 4 (for 1 pig) after the first immunization and on 2 animals after the second immunization. In most cases, an enrichment of the blood sample was necessary for bacterial isolation. The injection point did not show any local lesion. Finally, the vaccination had only a temporary effect on the growth of the vaccinated animals. The three experimental groups presented a similar growth and a daily weight gain after 2 weeks post-vaccination (Table 6). All those results show that the J119 strain is virulent and can be used safely as a vaccine.

Table 5 Number of animals having a high body temperature (> 39. 9°C) Group 1 Group 2 Group 3 Vaccinated Non- Control Characteristics vaccinated n=7 n=7 n=6 After 1St dose 4h 7 0 0 of vaccine 24h 0 0 0 After 2nd dose 4h 4 0 0 of vaccine 24h 0 0 0 After challenge 24h 7* 7* 0 48h 3* 7* 0 72h 1* 4* 0 * Temperature significantly higher in animals of group 2' (non vaccinated) Table 6 Daily weight gain in piglets (in grams) Periods Control Vaccinated non vaccinated After first immunization J4 to JO 375 450 400 JO to J3 666 600 633 J3 to J7 675 575 725 J7 to J11 525 400 675 J 11 to J 14 1033 1033 866 After second immunization JO to J4 800 725 750 J4 to J7 700 666 800 J7 to J11 700 725 625 J 11 to J 14 833 866 966 After experimental infection JO to J4 775-225 (n=6)-500 (n=7) J4 to J7 733 400 (n=6) 466 (n=5) J7 to J11 875 1125 (n=6) 300 (n=5) J 11 to J 14 800 866 (n=6) 890 (n=3) JO to J11 804 549 (n=6) 43 (n=5)

Experimental challenge Eighteen hours after the experimental challenge, a temperature increase was observed in all pigs from vaccinated and non-vaccinated groups (see Table 5), temperature being higher in the non-vaccinated group. Two days after the challenge, some vaccinated piglets had hyperthermia. In the non-vaccinated group, all piglets still had hyperthermia.

In the vaccinated group, all pigs survived, whereas in the non- vaccinated group, 4 piglets (57% mortality) died or were killed for humanity reasons (Table 7).

In the vaccinated group, 3 piglets presented slight locomotive problems. On the other hand, those problems were significantly higher in all animals of the non-vaccinated group and they persisted until euthanasia. The articulations of these animals were soared, hot and swollen at palpation. Those manifestations have been evidenced during the 2 weeks pi and they have been accompanied with great difficulties to move. Four out of seven animals have been observed to be shaking 24 hours after infection. Nervous symptoms (convulsions) have also been observed in one piglet. Surviving pigs of the 3 groups have been sacrificed two weeks after the infection with the 735 strain.

A post-mortem examination of all the animals was performed with a particular interest for the liver, the heart, the spleen, the lungs, the synovial membranes and the brain.

Table 7 Effects of challenge on animals Animals groups Group 1 Group 2 Group 3 Characteristics Vaccinated non-vaccinated control n=7 n=7 n=6 Mortality* 0 4 (57%) 0 Clinical symptoms locomotive problems + +++ shaking +/-+++ -nervous symptoms 0 1 0 Microscopics lesions Organs: liver, rate, lungs, 0 (0%) 6 (86%) 0 heart articulations 3 (43%) 4 (57%) 0 - brain 0 0% 3 (43%) 0 Bacterial isolation -blood (J4) ** + ++++ O organs 1 (14%) 6 (86%) 0 -articulations 4 (57%) 7 (100%) 0 *Natural or euthanasia for humanity reasons **Mean numbers of bacteria recovered from blood

In vaccinated pigs, most organs were not affected (see Table 7).

For the 7 pigs, only the articulations presented inflammatory symptoms.

A slight fibrin deposit and a synovial liquid excess were observed in articulations of 3 pigs. Macroscopic lesions were observed in non- vaccinated pigs, particularly at the pleura, pericardium and peritoneum.

Fibrin deposits were observed on the liver and the spleen of animals. A red pneumonia and a fibrinal pleurisy were also observed. The challenge also induces a pericarditis in most (5/7) non-vaccinated pigs.

Concerning bacterial isolation from blood after the challenge, the 735 strain was recovered from all vaccinated and non-vaccinated pigs (see Table 7) S. suis concentrations are maximal at 24 hours pi. The

non-vaccinated pigs were the most contaminated, showing concentration near to 106 CFU/ml, which 1000 times higher than in the vaccinated group. Differences in the number of bacteria isolated from blood in both groups remained after 4 days pi (Table 7).

Bacterial isolation was also carried out from different organs.

Most organs from animals of the non-vaccinated group cultured positive for S. suis type 2. In this group, the contamination of the organs from the thoracic cage and of the abdominal cavity by S. suis 735 was often observed (see Table 7).

While the invention has been described in connection with specific embodiments thereof, it will be understood that it is capable of further modifications and this application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains and as may be applied to the essential features herein before set forth, and as follows in the scope of the appended claims.