The Journal of
Medical Microbiology Vol. 11, No. 4 IMMU N O L O G I C A L RESPONSE TO A S T RAI N OF S T A P H Y L O C O C C U S E P I D E R M I D I S I N T H E RABBIT: P R O D U C T I O N O F PROTECTIVE ANTIBODY
K. YOSHIDA AND Y. ICHIMAN Department of Microbiology, St Marianna University School of Medicine, 2095 Sugao, Takatu-ku, Kawasaki 213, Japan
STXPHYLOCOCCUS EpIDERMIDIS strains have been regarded as being nonpathogenic for animals. Recently, however, a strain of S. epidermidis exhibiting mouse virulence related to antiphagocytic activity was isolated from a human source. Mice immunised with this strain showed significant resistance to homologous challenge which appeared 10 to 20 days after immunisation (Yoshida, Ichiman and Ohtomo, 1976). This phenomenon was regarded as being similar to that observed with the Smith diffuse strain of S. aureus in mice and rabbits (Yoshida and Ekstedt, 1968; Ekstedt and Yoshida, 1969). We have now attempted to find out whether the immunological properties of the strain and the nature of the protective antibody coincide with those observed with encapsulated strains of S. aureus. MATERIALS AND METHODS Strain. Strain 1142 of S. epidermidis, which was described elsewhere (Yoshida et al., 1976), was used. It gave negative reactions for coagulase, clumping factor, mannitol fermentation, deoxyribonuclease and acid phosphatase by the methods of Yoshida, Smith and Naito (1971). Animals. Mice of strain DD (Nihon Clea Co. Ltd, Tokyo) weighing c. 17 g and normal adult Flemish cross rabbits weighing c. 2 - 0 kg (Nihon Clea Co. Ltd, Tokyo, Japan) were used. Immunisation schedule. Strain 1142 was cultured overnight at 37°C in Brain Heart Infusion (BHI,Difco) broth, was washed once with saline (0.85% NaCl w/v in water) and resuspended in a similar solution. The turbidity of the cell suspension was nephelometrically adjusted to El -0 at 430 nm, and the suspension was then autoclaved at 121°C for 15 min. One ml of this cell suspensioncontained 8.35 x lo8colony forming units (c.f.u.) as determined by a surfaceviable-count method. The cell suspension was injected intravenouslyinto arabbit, 0.5 ml on 3 successive days. Ten weeks later, a single intravenous injection of a similar Received 2 May 1977; revised version accepted 2 Nov. 1977. 1. MED. MICROBI0L.-VOL.
11 (1978)
371
2 B
K. YOSHIDA AND Y . ICHIMAN
372
dose of vaccine was made. After the initial immunisation, 5 ml of blood were withdrawn weekly for 16 weeks and the serum was separated. Procedurefor the examination of mouse virulence. A similar but unheated suspension of strain 1142 was injected intraperitoneally in 0.5-ml amounts into each of five mice. Deaths were recorded for 4 weeks after challenge. Passive protection tests. Rabbit antisera were serially diluted two-fold with sterile saline and 0.5-ml amounts of the dilutions were injected intraperitoneally into groups of five mice. Thirty minutes later, the mice were each challenged intraperitoneally with 0.5 mi of the cell suspension containing c. 8 x lo8c.f.u. of strain 1142 as above, and the number of deaths was recorded for 4 weeks. The mouse-passive-protective activities of IgM- and IgG-rich fractions, obtained by sucrose-gradient ultracentrifugation as described by Yoshida and Ekstedt (1968) were determined similarly. Sensitised haemagglutinationtest. Ten mg of the surface polysaccharide antigen extracted from strain 1142 by the method of Yoshida et al. (1976), were dissolved in 10 ml of 0 . 0 6 7 ~phosphate-buffered saline (PBS), p H 6.5; 0.25 ml of sheep RBC was added to this and the mixture kept at 37°C for 2 h. The cells were washed three times with PBS, and a 0.25% (v/v) suspension of them was prepared. Rabbit antisera, which had been absorbed with similar volumes of packed sheep RBC, were diluted 1 in 10 and then serially diluted in two-fold steps with PBS. With these sera, haemagglutination titres were determined as previously described (Yoshida, Mizunari and Hattori, 1964). Mercaptoethanol treatment and IgG absorption of the sera. Rabbit antisera were treated with 0 - 2 2-mercaptoethanol ~ (2-ME) and iodoacetic acid (Yoshida and Ekstedt, 1968). Also, to remove the IgG from rabbit antisera, 1 .O ml of anti-rabbit IgG goat serum (Hyland) was added to an equal volume of the rabbit antiserum; after 2 h at 37°C the mixture was centrifuged at 7000 g for 15 min. and the supernate separated. Sucrose-density-gradientfractionation of sera. The serum sample was centrifuged through 5 - 0ml of 5 2 0 % sucrose gradient prepared in PBS at 86 OOO g for 18 h in an ultracentrifuge (Model 55P-2, Hitachi Co. Ltd, Tokyo, Japan) with an RPS50 head. Fractions of c. 0.2 ml
10 240
co
I
n
FIG.1 .-Haemagglutination titres of serum from rabbit immunised with strain 1142 of Staphylococcus epidermidis. Symbols: solid line, untreated; broken line, mercaptoethanol-treated serum. Arrows indicate times of injections.
PROTECTIVE ANTIBOD Y AGAINST STMH. EPIDERMIDIS
373
were then collected and diluted to 1 -0ml with 10% sucrose; the optical densities were read at 280 nm by means of a spectrophotometer (Model UV-200, Shimazu Co. Ltd, Kyoto, Japan).
RESULTS Antibody response to immunisation with S. epidermidis One week after the initial immunisation with heat-killed vaccine of strain 1142, the sensitised haemagglutination (HA) titre of the untreated serum reached 640. It continued at this level for 4 weeks and then gradually declined until the 10th week (fig. 1). When a booster injection was given at the 10th week, the titre rose rapidly to 10240 at the 11th week but had decreased to 640 at the 13th week and remained at this level until the 16th week when the examination was terminated. In sera that had been treated with 2-ME, low HA activity appeared as early as the 3rd week, but the titre remained constant at 20 between the 4th and 10th weeks. After the booster injection, however, the HA titre increased to 160 and remained at this level from the 11th to the 13th week, but had fallen to 20 by the 15th week. Rabbit antiserum showed passive-protective activity in mice as early as the 1st week after vaccination (table I), and the " primary " response was greatest at the 2nd week. It tended to decrease progressively after the 4th week, and was very weak by the 6th week. After the booster injection, the highest activity was shown in the 11th and 12th weeks; then there was a rapid decline, and no activity was shown at the 16th week. A comparison of the HA titres and the relative passive-protective activity of the sera suggested that protection was associated with 2-ME-sensitive antibody. TABLE I Mouse passive-protective activity of rabbit antisera taken at weekly intervals after irnmunisation with S. epidermidis strain 1142 Weeks after immunisation
O*
1 2 3 4 6 8 lot 11 12 13 14 16
HA titre
0 640 640 640 640 320 160 160 10 240 2 560 640 640 640
Number of mice deadlnumber challenged after receiving serum diluted 1 in '5
10
515 015 015 015
115
115
3:
215 415
...
015
115 215
3;
515 515
20
...
415 315 415 315 415 315
... ...
* Control. t Booster injection given.
40
so
...
... ... ...
515 415 515 515
515
... ...
...
3; ...
... ... .-.
...
...
K . YOSHIDA AND Y. ICHIMAN
374
Nature of protective antibody to S. epidermidis Experiments were therefore carried out to confirm that passive-protective activity was associated with IgM antibody. As shown in table 11, it was 2-ME sensitive and was not absorbed from the serum by rabbit anti-IgG goat serum. To elucidate further the nature of the protective activity, rabbit-antiserum fractions obtained by sucrose-density-gradient ultracentrifugation were examined for passive protective activity in the mouse (table 111). The results showed that 280 pg of bottom fraction, rich in IgM, was capable of passively protecting mice against homologous challenge. More than 910 pg of the upper fraction, rich both in IgA and IgG, had no such activity. Also, most of the HA activity was present in the bottom fraction (fig. 2).
DISCUSSION In a previous contribution (Yoshida et al., 1976) it was reported that strain 1142 of S. epidermidis, which was used in the present experiments, possessed a protective antigen on the cell surface and that this was associated with antiTABLE I1 Eflect of mercaptoethanol(2-ME) and IgG absorption on passive-protective activity of hyperimmune rabbit antiserum in mice Number of mice deadlnumber challenged
Serum given (and treatment)* I
Hyperimmune (2-ME) Hyperimmune (absorption with anti-rabbit IgG) Hyperimmune (untreated) Normal rabbit (untreated)
* The hyperimmune serum had been collected in the 11th week after immunisation. It was given intraperitoneally in a dose of 0.5 ml per mouse. TABLE I11 Mouse passive-protective activity of immunoglobulins separated by sucrose-density-gradient ultracentrifugationof serum taken from rabbit immunised with strain 1142 Passive immunisation of normal mice with IgM-rich fraction with IgG-rich fraction
I
Number of mice dead/number challenged
{E:i { 'E;i
normal rabbit serum (control)
51s
375
PROTECTIVE ANTIBODY AGAINST STAPH. EPIDERMIDIS
"Ol
-16
-8
Q) c,
.-L
c,
< I
I' I
HA titre 8
2
b
4
I
6
HA titre I
8
I
I
I
10 12 14 Tube number
I
16
1
18
6
20
.22
FIG.2.-Haemagglutination (HA) titres of fractions obtained by sucrose-density ultracentrifugation from a rabbit antiserum collected at the 11th week after immunisation.
phagocytic activity. Further, mice immunised with this organism showed significant resistance to challenge with it during the ensuing 10-20 days. These phenomena were similar to those observed with the Smith diffuse strain of S. aureus by Yoshida and Ekstedt (1968). In the experiments now recorded, the production of antibody in the rabbit against the surface polysaccharide antigen was monitored by means of the HA test. In the primary response, most of the HA was 2-ME sensitive, but greater amounts of 2-ME-resistant HA antibody were detected in antisera obtained after a booster injection. The HA titres were comparable with the relative passive-protective activity of sera in mice. Further, protective activity was removed from antiserum by 2-ME treatment but not by absorption with antiIgG serum. Also, after fractionation by sucrose-density-gradient ultracentrifugation of an antiserum that showed the highest titre in the HA test and the greatest activity in the protection test in the course of the secondary response, passive-protective activity, and most of the HA activity, was detected in the IgM-rich fraction, which would have contained only minor amounts of IgA. However, no passive protective activity was found in the IgG-rich fraction,
376
K. YOSHIDA AND Y. ICHIMAN
which would have contained a greater amount of IgA than the IgM-rich fraction. These results indicate that the protective antibody would positively be IgM-globulin in nature. Yoshida et al. (1976) observed that active immunisation of mice with 10 pg of the surface polysaccharide gave protection against homologous challenge, and that the passive-protective activity of rabbit antiserum was completely removed by absorption with this antigen. Thus, the protective antibody in rabbit antiserum may be regarded as IgM formed in response to the surface antigen. With regard to the protection-inducing antigen in S. aureus strains, this has been regarded as being present only in encapsulated strains (Morse, 1962 ; Fisher, Delvin and Erlandson, 1963;Rogers, 1962).The biological and immunological characters of the protection-inducing polysaccharide of S. epidermidis strain 1142 suggest that this is a capsular antigen. The occurrence and distribution of strains with similar properties among isolates of S. epidermidis are at present under investigation in our laboratory. SUMMARY
A rabbit was immunised with a heat-killed vaccine prepared from strain 1142 of Staphylococcus epidermidis. The HA titres of the antisera against a surface-polysaccharide antigen extracted from the homologous strain were highest in the 2nd-4th weeks after immunisation and subsequently declined. Only a minor amount of 2-ME-resistant antibody was formed, and this at a late stage of the reaction. After a booster injection of the vaccine at the 10th week, there was a significantly greater HA-antibody response and a greater amount of 2-ME-resistant antibody was formed. The relative passive-protective activity of the sera in mice corresponded to their content of 2-ME-sensitive HA antibody. The protective activity of the sera was 2-ME sensitive and was not removed by absorption with anti-rabbit IgG goat serum. Further, 280 pg of IgM-rich serum fraction obtained by sucrosedensity-gradient ultracentrifugation passively protected against homologous challenge infection in mice, but even 910 pg of IgG-rich fraction did not. These results indicated that the protective antibody was of the IgM class. REFERENCES EKSTFDT,R. D. AND YOSHIDA, K. 1969. Immunity to staphylococcal infection in mice: effect of living versus killed vaccine, role of circulating antibody, and induction of protection-inducing antigen(s) in vivo. J. Bact., 100, 745. FISHER, M. W., DELVIN, H. B. AND ERLANDSON, A. L. 1963. A new staphylococcal antigen: its preparation and immunizing activity against experimental infections. Nature, Lond., 199, 1074. MORSE,S. I. 1962. Isolation and properties of a surface antigen of Staphylococcus uureus. J. exp. Med., 115,295. ROGERS, D. E. 1962 Staphylococci and man. J. Am. Med. Ass., 181, 38. YOSHIDA, K. AND EKSTEDT,R. D. 1968. Antibody response to Staphylococcus aureus in rabbits: sequence of immunoglobulin synthesis and its correlation with passive protection in mice. J. Bact., 96, 1540.
PROTECTI YE ANTIBOD Y AGAINST STMH. EPIDERMIDIS
377
YOSHIDA, K., ICHIMAN, Y. AND OHTOMO, T. 1976. Relation of antiphagocytic activity of a strain of Staphylococcus epidermidis to the induction of resistance in mice. Zentbl. Bakt. ParasitKde, I Abt. Orig. A, Suppl. 5, 819. YOSHIDA,K., MIZUNARI, S. AND HAmom, H. 1964. Immunological studies on " surface antigen " extracted from Staphylococcus aureus. I. Isolation of the antigen and properties of the immune rabbit serum. Jap. J. Microbiol., 8, 67. YOSHIDA, K., S m , M. R. AND NAITO,Y. 1971. Demonstration of serologically different capsular types among strains of Staphylococcus aureus by the serum-soft agar technique. Infect. Immun., 3, 535.
Medical Microbiology Vol. 11, No. 4 IMMU N O L O G I C A L RESPONSE TO A S T RAI N OF S T A P H Y L O C O C C U S E P I D E R M I D I S I N T H E RABBIT: P R O D U C T I O N O F PROTECTIVE ANTIBODY
K. YOSHIDA AND Y. ICHIMAN Department of Microbiology, St Marianna University School of Medicine, 2095 Sugao, Takatu-ku, Kawasaki 213, Japan
STXPHYLOCOCCUS EpIDERMIDIS strains have been regarded as being nonpathogenic for animals. Recently, however, a strain of S. epidermidis exhibiting mouse virulence related to antiphagocytic activity was isolated from a human source. Mice immunised with this strain showed significant resistance to homologous challenge which appeared 10 to 20 days after immunisation (Yoshida, Ichiman and Ohtomo, 1976). This phenomenon was regarded as being similar to that observed with the Smith diffuse strain of S. aureus in mice and rabbits (Yoshida and Ekstedt, 1968; Ekstedt and Yoshida, 1969). We have now attempted to find out whether the immunological properties of the strain and the nature of the protective antibody coincide with those observed with encapsulated strains of S. aureus. MATERIALS AND METHODS Strain. Strain 1142 of S. epidermidis, which was described elsewhere (Yoshida et al., 1976), was used. It gave negative reactions for coagulase, clumping factor, mannitol fermentation, deoxyribonuclease and acid phosphatase by the methods of Yoshida, Smith and Naito (1971). Animals. Mice of strain DD (Nihon Clea Co. Ltd, Tokyo) weighing c. 17 g and normal adult Flemish cross rabbits weighing c. 2 - 0 kg (Nihon Clea Co. Ltd, Tokyo, Japan) were used. Immunisation schedule. Strain 1142 was cultured overnight at 37°C in Brain Heart Infusion (BHI,Difco) broth, was washed once with saline (0.85% NaCl w/v in water) and resuspended in a similar solution. The turbidity of the cell suspension was nephelometrically adjusted to El -0 at 430 nm, and the suspension was then autoclaved at 121°C for 15 min. One ml of this cell suspensioncontained 8.35 x lo8colony forming units (c.f.u.) as determined by a surfaceviable-count method. The cell suspension was injected intravenouslyinto arabbit, 0.5 ml on 3 successive days. Ten weeks later, a single intravenous injection of a similar Received 2 May 1977; revised version accepted 2 Nov. 1977. 1. MED. MICROBI0L.-VOL.
11 (1978)
371
2 B
K. YOSHIDA AND Y . ICHIMAN
372
dose of vaccine was made. After the initial immunisation, 5 ml of blood were withdrawn weekly for 16 weeks and the serum was separated. Procedurefor the examination of mouse virulence. A similar but unheated suspension of strain 1142 was injected intraperitoneally in 0.5-ml amounts into each of five mice. Deaths were recorded for 4 weeks after challenge. Passive protection tests. Rabbit antisera were serially diluted two-fold with sterile saline and 0.5-ml amounts of the dilutions were injected intraperitoneally into groups of five mice. Thirty minutes later, the mice were each challenged intraperitoneally with 0.5 mi of the cell suspension containing c. 8 x lo8c.f.u. of strain 1142 as above, and the number of deaths was recorded for 4 weeks. The mouse-passive-protective activities of IgM- and IgG-rich fractions, obtained by sucrose-gradient ultracentrifugation as described by Yoshida and Ekstedt (1968) were determined similarly. Sensitised haemagglutinationtest. Ten mg of the surface polysaccharide antigen extracted from strain 1142 by the method of Yoshida et al. (1976), were dissolved in 10 ml of 0 . 0 6 7 ~phosphate-buffered saline (PBS), p H 6.5; 0.25 ml of sheep RBC was added to this and the mixture kept at 37°C for 2 h. The cells were washed three times with PBS, and a 0.25% (v/v) suspension of them was prepared. Rabbit antisera, which had been absorbed with similar volumes of packed sheep RBC, were diluted 1 in 10 and then serially diluted in two-fold steps with PBS. With these sera, haemagglutination titres were determined as previously described (Yoshida, Mizunari and Hattori, 1964). Mercaptoethanol treatment and IgG absorption of the sera. Rabbit antisera were treated with 0 - 2 2-mercaptoethanol ~ (2-ME) and iodoacetic acid (Yoshida and Ekstedt, 1968). Also, to remove the IgG from rabbit antisera, 1 .O ml of anti-rabbit IgG goat serum (Hyland) was added to an equal volume of the rabbit antiserum; after 2 h at 37°C the mixture was centrifuged at 7000 g for 15 min. and the supernate separated. Sucrose-density-gradientfractionation of sera. The serum sample was centrifuged through 5 - 0ml of 5 2 0 % sucrose gradient prepared in PBS at 86 OOO g for 18 h in an ultracentrifuge (Model 55P-2, Hitachi Co. Ltd, Tokyo, Japan) with an RPS50 head. Fractions of c. 0.2 ml
10 240
co
I
n
FIG.1 .-Haemagglutination titres of serum from rabbit immunised with strain 1142 of Staphylococcus epidermidis. Symbols: solid line, untreated; broken line, mercaptoethanol-treated serum. Arrows indicate times of injections.
PROTECTIVE ANTIBOD Y AGAINST STMH. EPIDERMIDIS
373
were then collected and diluted to 1 -0ml with 10% sucrose; the optical densities were read at 280 nm by means of a spectrophotometer (Model UV-200, Shimazu Co. Ltd, Kyoto, Japan).
RESULTS Antibody response to immunisation with S. epidermidis One week after the initial immunisation with heat-killed vaccine of strain 1142, the sensitised haemagglutination (HA) titre of the untreated serum reached 640. It continued at this level for 4 weeks and then gradually declined until the 10th week (fig. 1). When a booster injection was given at the 10th week, the titre rose rapidly to 10240 at the 11th week but had decreased to 640 at the 13th week and remained at this level until the 16th week when the examination was terminated. In sera that had been treated with 2-ME, low HA activity appeared as early as the 3rd week, but the titre remained constant at 20 between the 4th and 10th weeks. After the booster injection, however, the HA titre increased to 160 and remained at this level from the 11th to the 13th week, but had fallen to 20 by the 15th week. Rabbit antiserum showed passive-protective activity in mice as early as the 1st week after vaccination (table I), and the " primary " response was greatest at the 2nd week. It tended to decrease progressively after the 4th week, and was very weak by the 6th week. After the booster injection, the highest activity was shown in the 11th and 12th weeks; then there was a rapid decline, and no activity was shown at the 16th week. A comparison of the HA titres and the relative passive-protective activity of the sera suggested that protection was associated with 2-ME-sensitive antibody. TABLE I Mouse passive-protective activity of rabbit antisera taken at weekly intervals after irnmunisation with S. epidermidis strain 1142 Weeks after immunisation
O*
1 2 3 4 6 8 lot 11 12 13 14 16
HA titre
0 640 640 640 640 320 160 160 10 240 2 560 640 640 640
Number of mice deadlnumber challenged after receiving serum diluted 1 in '5
10
515 015 015 015
115
115
3:
215 415
...
015
115 215
3;
515 515
20
...
415 315 415 315 415 315
... ...
* Control. t Booster injection given.
40
so
...
... ... ...
515 415 515 515
515
... ...
...
3; ...
... ... .-.
...
...
K . YOSHIDA AND Y. ICHIMAN
374
Nature of protective antibody to S. epidermidis Experiments were therefore carried out to confirm that passive-protective activity was associated with IgM antibody. As shown in table 11, it was 2-ME sensitive and was not absorbed from the serum by rabbit anti-IgG goat serum. To elucidate further the nature of the protective activity, rabbit-antiserum fractions obtained by sucrose-density-gradient ultracentrifugation were examined for passive protective activity in the mouse (table 111). The results showed that 280 pg of bottom fraction, rich in IgM, was capable of passively protecting mice against homologous challenge. More than 910 pg of the upper fraction, rich both in IgA and IgG, had no such activity. Also, most of the HA activity was present in the bottom fraction (fig. 2).
DISCUSSION In a previous contribution (Yoshida et al., 1976) it was reported that strain 1142 of S. epidermidis, which was used in the present experiments, possessed a protective antigen on the cell surface and that this was associated with antiTABLE I1 Eflect of mercaptoethanol(2-ME) and IgG absorption on passive-protective activity of hyperimmune rabbit antiserum in mice Number of mice deadlnumber challenged
Serum given (and treatment)* I
Hyperimmune (2-ME) Hyperimmune (absorption with anti-rabbit IgG) Hyperimmune (untreated) Normal rabbit (untreated)
* The hyperimmune serum had been collected in the 11th week after immunisation. It was given intraperitoneally in a dose of 0.5 ml per mouse. TABLE I11 Mouse passive-protective activity of immunoglobulins separated by sucrose-density-gradient ultracentrifugationof serum taken from rabbit immunised with strain 1142 Passive immunisation of normal mice with IgM-rich fraction with IgG-rich fraction
I
Number of mice dead/number challenged
{E:i { 'E;i
normal rabbit serum (control)
51s
375
PROTECTIVE ANTIBODY AGAINST STAPH. EPIDERMIDIS
"Ol
-16
-8
Q) c,
.-L
c,
< I
I' I
HA titre 8
2
b
4
I
6
HA titre I
8
I
I
I
10 12 14 Tube number
I
16
1
18
6
20
.22
FIG.2.-Haemagglutination (HA) titres of fractions obtained by sucrose-density ultracentrifugation from a rabbit antiserum collected at the 11th week after immunisation.
phagocytic activity. Further, mice immunised with this organism showed significant resistance to challenge with it during the ensuing 10-20 days. These phenomena were similar to those observed with the Smith diffuse strain of S. aureus by Yoshida and Ekstedt (1968). In the experiments now recorded, the production of antibody in the rabbit against the surface polysaccharide antigen was monitored by means of the HA test. In the primary response, most of the HA was 2-ME sensitive, but greater amounts of 2-ME-resistant HA antibody were detected in antisera obtained after a booster injection. The HA titres were comparable with the relative passive-protective activity of sera in mice. Further, protective activity was removed from antiserum by 2-ME treatment but not by absorption with antiIgG serum. Also, after fractionation by sucrose-density-gradient ultracentrifugation of an antiserum that showed the highest titre in the HA test and the greatest activity in the protection test in the course of the secondary response, passive-protective activity, and most of the HA activity, was detected in the IgM-rich fraction, which would have contained only minor amounts of IgA. However, no passive protective activity was found in the IgG-rich fraction,
376
K. YOSHIDA AND Y. ICHIMAN
which would have contained a greater amount of IgA than the IgM-rich fraction. These results indicate that the protective antibody would positively be IgM-globulin in nature. Yoshida et al. (1976) observed that active immunisation of mice with 10 pg of the surface polysaccharide gave protection against homologous challenge, and that the passive-protective activity of rabbit antiserum was completely removed by absorption with this antigen. Thus, the protective antibody in rabbit antiserum may be regarded as IgM formed in response to the surface antigen. With regard to the protection-inducing antigen in S. aureus strains, this has been regarded as being present only in encapsulated strains (Morse, 1962 ; Fisher, Delvin and Erlandson, 1963;Rogers, 1962).The biological and immunological characters of the protection-inducing polysaccharide of S. epidermidis strain 1142 suggest that this is a capsular antigen. The occurrence and distribution of strains with similar properties among isolates of S. epidermidis are at present under investigation in our laboratory. SUMMARY
A rabbit was immunised with a heat-killed vaccine prepared from strain 1142 of Staphylococcus epidermidis. The HA titres of the antisera against a surface-polysaccharide antigen extracted from the homologous strain were highest in the 2nd-4th weeks after immunisation and subsequently declined. Only a minor amount of 2-ME-resistant antibody was formed, and this at a late stage of the reaction. After a booster injection of the vaccine at the 10th week, there was a significantly greater HA-antibody response and a greater amount of 2-ME-resistant antibody was formed. The relative passive-protective activity of the sera in mice corresponded to their content of 2-ME-sensitive HA antibody. The protective activity of the sera was 2-ME sensitive and was not removed by absorption with anti-rabbit IgG goat serum. Further, 280 pg of IgM-rich serum fraction obtained by sucrosedensity-gradient ultracentrifugation passively protected against homologous challenge infection in mice, but even 910 pg of IgG-rich fraction did not. These results indicated that the protective antibody was of the IgM class. REFERENCES EKSTFDT,R. D. AND YOSHIDA, K. 1969. Immunity to staphylococcal infection in mice: effect of living versus killed vaccine, role of circulating antibody, and induction of protection-inducing antigen(s) in vivo. J. Bact., 100, 745. FISHER, M. W., DELVIN, H. B. AND ERLANDSON, A. L. 1963. A new staphylococcal antigen: its preparation and immunizing activity against experimental infections. Nature, Lond., 199, 1074. MORSE,S. I. 1962. Isolation and properties of a surface antigen of Staphylococcus uureus. J. exp. Med., 115,295. ROGERS, D. E. 1962 Staphylococci and man. J. Am. Med. Ass., 181, 38. YOSHIDA, K. AND EKSTEDT,R. D. 1968. Antibody response to Staphylococcus aureus in rabbits: sequence of immunoglobulin synthesis and its correlation with passive protection in mice. J. Bact., 96, 1540.
PROTECTI YE ANTIBOD Y AGAINST STMH. EPIDERMIDIS
377
YOSHIDA, K., ICHIMAN, Y. AND OHTOMO, T. 1976. Relation of antiphagocytic activity of a strain of Staphylococcus epidermidis to the induction of resistance in mice. Zentbl. Bakt. ParasitKde, I Abt. Orig. A, Suppl. 5, 819. YOSHIDA,K., MIZUNARI, S. AND HAmom, H. 1964. Immunological studies on " surface antigen " extracted from Staphylococcus aureus. I. Isolation of the antigen and properties of the immune rabbit serum. Jap. J. Microbiol., 8, 67. YOSHIDA, K., S m , M. R. AND NAITO,Y. 1971. Demonstration of serologically different capsular types among strains of Staphylococcus aureus by the serum-soft agar technique. Infect. Immun., 3, 535.
