Veterinary Microbiology, 30 ( 1992 ) 59-71 Elsevier Science Publishers B.V., A m s t e r d a m
59
Production and characterization of two Streptococcus suis capsular type 2 mutants M. Gottschalk, R. Higgins, M. Jacques and D. Dubreuil Groupe de Recherche sur les Maladies In~,ctieuses du Porc. Facult~ de mOdecine vOtkrinaire, Universitk de MontrOal. C.P. 5000. St-Hyacinthe, Qug'bec, J2S 7C6, Canada (Accepted 15 June 1991 )
ABSTRACT Gottschalk, M., Higgins, R., Jacques, M. and Dubreuil, D., 1992. Production and characterization of two Streptococcus suis capsular type 2 mutants. Vet. Microbiol., 30:59-71. Two avirulent mutants of Streptococcus suis capsular type 2 (M2 and M42 ) were produced from a highly virulent strain. Mutant M2, obtained after serial subcultures of the parent strain in the presence of rabbit anti-capsular type 2 serum, no longer possessed the type-specific capsular antigen, as demonstrated by serotyping methods and immunoelectron microscopy. The Lancefield group D antigen could not be detected on the cell surface of this mutant using the immunogold labelling technique. SDS-PAGE of lysozyme treated cells demonstrated that a 44 kDa protein which was present in the parent strain, was absent in mutant M2. Immunoblotting using rabbit whole cell homologous antiserum revealed that the protein was strongly immunogenic. Mutant M2 was totally avirulent in mice, and the homologous antiserum completely failed to protect mice against challenge with the parent strain. However, mutant M42, obtained after passages of the parent strain at 42 °C, remained capsulated but lacked the same 44 kDa protein as mutant M2. The quantity of sialic acid present in the capsule was similar to that of the parent strain. Despite the presence of antibodies against the capsule, antiserum prepared against M42 only partially protected mice against a challenge with the parent strain. The 44 kDa cell wall protein could act as a virulence factor as well as an important immunogen of S. suis capsular type 2.
INTRODUCTION
Streptococcus suis is responsible for a wide range of infections in swine (Higgins et al., 1990b; Vecht et al., 1985 ) and other animal species (Devriese et al., 1990; Higgins et al., 1990a). It shares antigens with Lancefield group D streptococci, from which it appears to be genetically different (Kilpper-B~iltz and Schleifer, 1987). So far, there are 23 identified capsular types of S. suis (Gottschalk et al., 1989). With the exception of the Scandinavian countries and Australia, where the capsular types 7 and 9, respectively, are the most prevalent (Gogolewski et al., 1990; Perch et al., 1983 ), most other countries report that capsular type 2 ofS. suis is more frequently isolated from diseased 0 3 7 8 - 1 1 3 5 / 9 2 / $ 0 5 . 0 0 © 1992 Elsevier Science Publishers B.V. All rights reserved.
60
M. GOTTSCHALK ET AL.
pigs than any other capsular type, and it is mainly associated with septicemia, meningitis and pneumonia (Clifton-Hadley, 1986; Higgins et al., 1990b). Moreover, this capsular type has repeatedly been isolated from cases of septicemia and meningitis in m a n as a result of contact with pigs or pig products (Arends and Zanen, 1988 ). Virulence factors of S. suis have not been identified, and attempts to control the infection in the field by vaccination have given equivocal results. The presence of sialic acid as a component of the capsular material of S. suis capsular types 1 and 2 (Elliott and Tai, 1978) suggests, as is the case for Escherichia coli K 1, Neisseria meningitidis and Streptococcus agalactiae, that these sialic acid-containing polysaccharides might play an important role in protecting microorganisms from the host defense system by preventing the activation of the alternative pathway of complement (Shigeoka et al., 1983). However, pathogenic and non-pathogenic S. suis capsular type 2 strains were described as phagocytosed at a similar rate in the absence of type-specific antibodies, suggesting that the capsule might not play an evident antiphagocytic role and that other virulence factors must be present to inhibit intracellular killing of pathogenic strains (Williams, 1990). Fimbriae and hemagglutinins were also described, but their exact function in S. suis infection remains unclear (Gottschalk et al., 1990; Jacques et al., 1990). In addition, cell wall "M-like" proteins have been described in S. suis capsular type 2 isolates recovered from cases of meningitis, but not in isolates from cases of pneumonia or from clinically healthy animals (Vecht et al., 1989). Antibodies against certain cell wall proteins were present in S. suis capsular type 2 hyperimmune protective serum, but absent in pre-immune non-protective serum (Holt et al., 1989). The study of mutants defective in surface components could help to identify some of the virulence factors of S. suis. In the present study, two S. suis capsular type 2 mutants were obtained from a highly virulent strain, and their characteristics compared to those of the parent strain. MATERIALS AND METHODS
Production of mutants S. suis capsular type 2 strain 89-1591, originally isolated from a case of septicemia and meningitis in a pig, was streaked onto blood agar plates (tryptic soy agar containing 5% bovine blood). For the production of mutants M 1 and M2 the following technique was used (Yeung and Mattingly, 1983): a single colony was chosen and kept as the parent strain. Cells from the same colony were inoculated into 5 ml Todd-Hewitt broth to which 1 ml of rabbitanti-capsular type 2 antiserum was added. A Todd-Hewitt broth containing the same proportion of normal rabbit serum was used as a control. After 6 h of incubation at 37 ° C, a solid clump at the bottom of the tube containing the
TWO S. SUIS CAPSULAR TYPE 2 MUTANTS
61
capsular type 2 antiserum was observed. After gentle agitation, 10 #1 from the non-agglutinated phase of the culture were transferred into 5 ml of ToddHewitt broth supplemented with 200/tl of anti-capsular type 2 antiserum, and incubated for 18 h; the procedure was repeated several times. After each transfer, cells from the non-agglutinated phase were streaked onto a blood agar plate, and five colonies from each passage were then tested by three different capsular typing methods (described below). Mutant M 1 was obtained after 16 passages whereas m u t a n t M2 was obtained after 33 passages. As both mutants presented exactly the same characteristics, only mutant M2 was studied further. Mutant M42 was obtained after 50 subcultures on tryptic soy agar at 42°C. The parent strain as well as the mutant strains were tested for their biochemical and enzymic profiles using conventional tests (Higgins and Gottschalk, 1990) and a multitest system (Rapid-Strep, API System, Laboratory Products Ltd, St-Laurent, Qu6., Canada.
Capsular typing methods Capsular typing was based on three different techniques: capsular reaction (Neufeld), capillary precipitation and coagglutination tests, as described earlier (Higgins and Gottschalk, 1990). The Lancefield group D antigen was detected in sonicated bacterial suspensions using a coagglutination reagent (Phadebact, Pharmacia Diagnostic, Uppsala, Sweden).
Electron microscopy Preparation for transmission electron microscopy was carried out as described previously (Jacques et al., 1990). When present, capsular material was stabilized with serotype-specific antiserum and stained with ruthenium red. Controls without stabilization or treated with normal rabbit serum were performed simultaneously. Thin sections were examined with an electron microscope (Philips 201 ) at an accelerating voltage of 60 kV. Dense bacterial suspensions were also examined after negative staining (Jacques et al., 1990). A drop of each preparation was placed on a 200-mesh Formvar-coated electron microscope grid and blotted until partially dry. A drop of 2% (w/v) phosphotungstate (pH 7.0) was then applied to the grids, which were observed with the electron microscope. For the localization of the cell wall antigens, grids with bacteria were treated with rabbit anti-Lancefield group D antiserum (Wellcome, Beckenham, England). After washing, gold labeled goat anti-rabbit IgG (5 n m gold colloidal particles, E.Y. Labs, Inc. Biochemical Division, San Mateo, CA) was added. Grids were observed as described above. A strain of Enterococcus faecalis (ATCC 19433) was used as positive control.
62
M. G O T T S C H A E K ET AL.
Quantitation of sialic acid levels Sialic acid was extracted from freeze-dried cells by acid and heat treatments as described previously (Shigeoka et al., 1983 ). Sialic acid levels were determined by the thiobarbituric acid method of Aminoff ( 1961 ).
Production of antisera Production of antisera against the reference strain of S. suis capsular type 2 (NCTC 10234), the parent strain (89-1541), and mutants M2 and M42 was performed as previously described (Higgins and Gottschalk, 1990 ). Capsular reaction and slide agglutination tests were used for the titration of antisera.
Mouse lethality test A virulence test was performed by intraperitoneal injection of 4-week-old mice (strain CF1 ) (Williams et al., 1988; Gottschalk et al., 1990). The 50% lethal dose (LDso) was estimated with groups of five mice. Bacteria were grown in Todd-Hewitt broth (Difco Laboratories, Detroit, Mi), harvested in log-phase growth (5-6 h), suspended in phosphate-buffered saline and adjusted to an optical density of 0.8 (540 n m ) which corresponded to a concentration of approximately 109 C F U / m l . The exact number of viable cells present was counted by a pour plate method, using blood-agar plates incubated for 18 h at 37°C. Death or presence of clinical signs, such as meningitis, was monitored over the next 7 d. The test was repeated four times for each strain.
Mouse protection test For the mouse protection test, five groups of twenty 4-week-old mice (strain CF1 ) were inoculated with a 5-6 h culture of the parent strain mixed with normal rabbit serum (control), or antiserum raised against the parent strain (non-diluted and diluted 1/2), the M2 strain and the M42 strain. In each experiment, the LDso of the inoculum was measured. Colony forming units were determined by a pour plate m e t h o d immediately after animal injections to monitor the exact numbers of viable bacteria inoculated. In this way it was estimated that each inoculum used for the intraperitoneal challenge of mice, contained 12-15 times the LDs0. Dead mice were counted and removed from their cages at 24 h intervals. A culture from heart blood was made onto blood agar plates from approximately 50% of mice which died after challenge in each group, and bacterial identity was confirmed by the capsular reaction test.
SDS-polyacrylamide gel electrophoresis The parent strain and the mutant M2 were grown in 7 ml of Todd-Hewitt broth at 37°C to early stationary phase (6-7 h). In all experiments, mutant M42 was incubated at 42°C. Each inoculum was transferred into 200 ml of Todd-Hewitt broth, incubated for 18 h, collected by centrifugation, washed
T W O X SUIS C A P S U L A R T Y P E 2 M U T A N T S
63
and lysed by sonication. Cell envelopes were obtained after centrifugation (40 000 g, 30 rain), and treated with lysozyme (5 m g / m l ) for 4 h at 37°C. Vertical slab gels of 12% polyacrylamide were employed with a 4.5% stacking gel as previously described (Whiley et al., 1981 ).
Western blotting After SDS-PAGE, separated material was transferred from the slab gel to nitrocellulose membrane by the methanol-Tris-glycine system (Towbin et al., 1979). Electroblotting was performed in a transblot apparatus (Hoefer Scientific Instruments, San Francisco, California) for 18 h at 60 V. After blocking unreacted sites with 2% (w/v) casein solution, the nitrocellulose was incubated for 2 h with the appropriate dilution ofantisera in the same solution. After washing in Tris-NaC1, the sheets were incubated with a peroxidase-labelled IgG fraction of goat antiserum raised against rabbit IgG (Daymar Laboratories, Toronto, Ontario) for 90 min at a dilution of 1 : 500 in a 2% solution of caseine in 10 m M Tris-NaC1. After repeated washing, the presence of bound antigens was visualized by reacting the nitrocellulose membrane with 0.06% 4-chloro-l-naphtol (Sigma Chemical Company, St-Louis, MO) in cold methanol mixed to 0.02% H202 in Tris-NaC1. Apparent molecular weights were calculated by comparison with standards of known molecular weight. RESULTS Conventional tests and the Rapid-Strep system showed that the parent as well as both mutants presented the same biochemical profle. They were identified as typical S. suis. The three strains coagglutinated, after sonication, with group D antiserum. Mutant M42 still showed a typical reaction with anti-capsular type 2 reference serum, whereas mutant M2 was untypeable with serotyping methods. Mutant M2 was confirmed to be non-capsulated by immunoelectron microscopy (Fig. 1 ). Mutant M42 was capsulated, but the capsular material distribution on the cell surface was rather irregular (Fig. 1 ). When the content of sialic acid was evaluated, the parent strain and mutant M42 presented similar values (3.05 X 10 -2 and 2.2 X 10-2/2M/mg of bacteria, respectively). All mutants possessed fimbriae which were morphologically identical to those of the parent strain (data not shown ). When antisera were produced against the parent strain and mutants M2 and M42, anti-capsular type 2 antibodies were present in antisera raised against the parent strain (titer:64) and mutant M42 (titer:32), but not in antiserum raised against mutant M2 (Table 1 ). Despite the absence of the capsule, mutant M2 did not produce antibodies, detectable by the slide agglutination test, against the group antigen, as a negative reaction could be observed with the E. faecalis strain (Table 1 ). When whole cells were treated
64
M. G O T T S C H A L K ET AL.
(2
Fig. 1. Transmission electron micrographs of thin sections of cells of Streptococcus suis exposed to anti-capsular type 2 serum and stained with ruthenium red. A: parent strain (89-1591); B: mutant M2; C: mutant M42. Bar: 200 nm. w i t h specific r a b b i t a n t i - L a n c e f i e l d g r o u p D I g s , a n d t h e n w i t h g o a t a n t i - r a b b i t colloidal g o l d - c o u p l e d Igs a n d o b s e r v e d u n d e r the e l e c t r o n m i c r o s c o p e , a few gold p a r t i c l e s c o u l d b e o b s e r v e d o n this m u t a n t as well as o n the well
65
TWO S. SUIS C A P S U L A R TYPE 2 M U T A N T S
iii
O°
7 Fig. 2. Negative staining of bacterial cells incubated with rabbit Lancefield group D antiserum and colloidal gold labeled sheep anti-rabbit immunoglobulin G. A: Enterococcus faecalis; B: Streptococcus suis parent strain (89-1591 ); C: mutant M2. capsulated p ar en t strain (Fig. 2). In contrast, anti-group D a n t i b o d y labeled with colloidal gold was f o u n d to bi nd ove r the entire surface o f E . faecalis. Both m u t a n t s were non-virulent for mice. T he LDso for the parent strain
66
M. GOTTSCHALK ET AL
TABLE 1 Serotyping results using antisera raised against tants M2 and M42.
Streptococcus
suis
parent strain (89-1591 ) and mu-
Antiserum raised against I
Antigen
Parent strain
Mutant M2
Mutant M42
Capsular type 2 (reference strain NCTC 10237) Parent strain (89-1591 ) Mutant M22 Mutant M42
+ (titer:64)
-
+ (titer:32)
+ -+
+ -
+ -+
E. ,faecalis 3
_
_
_
~The following tests were used: coagglutination, capillary precipitation and capsular reaction. 2Positive only with the coagglutination test. 3Only the coagglutination test was performed. TABLE 2 Protection of mice challenged with S t r e p t o c o c c u s heterologous antisera from immunized rabbits
suis
parent strain (89-1591) by homologous and
Serum used
No. deaths/No, tested ~
% of protection
Normal rabbit serum Antiserum raised against: Parent strain (non-diluted) (dilution 1:2) Mutant M2 Mutant M42
20/20
0
1/20 0/20 18/20 8/20
95 100 10 60
NOTE: The number of streptococci in each inoculum was equivalent to 12-15 times the LDso. IDeath occurring 72 hours after inoculation of mice.
was 3 ( + 1.4) × l07 C F U , whereas the m u t a n t strains did not kill the mice even at 109 CFU. Antisera raised against the parent as well as the m u t a n t strains (M2 and M42) were used for the mouse protection test (Table 2). Mice were challenged with a fresh 5-h suspension of the parent strain, representing 1215 ×LDso. Very good protection was obtained with the homologous antiserum (95%). Only one mouse died, from which no bacteria could be recovered. A dilution of this antiserum ( 1 : 2 ) protected 100% of the mice. Antiserum raised against the mutant M2 only protected 10% of the mice, whereas anti-mutant M42 protected 60% of the mice. The parent strain 89-1591 was isolated from the heart blood o f all mice that died after challenge. After SDS-PAGE of cell envelopes, all strains presented several bands which clearly stained with Coomassie Brilliant Blue. Only slight differences were
67
T W O S. SUIS C A P S U L A R T Y P E 2 M U T A N T S
E
E
o_
4 ~67KD
.q45 KD
" 9 1 30KD Fig. 3. Western blotting of Streptococcus suis parent strain (89-1591 ) a n d m u t a n t s M 1, M2 a n d M42, using a n t i s e r u m raised against the p a r e n t strain. P: p a r e n t strain; M 1: m u t a n t M 1; M2: m u t a n t M2; M42: m u t a n t M42. TABLE3 Summary of differences observed among the Streptococcus suis parent strain (89-1591 ), mutants M42 and M2 Characteristic
Parent strain
Mutant M42
Mutant M2
Presence of capsule Sialic acid content ( a M / m g ) 2 Capsular type 2 antibodies 3 Virulence for mice 44 kDa protein 44 kDa protein antibodies 3 Protection
+
+1
m
3.05 X 10- 2
2.2)< 10 -2
+ (titer:64
+ (titer:32)
+ + + +
+
~The layer of the capsular material was more irregular than that of the parent strain. 2#M sialic acid/bacterial weight in mg. 3production of antibodies in rabbits.
observed among the high molecular weight bands. One specific band of approximately 44 kDa was present in the parent but not in mutant strains. Western blot analysis showed that this band was recognized in the parent strain by homologous whole cell antiserum. Using the same antiserum, this band was not observed in mutant strains (Fig. 3). A summary of the differences observed between the parent strain and both mutants is shown in Table 3.
68
M. GOTTSCHALK ET AL.
DISCUSSION
The virulence factors ofS. suis are not well known. Surface components are often believed to be important virulence determinants. Obtaining mutants defective in certain of these structures could offer an opportunity to examine the role of these components more specifically. Arends and Zanen ( 1988 ) suggested that S. suis capsular type 2 has special invasive properties due to the presence of a capsule which contains sialic acid. Other studies with E. coli and S. agalactiae identified the capsule as a major virulence determinant (Schiffer et al., 1976; Shigeoka et al., 1983 ). However, Clifton-Hadley et al. ( 1986 ) indicated that the pathogenicity of different isolates of S. suis capsular type 2 varies independently on the presence or the absence of the capsule. Other non-capsulated S. suis have been reported in the past (Elliott, 1966; Pedersen et al., 1984), but they were directly isolated from animals, and defined as non-capsulated based on a similar S. suis biochemical profile and on their lack of reaction with capsular types 1 and 2 antisera. One of these isolates (A 227), reported at first as non-capsulated (Elliott, 1966 ), was later proposed as a provisional capsular type 3 (Elliott et al., 1977 ), and finally as a capsular type 6 (Perch et al., 1983). With the methodology used in the present study, a non-capsulated and completely avirulent mutant M2 was obtained. The absence of capsular material was confirmed by immunoelectron microscopy, and the virulence was assessed in a mouse model, which had already been demonstrated to be a suitable model for studies on the pathogenesis ofS. suis capsular type 2 (Holt et al., 1989; Williams et al., 1988; Williams, 1990). Although mutant M2 was devoid of capsule, antibodies against the group antigen were not produced when it was inoculated into rabbits. The experiments conducted in the present study demonstrated that group antigen was not exposed at the cell surface, which could explain the necessity for sonicating cells for identification (Perch et al., 1983). Immunoelectrophoretic studies carried out by Elliott et al. ( 1977 ) had already suggested that, in S. suis strains, most of the Lancefield group D antigen could be found as lipid-bound teichoic acid. This lipoteichoic acid, when exposed at the cell surface, has been implicated in the adherence of other streptococci to host tissues (Beachey, 1975; Nealon and Mattingly, 1984 ). The presence of the capsule does not appear to be the sole determinant of virulence, since mutant M42, which was capsulated and typeable with capsular type 2 antiserum, was avirulent. The quantity of sialic acid present in the capsule of this m u t a n t was similar to that of the highly virulent parent strain. The unique difference which could be demonstrated between both strains, was the absence of a 44 kDa cell wall protein. Williams et al. (1990) proposed that phagocytosed capsulated, non-pathogenic isolates were killed whereas intracellular pathogenic organisms survived and replicated within
TWO S. SUIS CAPSULAR TYPE 2 MUTANTS
69
macrophages in the absence ofS. suis capsular type 2 antiserum and complement. The presence of the cell wall protein could possibly interfere with the intracellular destruction of pathogenic isolates. As the capsular material in mutant M42 presented a rather irregular disposition and seemed to be more loosely attached to the cell wall, as shown by immunoelectron microscopy, a possible role of the protein in capsule linkage should not be ruled out. In the present study, no differences were found among the high molecular weight proteins, as previously described by Vecht et al. ( 1989 ). In our study, antibodies against the capsular material and against the 44 kDa protein were needed to ensure complete protection. Holt et al. (1989) suggested that antibodies against a similar protein were present in protective sera, but not in non-protective sera. They postulated that, since the capsular polysaccharide is poorly immunogenic, the presence of proteins could increase the antibody response against the capsule and enhance protection. However, in the present study, it was demonstrated that the concentration of antibodies against the capsular material was not responsible for the difference of protection observed between the serum raised against the mutant M42 and that produced with the parent strain. These results could indicate that the 44 kDa protein seems to be an important immunogen on its own. In fact, Holt et al. (1990) have recently demonstrated that, despite the presence of antibodies against the capsule in both cases, a strong protection was achieved after the inoculation of formalin-killed, but not heat-killed, S. suis cells, indicating that heat-sensitive antigens, such as proteins, are also responsible for protection. In the present study, no protection was obtained when there were no antibodies against the capsule and the 44 kDa protein, as was the case with serum raised against m u t a n t M2. In conclusion, a 44 kDa protein seems to act as a virulence factor ofS. suis type 2, and the presence of antibodies against this protein appears to be necessary to obtain complete protection against the disease. ACKNOWLEDGEMENTS
We acknowledge the invaluable technical assistance of Bernadette Foiry, Sophie Radacovici and Lyne Pelletier. We also thank Charles Dozois for reviewing the manuscript. This work was supported in part by the Fonds pour la Formation de chercheurs et l'aide/l la recherche (grant 91-EQ-4101 ) and the Minist~re de l'Enseignement Sup6rieur et de la Science.
REFERENCES Aminoff, D., 1961. Methods for the quantitative estimation of N-acetylneuraminic acid and their application to hydrolysates of sialomucoids. Biochem. J., 87: 384-392.
70
M. G O T T S C H A L K ET AL.
Arends, J. and Zanen, H., 1988. Meningitis caused by Streptococcus suis in humans. Rev. Infect. Dis., 10: 131-137. Beachey, E., 1975. Binding of group A streptococci to human oral mucosal cells by lipoteichoic acid. Trans. Assoc. Am. Physicians, 88: 285-292. Clifton-Hadley, F., Alexander, T. and Enright, M., 1986. The epidemiology, diagnosis, treatment and control of Streptococcus suis type 2 infection. Proc. Am. Assoc. Swine Pract., Minneapolis, Minnesota, pp. 473-491. Devriese, L.A., Sustronk, B., Maenhout, T. and Haesebrouck, F., 1990. Streptococcus suis in a horse. Vet. Rec., 127:68. Elliott, S.D., 1966. Streptococcal infections in young pigs. I. An immunochemical study of the causative agent (PM Streptococcus), J. Hyg., 64:205-212. Elliott, S.D. and Tai, J., 1978. The type specific polysaccharides of Streptococcus suis. J. Exp. Med., 148: 1699-1704. Elliott, S.D., McCarty, M. and Lancefield, R., 1977. Teichoic acids of group D streptococci with special reference to strains form pig meningitis (Streptococcus suis). J. Exp. Med., 145: 490499. Gogoloswski, R.P., Cook, R.W. and O'Connell, C.J., 1990. Streptococcus suis serotypes associated with disease in weaned pigs. Aust. Vet. J., 67: 202-204. Gottschalk, M., Higgins, R., Jacques, M., Mittal, K. and Henrichsen, J., 1989. Description of 14 new capsular types of Streptococcus suis. J. Clin. Microbiol., 27: 2633-2636. Gottschalk, M., Lebrun, A., Jacques, M. and Higgins, R., 1990. Hemagglutination properties of Streptococcus suis. J. Clin. Microbiol., 28:2156-2158. Higgins, R. and Gottschalk, M., 1990. An update on Streptococcus suis identification. J. Vet. Diagn. Invest., 2: 249-252. Higgins, R., Gottschalk, M., Fecteau, G., Sauvageau, R., De Guise, S., Du Tremblay, D., 1990a. Isolation of Streptococcus suis from cattle. Can. Vet. J., 31 : 529. Higgins, R., Gottschalk, M., Mittal, K. and Beaudoin, M., 1990b. Streptococcus suis in swine. A sixteen month study. Can. J. Vet. Res., 54: 170-173. Holt, M., Enright, M. and Alexander, T., 1989. Studies of the protective effect of different fractions of sera from pigs immune to Streptococcus suis type 2 infections. J. Comp. Pathol., 100: 435-442. Holt, M., Enright, M. and Alexander, T., 1990. Immunization of pigs with killed cultures of Streptococcus suis type 2. Res. Vet. Sci., 48: 23-27. Jacques, M., Gottschalk, M., Foiry, B. and Higgins, R., 1990. Ultrastructural study of surface components of Streptococcus suis. J. Bacteriol., 172: 2833-2838. Kilpper-B~ltz, R. and Schleifer, K., 1987. Streptococcus suis sp. nov., nom. rev. Int. J. Syst. Bacteriol., 37: 160-162. Nealon, T. and Mattingly, S., 1984. Role of cellular lipoteichoic acids in mediating adherence ofserotype III strains of group B streptococci to human embryonic, fetal, and adult epithelial cells. Infect. Immun., 43: 523-530. Pedersen, K., Henrichsen, J. and Perch, B., 1984. The bacteriology of endocarditis in slaughter pigs. Acta Path. Microbiol. lmmunol. Scand. Sect. B, 92: 237-238. Perch, B., Pedersen, K. and Henrichsen, J., 1983. Serology of capsulated streptococci pathogenic for pigs: six new serotypes of Streptococcus suis. J. Clin. Microbiol., 17: 993-996. Schiffcr, M., Oliveira, E., Glode, M., McCraken, G., Sarfl; J. and Robbins, J., 1976. A review: relation between invasiveness and the K1 capsular polysaccharide of Escherichia coli. Pediatr. Res., 10: 82-87. Shigeoka, A., Rote, N., Santos, J. and Hill, H., 1983. Assessment of the virulence factors of Group B streptococci: correlation with sialic acid content. J. Infect. Dis., 147: 857-863. Towbin, H., Staehelin, T. and Gordon, J., 1979. Electrophoretic transfer of proteins from poly-
TWO S. SUIS CAPSULAR TYPE 2 MUTANTS
71
acrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci., 79: 4350-4354. Vecht, U., van Leengoed, L.A. and Verheijen, E., 1985. Streptococcus suis infections in pigs in the Netherlands (Part 1 ). Vet. Q., 7: 315-321. Vecht, U., Arends, J., van der Molen, E. and van Leengoed, L.A., 1989. Differences in virulence between two strains of Streptococcus suis type 2 after experimentally induced infection of newborn germ-free pigs. Am. J. Vet. Res., 50: 1037-1043. Whiley, R.A., Hardie, J.M. and Jackman, J.H., 1981. SDS-polyacrylamide gel electrophoresis of oral streptococci. In: Basic Concepts of Streptococci and Streptococcal Diseases. Reedbooks Ltd. Fox Lane North, Chertsey, Surrey. pp. 61-62. Williams, A., 1990. Relationship between intracellular survival in macrophages and pathogenicity of Streptococcus suis type 2 isolates. Microb. Pathog., 8: 189-196. Williams, A., Blakemore, W. and Alexander, T., 1988. A murine model of Streptococcus suis type 2 meningitis in the pig. Res. Vet. Sci., 45: 394-399. Yeung, M. and Mattingly, S., 1983. Isolation and characterization of type III group B streptococcal mutants defective in biosynthesis of the type specific antigen. Infect. Immun., 42: 141-151.
59
Production and characterization of two Streptococcus suis capsular type 2 mutants M. Gottschalk, R. Higgins, M. Jacques and D. Dubreuil Groupe de Recherche sur les Maladies In~,ctieuses du Porc. Facult~ de mOdecine vOtkrinaire, Universitk de MontrOal. C.P. 5000. St-Hyacinthe, Qug'bec, J2S 7C6, Canada (Accepted 15 June 1991 )
ABSTRACT Gottschalk, M., Higgins, R., Jacques, M. and Dubreuil, D., 1992. Production and characterization of two Streptococcus suis capsular type 2 mutants. Vet. Microbiol., 30:59-71. Two avirulent mutants of Streptococcus suis capsular type 2 (M2 and M42 ) were produced from a highly virulent strain. Mutant M2, obtained after serial subcultures of the parent strain in the presence of rabbit anti-capsular type 2 serum, no longer possessed the type-specific capsular antigen, as demonstrated by serotyping methods and immunoelectron microscopy. The Lancefield group D antigen could not be detected on the cell surface of this mutant using the immunogold labelling technique. SDS-PAGE of lysozyme treated cells demonstrated that a 44 kDa protein which was present in the parent strain, was absent in mutant M2. Immunoblotting using rabbit whole cell homologous antiserum revealed that the protein was strongly immunogenic. Mutant M2 was totally avirulent in mice, and the homologous antiserum completely failed to protect mice against challenge with the parent strain. However, mutant M42, obtained after passages of the parent strain at 42 °C, remained capsulated but lacked the same 44 kDa protein as mutant M2. The quantity of sialic acid present in the capsule was similar to that of the parent strain. Despite the presence of antibodies against the capsule, antiserum prepared against M42 only partially protected mice against a challenge with the parent strain. The 44 kDa cell wall protein could act as a virulence factor as well as an important immunogen of S. suis capsular type 2.
INTRODUCTION
Streptococcus suis is responsible for a wide range of infections in swine (Higgins et al., 1990b; Vecht et al., 1985 ) and other animal species (Devriese et al., 1990; Higgins et al., 1990a). It shares antigens with Lancefield group D streptococci, from which it appears to be genetically different (Kilpper-B~iltz and Schleifer, 1987). So far, there are 23 identified capsular types of S. suis (Gottschalk et al., 1989). With the exception of the Scandinavian countries and Australia, where the capsular types 7 and 9, respectively, are the most prevalent (Gogolewski et al., 1990; Perch et al., 1983 ), most other countries report that capsular type 2 ofS. suis is more frequently isolated from diseased 0 3 7 8 - 1 1 3 5 / 9 2 / $ 0 5 . 0 0 © 1992 Elsevier Science Publishers B.V. All rights reserved.
60
M. GOTTSCHALK ET AL.
pigs than any other capsular type, and it is mainly associated with septicemia, meningitis and pneumonia (Clifton-Hadley, 1986; Higgins et al., 1990b). Moreover, this capsular type has repeatedly been isolated from cases of septicemia and meningitis in m a n as a result of contact with pigs or pig products (Arends and Zanen, 1988 ). Virulence factors of S. suis have not been identified, and attempts to control the infection in the field by vaccination have given equivocal results. The presence of sialic acid as a component of the capsular material of S. suis capsular types 1 and 2 (Elliott and Tai, 1978) suggests, as is the case for Escherichia coli K 1, Neisseria meningitidis and Streptococcus agalactiae, that these sialic acid-containing polysaccharides might play an important role in protecting microorganisms from the host defense system by preventing the activation of the alternative pathway of complement (Shigeoka et al., 1983). However, pathogenic and non-pathogenic S. suis capsular type 2 strains were described as phagocytosed at a similar rate in the absence of type-specific antibodies, suggesting that the capsule might not play an evident antiphagocytic role and that other virulence factors must be present to inhibit intracellular killing of pathogenic strains (Williams, 1990). Fimbriae and hemagglutinins were also described, but their exact function in S. suis infection remains unclear (Gottschalk et al., 1990; Jacques et al., 1990). In addition, cell wall "M-like" proteins have been described in S. suis capsular type 2 isolates recovered from cases of meningitis, but not in isolates from cases of pneumonia or from clinically healthy animals (Vecht et al., 1989). Antibodies against certain cell wall proteins were present in S. suis capsular type 2 hyperimmune protective serum, but absent in pre-immune non-protective serum (Holt et al., 1989). The study of mutants defective in surface components could help to identify some of the virulence factors of S. suis. In the present study, two S. suis capsular type 2 mutants were obtained from a highly virulent strain, and their characteristics compared to those of the parent strain. MATERIALS AND METHODS
Production of mutants S. suis capsular type 2 strain 89-1591, originally isolated from a case of septicemia and meningitis in a pig, was streaked onto blood agar plates (tryptic soy agar containing 5% bovine blood). For the production of mutants M 1 and M2 the following technique was used (Yeung and Mattingly, 1983): a single colony was chosen and kept as the parent strain. Cells from the same colony were inoculated into 5 ml Todd-Hewitt broth to which 1 ml of rabbitanti-capsular type 2 antiserum was added. A Todd-Hewitt broth containing the same proportion of normal rabbit serum was used as a control. After 6 h of incubation at 37 ° C, a solid clump at the bottom of the tube containing the
TWO S. SUIS CAPSULAR TYPE 2 MUTANTS
61
capsular type 2 antiserum was observed. After gentle agitation, 10 #1 from the non-agglutinated phase of the culture were transferred into 5 ml of ToddHewitt broth supplemented with 200/tl of anti-capsular type 2 antiserum, and incubated for 18 h; the procedure was repeated several times. After each transfer, cells from the non-agglutinated phase were streaked onto a blood agar plate, and five colonies from each passage were then tested by three different capsular typing methods (described below). Mutant M 1 was obtained after 16 passages whereas m u t a n t M2 was obtained after 33 passages. As both mutants presented exactly the same characteristics, only mutant M2 was studied further. Mutant M42 was obtained after 50 subcultures on tryptic soy agar at 42°C. The parent strain as well as the mutant strains were tested for their biochemical and enzymic profiles using conventional tests (Higgins and Gottschalk, 1990) and a multitest system (Rapid-Strep, API System, Laboratory Products Ltd, St-Laurent, Qu6., Canada.
Capsular typing methods Capsular typing was based on three different techniques: capsular reaction (Neufeld), capillary precipitation and coagglutination tests, as described earlier (Higgins and Gottschalk, 1990). The Lancefield group D antigen was detected in sonicated bacterial suspensions using a coagglutination reagent (Phadebact, Pharmacia Diagnostic, Uppsala, Sweden).
Electron microscopy Preparation for transmission electron microscopy was carried out as described previously (Jacques et al., 1990). When present, capsular material was stabilized with serotype-specific antiserum and stained with ruthenium red. Controls without stabilization or treated with normal rabbit serum were performed simultaneously. Thin sections were examined with an electron microscope (Philips 201 ) at an accelerating voltage of 60 kV. Dense bacterial suspensions were also examined after negative staining (Jacques et al., 1990). A drop of each preparation was placed on a 200-mesh Formvar-coated electron microscope grid and blotted until partially dry. A drop of 2% (w/v) phosphotungstate (pH 7.0) was then applied to the grids, which were observed with the electron microscope. For the localization of the cell wall antigens, grids with bacteria were treated with rabbit anti-Lancefield group D antiserum (Wellcome, Beckenham, England). After washing, gold labeled goat anti-rabbit IgG (5 n m gold colloidal particles, E.Y. Labs, Inc. Biochemical Division, San Mateo, CA) was added. Grids were observed as described above. A strain of Enterococcus faecalis (ATCC 19433) was used as positive control.
62
M. G O T T S C H A E K ET AL.
Quantitation of sialic acid levels Sialic acid was extracted from freeze-dried cells by acid and heat treatments as described previously (Shigeoka et al., 1983 ). Sialic acid levels were determined by the thiobarbituric acid method of Aminoff ( 1961 ).
Production of antisera Production of antisera against the reference strain of S. suis capsular type 2 (NCTC 10234), the parent strain (89-1541), and mutants M2 and M42 was performed as previously described (Higgins and Gottschalk, 1990 ). Capsular reaction and slide agglutination tests were used for the titration of antisera.
Mouse lethality test A virulence test was performed by intraperitoneal injection of 4-week-old mice (strain CF1 ) (Williams et al., 1988; Gottschalk et al., 1990). The 50% lethal dose (LDso) was estimated with groups of five mice. Bacteria were grown in Todd-Hewitt broth (Difco Laboratories, Detroit, Mi), harvested in log-phase growth (5-6 h), suspended in phosphate-buffered saline and adjusted to an optical density of 0.8 (540 n m ) which corresponded to a concentration of approximately 109 C F U / m l . The exact number of viable cells present was counted by a pour plate method, using blood-agar plates incubated for 18 h at 37°C. Death or presence of clinical signs, such as meningitis, was monitored over the next 7 d. The test was repeated four times for each strain.
Mouse protection test For the mouse protection test, five groups of twenty 4-week-old mice (strain CF1 ) were inoculated with a 5-6 h culture of the parent strain mixed with normal rabbit serum (control), or antiserum raised against the parent strain (non-diluted and diluted 1/2), the M2 strain and the M42 strain. In each experiment, the LDso of the inoculum was measured. Colony forming units were determined by a pour plate m e t h o d immediately after animal injections to monitor the exact numbers of viable bacteria inoculated. In this way it was estimated that each inoculum used for the intraperitoneal challenge of mice, contained 12-15 times the LDs0. Dead mice were counted and removed from their cages at 24 h intervals. A culture from heart blood was made onto blood agar plates from approximately 50% of mice which died after challenge in each group, and bacterial identity was confirmed by the capsular reaction test.
SDS-polyacrylamide gel electrophoresis The parent strain and the mutant M2 were grown in 7 ml of Todd-Hewitt broth at 37°C to early stationary phase (6-7 h). In all experiments, mutant M42 was incubated at 42°C. Each inoculum was transferred into 200 ml of Todd-Hewitt broth, incubated for 18 h, collected by centrifugation, washed
T W O X SUIS C A P S U L A R T Y P E 2 M U T A N T S
63
and lysed by sonication. Cell envelopes were obtained after centrifugation (40 000 g, 30 rain), and treated with lysozyme (5 m g / m l ) for 4 h at 37°C. Vertical slab gels of 12% polyacrylamide were employed with a 4.5% stacking gel as previously described (Whiley et al., 1981 ).
Western blotting After SDS-PAGE, separated material was transferred from the slab gel to nitrocellulose membrane by the methanol-Tris-glycine system (Towbin et al., 1979). Electroblotting was performed in a transblot apparatus (Hoefer Scientific Instruments, San Francisco, California) for 18 h at 60 V. After blocking unreacted sites with 2% (w/v) casein solution, the nitrocellulose was incubated for 2 h with the appropriate dilution ofantisera in the same solution. After washing in Tris-NaC1, the sheets were incubated with a peroxidase-labelled IgG fraction of goat antiserum raised against rabbit IgG (Daymar Laboratories, Toronto, Ontario) for 90 min at a dilution of 1 : 500 in a 2% solution of caseine in 10 m M Tris-NaC1. After repeated washing, the presence of bound antigens was visualized by reacting the nitrocellulose membrane with 0.06% 4-chloro-l-naphtol (Sigma Chemical Company, St-Louis, MO) in cold methanol mixed to 0.02% H202 in Tris-NaC1. Apparent molecular weights were calculated by comparison with standards of known molecular weight. RESULTS Conventional tests and the Rapid-Strep system showed that the parent as well as both mutants presented the same biochemical profle. They were identified as typical S. suis. The three strains coagglutinated, after sonication, with group D antiserum. Mutant M42 still showed a typical reaction with anti-capsular type 2 reference serum, whereas mutant M2 was untypeable with serotyping methods. Mutant M2 was confirmed to be non-capsulated by immunoelectron microscopy (Fig. 1 ). Mutant M42 was capsulated, but the capsular material distribution on the cell surface was rather irregular (Fig. 1 ). When the content of sialic acid was evaluated, the parent strain and mutant M42 presented similar values (3.05 X 10 -2 and 2.2 X 10-2/2M/mg of bacteria, respectively). All mutants possessed fimbriae which were morphologically identical to those of the parent strain (data not shown ). When antisera were produced against the parent strain and mutants M2 and M42, anti-capsular type 2 antibodies were present in antisera raised against the parent strain (titer:64) and mutant M42 (titer:32), but not in antiserum raised against mutant M2 (Table 1 ). Despite the absence of the capsule, mutant M2 did not produce antibodies, detectable by the slide agglutination test, against the group antigen, as a negative reaction could be observed with the E. faecalis strain (Table 1 ). When whole cells were treated
64
M. G O T T S C H A L K ET AL.
(2
Fig. 1. Transmission electron micrographs of thin sections of cells of Streptococcus suis exposed to anti-capsular type 2 serum and stained with ruthenium red. A: parent strain (89-1591); B: mutant M2; C: mutant M42. Bar: 200 nm. w i t h specific r a b b i t a n t i - L a n c e f i e l d g r o u p D I g s , a n d t h e n w i t h g o a t a n t i - r a b b i t colloidal g o l d - c o u p l e d Igs a n d o b s e r v e d u n d e r the e l e c t r o n m i c r o s c o p e , a few gold p a r t i c l e s c o u l d b e o b s e r v e d o n this m u t a n t as well as o n the well
65
TWO S. SUIS C A P S U L A R TYPE 2 M U T A N T S
iii
O°
7 Fig. 2. Negative staining of bacterial cells incubated with rabbit Lancefield group D antiserum and colloidal gold labeled sheep anti-rabbit immunoglobulin G. A: Enterococcus faecalis; B: Streptococcus suis parent strain (89-1591 ); C: mutant M2. capsulated p ar en t strain (Fig. 2). In contrast, anti-group D a n t i b o d y labeled with colloidal gold was f o u n d to bi nd ove r the entire surface o f E . faecalis. Both m u t a n t s were non-virulent for mice. T he LDso for the parent strain
66
M. GOTTSCHALK ET AL
TABLE 1 Serotyping results using antisera raised against tants M2 and M42.
Streptococcus
suis
parent strain (89-1591 ) and mu-
Antiserum raised against I
Antigen
Parent strain
Mutant M2
Mutant M42
Capsular type 2 (reference strain NCTC 10237) Parent strain (89-1591 ) Mutant M22 Mutant M42
+ (titer:64)
-
+ (titer:32)
+ -+
+ -
+ -+
E. ,faecalis 3
_
_
_
~The following tests were used: coagglutination, capillary precipitation and capsular reaction. 2Positive only with the coagglutination test. 3Only the coagglutination test was performed. TABLE 2 Protection of mice challenged with S t r e p t o c o c c u s heterologous antisera from immunized rabbits
suis
parent strain (89-1591) by homologous and
Serum used
No. deaths/No, tested ~
% of protection
Normal rabbit serum Antiserum raised against: Parent strain (non-diluted) (dilution 1:2) Mutant M2 Mutant M42
20/20
0
1/20 0/20 18/20 8/20
95 100 10 60
NOTE: The number of streptococci in each inoculum was equivalent to 12-15 times the LDso. IDeath occurring 72 hours after inoculation of mice.
was 3 ( + 1.4) × l07 C F U , whereas the m u t a n t strains did not kill the mice even at 109 CFU. Antisera raised against the parent as well as the m u t a n t strains (M2 and M42) were used for the mouse protection test (Table 2). Mice were challenged with a fresh 5-h suspension of the parent strain, representing 1215 ×LDso. Very good protection was obtained with the homologous antiserum (95%). Only one mouse died, from which no bacteria could be recovered. A dilution of this antiserum ( 1 : 2 ) protected 100% of the mice. Antiserum raised against the mutant M2 only protected 10% of the mice, whereas anti-mutant M42 protected 60% of the mice. The parent strain 89-1591 was isolated from the heart blood o f all mice that died after challenge. After SDS-PAGE of cell envelopes, all strains presented several bands which clearly stained with Coomassie Brilliant Blue. Only slight differences were
67
T W O S. SUIS C A P S U L A R T Y P E 2 M U T A N T S
E
E
o_
4 ~67KD
.q45 KD
" 9 1 30KD Fig. 3. Western blotting of Streptococcus suis parent strain (89-1591 ) a n d m u t a n t s M 1, M2 a n d M42, using a n t i s e r u m raised against the p a r e n t strain. P: p a r e n t strain; M 1: m u t a n t M 1; M2: m u t a n t M2; M42: m u t a n t M42. TABLE3 Summary of differences observed among the Streptococcus suis parent strain (89-1591 ), mutants M42 and M2 Characteristic
Parent strain
Mutant M42
Mutant M2
Presence of capsule Sialic acid content ( a M / m g ) 2 Capsular type 2 antibodies 3 Virulence for mice 44 kDa protein 44 kDa protein antibodies 3 Protection
+
+1
m
3.05 X 10- 2
2.2)< 10 -2
+ (titer:64
+ (titer:32)
+ + + +
+
~The layer of the capsular material was more irregular than that of the parent strain. 2#M sialic acid/bacterial weight in mg. 3production of antibodies in rabbits.
observed among the high molecular weight bands. One specific band of approximately 44 kDa was present in the parent but not in mutant strains. Western blot analysis showed that this band was recognized in the parent strain by homologous whole cell antiserum. Using the same antiserum, this band was not observed in mutant strains (Fig. 3). A summary of the differences observed between the parent strain and both mutants is shown in Table 3.
68
M. GOTTSCHALK ET AL.
DISCUSSION
The virulence factors ofS. suis are not well known. Surface components are often believed to be important virulence determinants. Obtaining mutants defective in certain of these structures could offer an opportunity to examine the role of these components more specifically. Arends and Zanen ( 1988 ) suggested that S. suis capsular type 2 has special invasive properties due to the presence of a capsule which contains sialic acid. Other studies with E. coli and S. agalactiae identified the capsule as a major virulence determinant (Schiffer et al., 1976; Shigeoka et al., 1983 ). However, Clifton-Hadley et al. ( 1986 ) indicated that the pathogenicity of different isolates of S. suis capsular type 2 varies independently on the presence or the absence of the capsule. Other non-capsulated S. suis have been reported in the past (Elliott, 1966; Pedersen et al., 1984), but they were directly isolated from animals, and defined as non-capsulated based on a similar S. suis biochemical profile and on their lack of reaction with capsular types 1 and 2 antisera. One of these isolates (A 227), reported at first as non-capsulated (Elliott, 1966 ), was later proposed as a provisional capsular type 3 (Elliott et al., 1977 ), and finally as a capsular type 6 (Perch et al., 1983). With the methodology used in the present study, a non-capsulated and completely avirulent mutant M2 was obtained. The absence of capsular material was confirmed by immunoelectron microscopy, and the virulence was assessed in a mouse model, which had already been demonstrated to be a suitable model for studies on the pathogenesis ofS. suis capsular type 2 (Holt et al., 1989; Williams et al., 1988; Williams, 1990). Although mutant M2 was devoid of capsule, antibodies against the group antigen were not produced when it was inoculated into rabbits. The experiments conducted in the present study demonstrated that group antigen was not exposed at the cell surface, which could explain the necessity for sonicating cells for identification (Perch et al., 1983). Immunoelectrophoretic studies carried out by Elliott et al. ( 1977 ) had already suggested that, in S. suis strains, most of the Lancefield group D antigen could be found as lipid-bound teichoic acid. This lipoteichoic acid, when exposed at the cell surface, has been implicated in the adherence of other streptococci to host tissues (Beachey, 1975; Nealon and Mattingly, 1984 ). The presence of the capsule does not appear to be the sole determinant of virulence, since mutant M42, which was capsulated and typeable with capsular type 2 antiserum, was avirulent. The quantity of sialic acid present in the capsule of this m u t a n t was similar to that of the highly virulent parent strain. The unique difference which could be demonstrated between both strains, was the absence of a 44 kDa cell wall protein. Williams et al. (1990) proposed that phagocytosed capsulated, non-pathogenic isolates were killed whereas intracellular pathogenic organisms survived and replicated within
TWO S. SUIS CAPSULAR TYPE 2 MUTANTS
69
macrophages in the absence ofS. suis capsular type 2 antiserum and complement. The presence of the cell wall protein could possibly interfere with the intracellular destruction of pathogenic isolates. As the capsular material in mutant M42 presented a rather irregular disposition and seemed to be more loosely attached to the cell wall, as shown by immunoelectron microscopy, a possible role of the protein in capsule linkage should not be ruled out. In the present study, no differences were found among the high molecular weight proteins, as previously described by Vecht et al. ( 1989 ). In our study, antibodies against the capsular material and against the 44 kDa protein were needed to ensure complete protection. Holt et al. (1989) suggested that antibodies against a similar protein were present in protective sera, but not in non-protective sera. They postulated that, since the capsular polysaccharide is poorly immunogenic, the presence of proteins could increase the antibody response against the capsule and enhance protection. However, in the present study, it was demonstrated that the concentration of antibodies against the capsular material was not responsible for the difference of protection observed between the serum raised against the mutant M42 and that produced with the parent strain. These results could indicate that the 44 kDa protein seems to be an important immunogen on its own. In fact, Holt et al. (1990) have recently demonstrated that, despite the presence of antibodies against the capsule in both cases, a strong protection was achieved after the inoculation of formalin-killed, but not heat-killed, S. suis cells, indicating that heat-sensitive antigens, such as proteins, are also responsible for protection. In the present study, no protection was obtained when there were no antibodies against the capsule and the 44 kDa protein, as was the case with serum raised against m u t a n t M2. In conclusion, a 44 kDa protein seems to act as a virulence factor ofS. suis type 2, and the presence of antibodies against this protein appears to be necessary to obtain complete protection against the disease. ACKNOWLEDGEMENTS
We acknowledge the invaluable technical assistance of Bernadette Foiry, Sophie Radacovici and Lyne Pelletier. We also thank Charles Dozois for reviewing the manuscript. This work was supported in part by the Fonds pour la Formation de chercheurs et l'aide/l la recherche (grant 91-EQ-4101 ) and the Minist~re de l'Enseignement Sup6rieur et de la Science.
REFERENCES Aminoff, D., 1961. Methods for the quantitative estimation of N-acetylneuraminic acid and their application to hydrolysates of sialomucoids. Biochem. J., 87: 384-392.
70
M. G O T T S C H A L K ET AL.
Arends, J. and Zanen, H., 1988. Meningitis caused by Streptococcus suis in humans. Rev. Infect. Dis., 10: 131-137. Beachey, E., 1975. Binding of group A streptococci to human oral mucosal cells by lipoteichoic acid. Trans. Assoc. Am. Physicians, 88: 285-292. Clifton-Hadley, F., Alexander, T. and Enright, M., 1986. The epidemiology, diagnosis, treatment and control of Streptococcus suis type 2 infection. Proc. Am. Assoc. Swine Pract., Minneapolis, Minnesota, pp. 473-491. Devriese, L.A., Sustronk, B., Maenhout, T. and Haesebrouck, F., 1990. Streptococcus suis in a horse. Vet. Rec., 127:68. Elliott, S.D., 1966. Streptococcal infections in young pigs. I. An immunochemical study of the causative agent (PM Streptococcus), J. Hyg., 64:205-212. Elliott, S.D. and Tai, J., 1978. The type specific polysaccharides of Streptococcus suis. J. Exp. Med., 148: 1699-1704. Elliott, S.D., McCarty, M. and Lancefield, R., 1977. Teichoic acids of group D streptococci with special reference to strains form pig meningitis (Streptococcus suis). J. Exp. Med., 145: 490499. Gogoloswski, R.P., Cook, R.W. and O'Connell, C.J., 1990. Streptococcus suis serotypes associated with disease in weaned pigs. Aust. Vet. J., 67: 202-204. Gottschalk, M., Higgins, R., Jacques, M., Mittal, K. and Henrichsen, J., 1989. Description of 14 new capsular types of Streptococcus suis. J. Clin. Microbiol., 27: 2633-2636. Gottschalk, M., Lebrun, A., Jacques, M. and Higgins, R., 1990. Hemagglutination properties of Streptococcus suis. J. Clin. Microbiol., 28:2156-2158. Higgins, R. and Gottschalk, M., 1990. An update on Streptococcus suis identification. J. Vet. Diagn. Invest., 2: 249-252. Higgins, R., Gottschalk, M., Fecteau, G., Sauvageau, R., De Guise, S., Du Tremblay, D., 1990a. Isolation of Streptococcus suis from cattle. Can. Vet. J., 31 : 529. Higgins, R., Gottschalk, M., Mittal, K. and Beaudoin, M., 1990b. Streptococcus suis in swine. A sixteen month study. Can. J. Vet. Res., 54: 170-173. Holt, M., Enright, M. and Alexander, T., 1989. Studies of the protective effect of different fractions of sera from pigs immune to Streptococcus suis type 2 infections. J. Comp. Pathol., 100: 435-442. Holt, M., Enright, M. and Alexander, T., 1990. Immunization of pigs with killed cultures of Streptococcus suis type 2. Res. Vet. Sci., 48: 23-27. Jacques, M., Gottschalk, M., Foiry, B. and Higgins, R., 1990. Ultrastructural study of surface components of Streptococcus suis. J. Bacteriol., 172: 2833-2838. Kilpper-B~ltz, R. and Schleifer, K., 1987. Streptococcus suis sp. nov., nom. rev. Int. J. Syst. Bacteriol., 37: 160-162. Nealon, T. and Mattingly, S., 1984. Role of cellular lipoteichoic acids in mediating adherence ofserotype III strains of group B streptococci to human embryonic, fetal, and adult epithelial cells. Infect. Immun., 43: 523-530. Pedersen, K., Henrichsen, J. and Perch, B., 1984. The bacteriology of endocarditis in slaughter pigs. Acta Path. Microbiol. lmmunol. Scand. Sect. B, 92: 237-238. Perch, B., Pedersen, K. and Henrichsen, J., 1983. Serology of capsulated streptococci pathogenic for pigs: six new serotypes of Streptococcus suis. J. Clin. Microbiol., 17: 993-996. Schiffcr, M., Oliveira, E., Glode, M., McCraken, G., Sarfl; J. and Robbins, J., 1976. A review: relation between invasiveness and the K1 capsular polysaccharide of Escherichia coli. Pediatr. Res., 10: 82-87. Shigeoka, A., Rote, N., Santos, J. and Hill, H., 1983. Assessment of the virulence factors of Group B streptococci: correlation with sialic acid content. J. Infect. Dis., 147: 857-863. Towbin, H., Staehelin, T. and Gordon, J., 1979. Electrophoretic transfer of proteins from poly-
TWO S. SUIS CAPSULAR TYPE 2 MUTANTS
71
acrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci., 79: 4350-4354. Vecht, U., van Leengoed, L.A. and Verheijen, E., 1985. Streptococcus suis infections in pigs in the Netherlands (Part 1 ). Vet. Q., 7: 315-321. Vecht, U., Arends, J., van der Molen, E. and van Leengoed, L.A., 1989. Differences in virulence between two strains of Streptococcus suis type 2 after experimentally induced infection of newborn germ-free pigs. Am. J. Vet. Res., 50: 1037-1043. Whiley, R.A., Hardie, J.M. and Jackman, J.H., 1981. SDS-polyacrylamide gel electrophoresis of oral streptococci. In: Basic Concepts of Streptococci and Streptococcal Diseases. Reedbooks Ltd. Fox Lane North, Chertsey, Surrey. pp. 61-62. Williams, A., 1990. Relationship between intracellular survival in macrophages and pathogenicity of Streptococcus suis type 2 isolates. Microb. Pathog., 8: 189-196. Williams, A., Blakemore, W. and Alexander, T., 1988. A murine model of Streptococcus suis type 2 meningitis in the pig. Res. Vet. Sci., 45: 394-399. Yeung, M. and Mattingly, S., 1983. Isolation and characterization of type III group B streptococcal mutants defective in biosynthesis of the type specific antigen. Infect. Immun., 42: 141-151.
