Microbial
Pathogenesis
1992;
Comparative immunogens Streptococcus Robert Department
(Received
12:
43
efficacy of autolysin and pneumolysin protecting mice against infection by pneumoniae
A. Lock,
David
Hansman
of Microbiology,
March
137-I
and James
Adelaide
27, 1991; accepted
ChildrenS
as
C. Paton”
Hospital,
North
in revised form October
Adelaide,
South
Australia,
13, 1991)
Lock, R. A. (Dept of Microbiology, Adelaide Children’s Hospital, North Adelaide, South Australia, 5006, Australia), D. Hansman and J. C. Paton. Comparative efficacy of autolysin and pneumolysin as immunogens protecting mice against infection by Streptococcus pneumoniae. Microbial
Pathogenesis
1992;
12:
137-l
43.
Previous studies on Streptococcus pneumoniae have established that the pneumococcal proteins autolysin (N-acetylmuramyl-L-alanine amidase) and pneumolysin both contribute significantly to the virulence of the organism. In the present work, autolysin and a defined toxoid derivative of pneumolysin were tested, individually and in combination, for efficacy in a mouse model as antigens protecting against challenge with virulent, wild-type S. pneumoniae. While each antigen alone provided significant protection, the degree of protection was not increased when the antigens were administered together. In an additional experiment, mice were challenged with a genetically-modified mutant strain of pneumococcus unable to express active pneumolysin. Pre-immunization of such mice with autolysin failed to provide any significant protection against the challenge. The results of this study suggest that the most important contribution made by autolysin to the virulence of S. pneumoniae may be its role in mediating the release of pneumolysin from the pneumococcal cytoplasm during infection. Key
words:
amidase;
autolysin;
pneumococcus;
pneumolysin;
Streptococcuspneumoniae.
introduction Current human vaccines against Streptococcus pneumoniae (the pneumococcus) consist of a complex mixture of relatively poorly-immunogenic capsular polysaccharides. Such formulations have serious inadequacies which might be alleviated by supplementation with, or conjugation to, appropriate protein antigens.‘,’ For some years, pneumolysin, a powerful haemolytic cytotoxin produced by S. pneumoniae, was suspected of involvement in the pathogenesis of the organism. Such involvement has recently been confirmed in our laboratory by the demonstration that genetically-constructed mutant pneumococcal strains lacking the ability to express active pneumolysin have greatly reduced virulence in mice.3 We have also shown4 that immunization of mice with purified pneumolysin confers significant non-serotypespecific protection against intranasal challenge with virulent, wild-type pneumococci.
“Author 0882-401
to whom O/92/0201
correspondence 37+07
should $03.00/O
be addressed. @ 1992
Academic
Press
Limited
138
I? A
Lock
era/
Clearly, pneumolysin deserves consideration as a protein component of new-generation anti-pneumococcal vaccines. The pneumolysin used in the present study was Pd-A, a genetically-modified toxoid in which the single cysteine residue of the original protein (at position 428) has been replaced by glycine.5 While pneumolysin derivatives are strong candidates for novel vaccine antigens, other pneumococcal proteins also deserve consideration. One particularly interesting possibility is the cell-wall amidase, autolysin. While autolysin would appear to have no directly toxic effects on the host, we have shown previously that the enzyme nevertheless contributes very significantly to the pathogenicity of S. pneumoniae.6 This was demonstrated by the reduced virulence of defined autolysin-negative pneumococcal mutants. In addition, pre-immunization of mice with autolysin was shown to provide significant partial protection against challenge by a virulent wildtype strain. In this study, we compared the efficacy of the pneumolysoid Pd-A and autolysin as protective antigens, and investigated the possibility that an additional protective effect might be obtained by administration of both antigens together. Results Antigens
The purity of the antigens used in this study was confirmed by SDS-polyacrylamide gel electrophoresis, after which both proteins were visualized as single bands (Fig. 1). The amidase had a mobility corresponding to a molecular mass of 36.5 kDa, as previously reported for autolysin.’ A 20 ng amount of the enzyme was sufficient to solubilize all tritiated-choline label from 100 ng of pneumococcal cell-wall substrate in 15 min at 37°C in PBS. The enzyme was also capable of lysing autolysin-negative M
I
2
Fig. 1. SDS-PAGE of purified antigens. Samples of either autolysin (lane 1) or Pd-A (lane 2) containing 5 pg of protein were subjected to SDS-PAGE (125% acrylamide) and stained, as described in Materials and methods. Lane M contained the following protein markers (from top to bottom): 92.5 kDa, 66.2 kDa, 45.0 kDa, 31 .O kDa, and 21.5 kDa.
Autolysin
and pneumolysin
as pneumococcal
immunogens
139
pneumococcal strains. The detergent-activated, choline-inhibited nature of its activity was characteristic of autolysin. After SDS-PAGE, purified Pd-A co-migrated exactly with native pneumolysin at a mobility corresponding to a molecular mass of 52 kDa. The specific haemolytic activity of Pd-A was about 6200 HU (haemolytic units) per mg protein, which is about 0.6% of the specific activity of native pneumolysin. Antibody
responses
of immunized
mice
After immunization, mouse sera were tested by gel double-immunodiffusion for the presence of antibodies directed against the appropriate antigens (not shown). The sera of mice immunized with autolysin each produced a strong, single precipitin band when tested against purified autolysin. Similarly, the sera of mice immunized with Pd-A each produced a strong, single precipitin band when tested against either Pd-A or purified native pneumolysin. Anit-pneumolysin titres of control sera were about 200 anti-HU/ml, while those of mice immunized with Pd-A were about 2300 anti-HU/ml. In parallel experiments (not shown), immunization of mice with native pneumolysin under the same regime resulted in sera with similar anti-haemolytic titres. Anti-autolysin titres of control sera were all less than 1 (i.e. control sera, even undiluted, did not significantly inhibit the activity of the enzyme). Anti-autolysin titres of sera from appropriately immunized mice ranged from 320 to 420. To confirm that such antibody levels were capable of inhibiting autolysis of pneumococci, S. pneumoniae D39 was grown in Todd-Hewitt broth +0.5% yeast extract, supplemented with a 1 :200 dilution of either control or autolysin-immunized mouse serum (Fig. 2). In accordance with previous studies,’ the presence of anti-autolysin almost completely inhibited autolysis of the pneumococci at the end of the growth phase, as judged by the absorbance of the culture at 600 nm. In the presence of control serum, pneumolysin
Time
(h)
Fig. 2. Effect of anti-autolysin on cellular autolysis and release of pneumolysin. S. pneumoniae D39 was grown in Todd-Hewitt broth +0.5% yeast extract, supplemented with a 1 :200 dilution of control mouse serum (a) or anti-autolysin serum (0). Growth of the culture was monitored by Absorbance at 600 nm (-). Samples were also withdrawn at the indicated times and microfuged, prior to determination of the amount of pneumolysin in the culture supernatant by haemolytic assay (---).
140
R. A. Lock
et a/
appeared in the culture supernatant simultaneously with the apparent onset of autolysis. However, in the presence of anti-autolysin, no pneumolysin had been released after 24 h of culture, compared with a figure of 25 HU/ml in the presence of control serum. After 36 h of culture, the respective amounts of pneumolysin in the culture supernatant were 6 and 27 HU/ml. Thus, exogenous anti-autolysin was capable of inhibiting autolysis and blocking the release of pneumolysin from the pneumococcal cells. Challenge
of mice
Following immunization with either Pd-A, autolysin, both antigens, or sham immunization with no protein, groups of 36 BALB/c mice were challenged intraperitoneally with about 6.5 x 1O’cfu per mouse of the virulent, wild-type S. pneumoniae strain D39. This dose was calculated to be 1O-20 times the 50% lethal dose (LD&. The subsequent survival times of the mice in this experiment are indicated in Fig. 3. Survival rates (number of mice surviving for more than 14 days after challenge) for the sham-immunized group, and for groups immunized with Pd-A, with autolysin, and with both antigens, were 9, 28, 23 and 23, respectively. The presence of pneumococci of the appropriate serotype in the heart blood of dead mice was confirmed by culture and Quellung reaction, Statistical analysis of these results, using the x2 test for two independent samples (1 -tailed), indicated no significant difference between immunized groups. For all immunized groups, however, the survival rate was significantly greater than for the control group (P < 0.0005 for the group immunized with Pd-A, and P < 0.005 for groups immunized with autolysin alone, or with autolysin plus Pd-A). These results indicated that immunization with a combination of pneumolysin and autolysin did not provide any protection additional to that obtained by administration of either antigen alone. This is consistent with the suggestion that the protective effect of autolysin is, in large part, due to inhibition of the autolysin-induced release of
Mice
survivmg 9 > 14 days ,4=-------------------------~
28
23
1312-
23
. .
.
IIIO2 a ?
90-
. .
7-
$6-
.
A .
54-
:
;
I CON
I Pd-A
”
I
i
.
:
I AL
I Pd-A+AL
Fig. 3. Effect of immunization with autolysin and pneumolysoid on survival time of mice challenged with S. pneumoniae D39. BALB/c mice were immunized with pneumolysoid (Pd-A). autolysin (AL), both antigens (Pd-A+AL), or were sham-immunized (CON), as described in the text. Mice were then challenged with S. pneumoniae D39, and their survival time was recorded. The number of mice in each group which were alive and well at the conclusion of the experiment (14 days after challenge) is also indicated.
Autolysin
and pneumolysin
as pneumococcal
immunogens
141
pneumolysin from the invading pneumococci. To test this hypothesis further, two groups of 36 BALB/c mice, one group sham-immunized and the other immunized with autolysin, were challenged intraperitoneally with 0.1 ml of an undiluted culture (about lo7 cfu) of S. pneumoniae PLN-A, a strain carrying a defined mutation in the pneumolysin gene such that no active pneumolysin is expressed.3 Because of the markedly reduced virulence of this strain compared with D39, the challenge dose was slightly less than the LDsO, and the experiment was prolonged to 28 days after challenge. Six of the control mice and seven of the autolysin-immunized mice died as a result of the challenge. The median survival times for the mice which died as a result of challenge (6.9 days for the control group, compared with 8.8 days for the immunized group) was also not significantly different (Mann-Whitney U test). Erythromycinresistant, pneumolysin-negative pneumococci of the appropriate serotype were detected in the heart blood of each of the mice that succumbed. Thus, immunization with autolysin provided no protection at all against challenge with pneumolysinnegative pneumococci. Discussion In accordance with previous studies, 3*4,6the present work indicates that both autolysin and pneumolysin, administered individually, are antigens capable of providing mice significant (albeit incomplete) protection against challenge with a virulent pneumococcal strain, D39. However, when mice were challenged intraperitoneally with D39, no additional protection was observed in mice which had received a combination of the autolysin and Pd-A. This result suggests, firstly, that autolysin may be no more effective than pneumolysoid as a protective antigen in protein-based pneumococcal vaccines, and secondly, that the inclusion of both proteins in such vaccines may not be warranted. The challenge experiment involving D39 also has bearing on the question of the exact role played by autolysin in pathogenesis. We have previously suggested6 that autolysin may function during infection by lysing a proportion of the invading pneumococci, thereby releasing potentially lethal toxins such as pneumolysin, neuraminidase,* and possibly other proteins, which would otherwise remain confined to the bacterial cytoplasm. The action of autolysin in the release of highly-inflammatory cell wall breakdown products which could contribute to pathogenesis is also wellestablished.g-” In the present experiments, the failure of autolysin to provide protection to mice additional to that induced by Pd-A, suggests that the most important role for autolysin in our model of infection may be to mediate the release of pneumolysin from S. pneumoniae. This conclusion is supported by the results of the experiment in which pre-immunization of mice with autolysin failed to provide any protection from challenge with a strain of S. pneumoniae specifically lacking a functional pneumolysin gene. In our model, the release by autolysin of toxic products other than pneumolysin may be of little significance to pathogenicity. In conclusion, the results of this study reinforce the view that, at present, pneumolysin toxoid remains the most appropriate candidate for protein supplementation of capsular polysaccharides in the next generation of human anti-pneumococcal vaccines. Materials
and methods
Preparation of pneumolysoid. Pd-A (pneumolysin cys-428 strain JMI 09 harbouring a pUCl8 derivative containing the by G. Boulnois. Pd-A was prepared by a modification of the
-+ gly) was prepared from mutant procedure
PL gene,5 kindly used previously
E. coli supplied for the
142
R. A. Lock
et al
preparation of pneumolysin from S. pneumoniae.4 Briefly, E. co/i were grown under aeration In the presence of ampicillin (50 LLg/ml) in batches of 8 litres of Trypticase soy broth (BBL, Cockeysville, MD) supplemented as previously described.4 The final A,,, of the culture was about 1 .O. Following concentration using an Amicon DC101 LA hollow fibre concentrator fitted with a 0.1 -/lrn exclusion cartridge, cells were pelleted by centrifugation and lysed by French press. Cell debris was removed by centrifugation, and Pd-A was purified by serial column chromatography through DEAE-cellulose (Whatman DE-52) and Sephacryl S-200 Superfine (Pharmacia AB, Uppsala). The presence of Pd-A was monitored by haemolytic assay,4 pooling fractions with activities > 100 haemolytic units (HU) per ml, where 1 HU was defined as that amount of activity required to lyse 50% of a 1% (v/v) suspension of packed human group 0 erythrocytes in phosphate-buffered saline (PBS) for 30 min at 37°C. The final yield of Pd-A was about 2 mg per litre of original culture. The protein was 295% pure judged by SDSPAGE. Preparation of autolysin. Autolysin was purified by affinity column chromatography using choline-Sepharose 66, as previously described6 from E. co/i strain DHI harbouring plasmid pGL80 (kindly provided by R. Lopez). This plasmid contains a 1.2 kb Hind III fragment of pneumococcal DNA carrying the complete pneumococcal autolysin gene.” The enzyme was assayed by solubilization of tritiated pneumococcal cell wall as previously described.‘j Each litre of culture yielded about 1 mg of autolysin which was >95% pure as judged by SDS-PAGE. Protein assay. Protein concentrations bovine serum albumin as the standard. SDS-PAGE. were stained
SDS-PAGE with Coomassie
was conducted Brilliant Blue
were
measured
essentially R250.
by
by
the
the
method
method
of
Bradford,13
of Laemmli14
with
and
gels
Immunization of mice. BALB/c mice (6-8 weeks old) were injected intraperitoneally with 0.2 ml volumes of an emulsion containing either autolysin, Pd-A, or both (20 pg per antigen) in PBS and Freund’s complete adjuvant (Commonwealth Serum Laboratories, Melbourne). At 14-day intervals, the mice were given two additional injections of antigen(s) emulsified in Freund’s incomplete adjuvant (CSL). Control mice received a similar course of injections from which protein was omitted. Blood samples were collected from the suborbital plexus 7 days after the last injection. Sera were tested for the presence of appropriate antibodies by gel double immunodiffusion, and anti-autolysin and anti-pneumolysin titres were also determined. Determination of anti-autolysin titres. Anti-autolysin titles previously described,6 and expressed as the reciprocal of the inhibition of the enzyme activity of 20 ng of purified autolysin. Determination their ability to were determined Construction pneumoniae type 2 strain,
of anti-pneumolysin titres. Sera of mice inhibit the haemolytic activity of native as previously described,4 and expressed of
strain D39,
the pneumolysin-negative PLN-A was constructed as previously reported.3
immunized pneumolysin. as anti-HU
pneumococcus. by insertion-duplication
Challenge of mice. One week after bleeding, the mice 100 ~1 of serum broth containing about 6.5~10~ cfu of times the intraperitoneal LD,, of D39 in BALB/c mice. intraperitoneally with 100 ~1 serum broth containing about S. pneumoniae strain PLN-A. The subsequent survival time
We thank Julie Goldfinch and Kylie Maddison helpful discussions. This work was supported Research Council of Australia.
of sera were serum dilution
The
determined causing
with Pd-A were Anti-pneumolysin per ml serum.
tested
as 50%
for titres
pneumolysin-negative S. mutagenesis of the virulent
were injected intraperitoneally with D39. This dose represents about 20 Alternatively, mice were challenged IO’ cfu of the pneumolysin-negative of each mouse was recorded.
for technical by a grant from
assistance the National
and Anne Berry for Health and Medical
Autolysin
and pneumolysin
as pneumococcal
immunogens
143
References 1.
2. 3. 4. 5.
6. 7. 8
9 10 11 12 13 14
La Force FM, Eickhoff TC. Pneumococcal vaccine: An emerging consensus, Ann Intern Med 1988; 108: 757-9. Davies AJ, Kumaratne DS. The continuing problem of pneumococcal infection. J Antimicrob Chemother 1988; 21: 387-91. Berry AM, Yother J, Briles DE, Hansman D, Paton JC. Reduced virulence of a defined pneumolysinnegative mutant of Streptococcus pneumoniae. Infect lmmun 1989; 57: 2037-42. Paton JC, Lock RA, Hansman DJ. Effect of immunization with pneumolysin on survival time of mice challenged with Streptococcus pneumoniae. Infect lmmun 1983; 40: 548-52. Saunders FK, Mitchell TJ, Walker JA, Andrew PW, Boulnois GJ. Pneumolysin, the thiol-activated toxin of Streptococcus pneumoniae, does not require a throl group for in vitro activity. Infect lmmun 1989; 57: 2547-52. Berry AM, Lock RA, Hansman D, Paton JC. Contribution of autolysin to virulence of Streptococcus pneumoniae. Infect lmmun 1989; 57: 2324-30. Briese T, Hakenbeck R. Interaction of the pneumococcal amidase with lipoteichoic acid and choline. Eur J Biochem 1985; 146: 417-27. Lock RA, Paton JC, Hansman D. Comparative efficacy of pneumococcal neuraminidase and pneumolysin as immunogens protective against Streptococcus pneumoniae. Microb Pathogen 1988; 5: 461-7. Chetty C, Kreger A. Characterization of pneumococcal purpura-producing principle. Infect lmmun 1980; 29: 158-64. Chetty C, Kreger A. Role of autolysin in generating pneumococcal purpura-producing principle. Infect lmmun 1981; 31: 339-44. Tuomanen E, Rich R. Zak 0. Induction of pulmonary inflammation by components of the pneumococcal cell surface. Am Rev Respir Dis 1987; 135: 869-74. Garcia P, Garcia JL, Garcia E, Lopez R. Nucleotide sequence and expression of the pneumococcal autolysin gene from its own promoter in Escherichia co/i. Gene 1986; 43: 265-72. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protern utilizing the principle of protein-dye binding. Anal Biochem 1976; 72: 248-54. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970; 227: 680-5.
Pathogenesis
1992;
Comparative immunogens Streptococcus Robert Department
(Received
12:
43
efficacy of autolysin and pneumolysin protecting mice against infection by pneumoniae
A. Lock,
David
Hansman
of Microbiology,
March
137-I
and James
Adelaide
27, 1991; accepted
ChildrenS
as
C. Paton”
Hospital,
North
in revised form October
Adelaide,
South
Australia,
13, 1991)
Lock, R. A. (Dept of Microbiology, Adelaide Children’s Hospital, North Adelaide, South Australia, 5006, Australia), D. Hansman and J. C. Paton. Comparative efficacy of autolysin and pneumolysin as immunogens protecting mice against infection by Streptococcus pneumoniae. Microbial
Pathogenesis
1992;
12:
137-l
43.
Previous studies on Streptococcus pneumoniae have established that the pneumococcal proteins autolysin (N-acetylmuramyl-L-alanine amidase) and pneumolysin both contribute significantly to the virulence of the organism. In the present work, autolysin and a defined toxoid derivative of pneumolysin were tested, individually and in combination, for efficacy in a mouse model as antigens protecting against challenge with virulent, wild-type S. pneumoniae. While each antigen alone provided significant protection, the degree of protection was not increased when the antigens were administered together. In an additional experiment, mice were challenged with a genetically-modified mutant strain of pneumococcus unable to express active pneumolysin. Pre-immunization of such mice with autolysin failed to provide any significant protection against the challenge. The results of this study suggest that the most important contribution made by autolysin to the virulence of S. pneumoniae may be its role in mediating the release of pneumolysin from the pneumococcal cytoplasm during infection. Key
words:
amidase;
autolysin;
pneumococcus;
pneumolysin;
Streptococcuspneumoniae.
introduction Current human vaccines against Streptococcus pneumoniae (the pneumococcus) consist of a complex mixture of relatively poorly-immunogenic capsular polysaccharides. Such formulations have serious inadequacies which might be alleviated by supplementation with, or conjugation to, appropriate protein antigens.‘,’ For some years, pneumolysin, a powerful haemolytic cytotoxin produced by S. pneumoniae, was suspected of involvement in the pathogenesis of the organism. Such involvement has recently been confirmed in our laboratory by the demonstration that genetically-constructed mutant pneumococcal strains lacking the ability to express active pneumolysin have greatly reduced virulence in mice.3 We have also shown4 that immunization of mice with purified pneumolysin confers significant non-serotypespecific protection against intranasal challenge with virulent, wild-type pneumococci.
“Author 0882-401
to whom O/92/0201
correspondence 37+07
should $03.00/O
be addressed. @ 1992
Academic
Press
Limited
138
I? A
Lock
era/
Clearly, pneumolysin deserves consideration as a protein component of new-generation anti-pneumococcal vaccines. The pneumolysin used in the present study was Pd-A, a genetically-modified toxoid in which the single cysteine residue of the original protein (at position 428) has been replaced by glycine.5 While pneumolysin derivatives are strong candidates for novel vaccine antigens, other pneumococcal proteins also deserve consideration. One particularly interesting possibility is the cell-wall amidase, autolysin. While autolysin would appear to have no directly toxic effects on the host, we have shown previously that the enzyme nevertheless contributes very significantly to the pathogenicity of S. pneumoniae.6 This was demonstrated by the reduced virulence of defined autolysin-negative pneumococcal mutants. In addition, pre-immunization of mice with autolysin was shown to provide significant partial protection against challenge by a virulent wildtype strain. In this study, we compared the efficacy of the pneumolysoid Pd-A and autolysin as protective antigens, and investigated the possibility that an additional protective effect might be obtained by administration of both antigens together. Results Antigens
The purity of the antigens used in this study was confirmed by SDS-polyacrylamide gel electrophoresis, after which both proteins were visualized as single bands (Fig. 1). The amidase had a mobility corresponding to a molecular mass of 36.5 kDa, as previously reported for autolysin.’ A 20 ng amount of the enzyme was sufficient to solubilize all tritiated-choline label from 100 ng of pneumococcal cell-wall substrate in 15 min at 37°C in PBS. The enzyme was also capable of lysing autolysin-negative M
I
2
Fig. 1. SDS-PAGE of purified antigens. Samples of either autolysin (lane 1) or Pd-A (lane 2) containing 5 pg of protein were subjected to SDS-PAGE (125% acrylamide) and stained, as described in Materials and methods. Lane M contained the following protein markers (from top to bottom): 92.5 kDa, 66.2 kDa, 45.0 kDa, 31 .O kDa, and 21.5 kDa.
Autolysin
and pneumolysin
as pneumococcal
immunogens
139
pneumococcal strains. The detergent-activated, choline-inhibited nature of its activity was characteristic of autolysin. After SDS-PAGE, purified Pd-A co-migrated exactly with native pneumolysin at a mobility corresponding to a molecular mass of 52 kDa. The specific haemolytic activity of Pd-A was about 6200 HU (haemolytic units) per mg protein, which is about 0.6% of the specific activity of native pneumolysin. Antibody
responses
of immunized
mice
After immunization, mouse sera were tested by gel double-immunodiffusion for the presence of antibodies directed against the appropriate antigens (not shown). The sera of mice immunized with autolysin each produced a strong, single precipitin band when tested against purified autolysin. Similarly, the sera of mice immunized with Pd-A each produced a strong, single precipitin band when tested against either Pd-A or purified native pneumolysin. Anit-pneumolysin titres of control sera were about 200 anti-HU/ml, while those of mice immunized with Pd-A were about 2300 anti-HU/ml. In parallel experiments (not shown), immunization of mice with native pneumolysin under the same regime resulted in sera with similar anti-haemolytic titres. Anti-autolysin titres of control sera were all less than 1 (i.e. control sera, even undiluted, did not significantly inhibit the activity of the enzyme). Anti-autolysin titres of sera from appropriately immunized mice ranged from 320 to 420. To confirm that such antibody levels were capable of inhibiting autolysis of pneumococci, S. pneumoniae D39 was grown in Todd-Hewitt broth +0.5% yeast extract, supplemented with a 1 :200 dilution of either control or autolysin-immunized mouse serum (Fig. 2). In accordance with previous studies,’ the presence of anti-autolysin almost completely inhibited autolysis of the pneumococci at the end of the growth phase, as judged by the absorbance of the culture at 600 nm. In the presence of control serum, pneumolysin
Time
(h)
Fig. 2. Effect of anti-autolysin on cellular autolysis and release of pneumolysin. S. pneumoniae D39 was grown in Todd-Hewitt broth +0.5% yeast extract, supplemented with a 1 :200 dilution of control mouse serum (a) or anti-autolysin serum (0). Growth of the culture was monitored by Absorbance at 600 nm (-). Samples were also withdrawn at the indicated times and microfuged, prior to determination of the amount of pneumolysin in the culture supernatant by haemolytic assay (---).
140
R. A. Lock
et a/
appeared in the culture supernatant simultaneously with the apparent onset of autolysis. However, in the presence of anti-autolysin, no pneumolysin had been released after 24 h of culture, compared with a figure of 25 HU/ml in the presence of control serum. After 36 h of culture, the respective amounts of pneumolysin in the culture supernatant were 6 and 27 HU/ml. Thus, exogenous anti-autolysin was capable of inhibiting autolysis and blocking the release of pneumolysin from the pneumococcal cells. Challenge
of mice
Following immunization with either Pd-A, autolysin, both antigens, or sham immunization with no protein, groups of 36 BALB/c mice were challenged intraperitoneally with about 6.5 x 1O’cfu per mouse of the virulent, wild-type S. pneumoniae strain D39. This dose was calculated to be 1O-20 times the 50% lethal dose (LD&. The subsequent survival times of the mice in this experiment are indicated in Fig. 3. Survival rates (number of mice surviving for more than 14 days after challenge) for the sham-immunized group, and for groups immunized with Pd-A, with autolysin, and with both antigens, were 9, 28, 23 and 23, respectively. The presence of pneumococci of the appropriate serotype in the heart blood of dead mice was confirmed by culture and Quellung reaction, Statistical analysis of these results, using the x2 test for two independent samples (1 -tailed), indicated no significant difference between immunized groups. For all immunized groups, however, the survival rate was significantly greater than for the control group (P < 0.0005 for the group immunized with Pd-A, and P < 0.005 for groups immunized with autolysin alone, or with autolysin plus Pd-A). These results indicated that immunization with a combination of pneumolysin and autolysin did not provide any protection additional to that obtained by administration of either antigen alone. This is consistent with the suggestion that the protective effect of autolysin is, in large part, due to inhibition of the autolysin-induced release of
Mice
survivmg 9 > 14 days ,4=-------------------------~
28
23
1312-
23
. .
.
IIIO2 a ?
90-
. .
7-
$6-
.
A .
54-
:
;
I CON
I Pd-A
”
I
i
.
:
I AL
I Pd-A+AL
Fig. 3. Effect of immunization with autolysin and pneumolysoid on survival time of mice challenged with S. pneumoniae D39. BALB/c mice were immunized with pneumolysoid (Pd-A). autolysin (AL), both antigens (Pd-A+AL), or were sham-immunized (CON), as described in the text. Mice were then challenged with S. pneumoniae D39, and their survival time was recorded. The number of mice in each group which were alive and well at the conclusion of the experiment (14 days after challenge) is also indicated.
Autolysin
and pneumolysin
as pneumococcal
immunogens
141
pneumolysin from the invading pneumococci. To test this hypothesis further, two groups of 36 BALB/c mice, one group sham-immunized and the other immunized with autolysin, were challenged intraperitoneally with 0.1 ml of an undiluted culture (about lo7 cfu) of S. pneumoniae PLN-A, a strain carrying a defined mutation in the pneumolysin gene such that no active pneumolysin is expressed.3 Because of the markedly reduced virulence of this strain compared with D39, the challenge dose was slightly less than the LDsO, and the experiment was prolonged to 28 days after challenge. Six of the control mice and seven of the autolysin-immunized mice died as a result of the challenge. The median survival times for the mice which died as a result of challenge (6.9 days for the control group, compared with 8.8 days for the immunized group) was also not significantly different (Mann-Whitney U test). Erythromycinresistant, pneumolysin-negative pneumococci of the appropriate serotype were detected in the heart blood of each of the mice that succumbed. Thus, immunization with autolysin provided no protection at all against challenge with pneumolysinnegative pneumococci. Discussion In accordance with previous studies, 3*4,6the present work indicates that both autolysin and pneumolysin, administered individually, are antigens capable of providing mice significant (albeit incomplete) protection against challenge with a virulent pneumococcal strain, D39. However, when mice were challenged intraperitoneally with D39, no additional protection was observed in mice which had received a combination of the autolysin and Pd-A. This result suggests, firstly, that autolysin may be no more effective than pneumolysoid as a protective antigen in protein-based pneumococcal vaccines, and secondly, that the inclusion of both proteins in such vaccines may not be warranted. The challenge experiment involving D39 also has bearing on the question of the exact role played by autolysin in pathogenesis. We have previously suggested6 that autolysin may function during infection by lysing a proportion of the invading pneumococci, thereby releasing potentially lethal toxins such as pneumolysin, neuraminidase,* and possibly other proteins, which would otherwise remain confined to the bacterial cytoplasm. The action of autolysin in the release of highly-inflammatory cell wall breakdown products which could contribute to pathogenesis is also wellestablished.g-” In the present experiments, the failure of autolysin to provide protection to mice additional to that induced by Pd-A, suggests that the most important role for autolysin in our model of infection may be to mediate the release of pneumolysin from S. pneumoniae. This conclusion is supported by the results of the experiment in which pre-immunization of mice with autolysin failed to provide any protection from challenge with a strain of S. pneumoniae specifically lacking a functional pneumolysin gene. In our model, the release by autolysin of toxic products other than pneumolysin may be of little significance to pathogenicity. In conclusion, the results of this study reinforce the view that, at present, pneumolysin toxoid remains the most appropriate candidate for protein supplementation of capsular polysaccharides in the next generation of human anti-pneumococcal vaccines. Materials
and methods
Preparation of pneumolysoid. Pd-A (pneumolysin cys-428 strain JMI 09 harbouring a pUCl8 derivative containing the by G. Boulnois. Pd-A was prepared by a modification of the
-+ gly) was prepared from mutant procedure
PL gene,5 kindly used previously
E. coli supplied for the
142
R. A. Lock
et al
preparation of pneumolysin from S. pneumoniae.4 Briefly, E. co/i were grown under aeration In the presence of ampicillin (50 LLg/ml) in batches of 8 litres of Trypticase soy broth (BBL, Cockeysville, MD) supplemented as previously described.4 The final A,,, of the culture was about 1 .O. Following concentration using an Amicon DC101 LA hollow fibre concentrator fitted with a 0.1 -/lrn exclusion cartridge, cells were pelleted by centrifugation and lysed by French press. Cell debris was removed by centrifugation, and Pd-A was purified by serial column chromatography through DEAE-cellulose (Whatman DE-52) and Sephacryl S-200 Superfine (Pharmacia AB, Uppsala). The presence of Pd-A was monitored by haemolytic assay,4 pooling fractions with activities > 100 haemolytic units (HU) per ml, where 1 HU was defined as that amount of activity required to lyse 50% of a 1% (v/v) suspension of packed human group 0 erythrocytes in phosphate-buffered saline (PBS) for 30 min at 37°C. The final yield of Pd-A was about 2 mg per litre of original culture. The protein was 295% pure judged by SDSPAGE. Preparation of autolysin. Autolysin was purified by affinity column chromatography using choline-Sepharose 66, as previously described6 from E. co/i strain DHI harbouring plasmid pGL80 (kindly provided by R. Lopez). This plasmid contains a 1.2 kb Hind III fragment of pneumococcal DNA carrying the complete pneumococcal autolysin gene.” The enzyme was assayed by solubilization of tritiated pneumococcal cell wall as previously described.‘j Each litre of culture yielded about 1 mg of autolysin which was >95% pure as judged by SDS-PAGE. Protein assay. Protein concentrations bovine serum albumin as the standard. SDS-PAGE. were stained
SDS-PAGE with Coomassie
was conducted Brilliant Blue
were
measured
essentially R250.
by
by
the
the
method
method
of
Bradford,13
of Laemmli14
with
and
gels
Immunization of mice. BALB/c mice (6-8 weeks old) were injected intraperitoneally with 0.2 ml volumes of an emulsion containing either autolysin, Pd-A, or both (20 pg per antigen) in PBS and Freund’s complete adjuvant (Commonwealth Serum Laboratories, Melbourne). At 14-day intervals, the mice were given two additional injections of antigen(s) emulsified in Freund’s incomplete adjuvant (CSL). Control mice received a similar course of injections from which protein was omitted. Blood samples were collected from the suborbital plexus 7 days after the last injection. Sera were tested for the presence of appropriate antibodies by gel double immunodiffusion, and anti-autolysin and anti-pneumolysin titres were also determined. Determination of anti-autolysin titres. Anti-autolysin titles previously described,6 and expressed as the reciprocal of the inhibition of the enzyme activity of 20 ng of purified autolysin. Determination their ability to were determined Construction pneumoniae type 2 strain,
of anti-pneumolysin titres. Sera of mice inhibit the haemolytic activity of native as previously described,4 and expressed of
strain D39,
the pneumolysin-negative PLN-A was constructed as previously reported.3
immunized pneumolysin. as anti-HU
pneumococcus. by insertion-duplication
Challenge of mice. One week after bleeding, the mice 100 ~1 of serum broth containing about 6.5~10~ cfu of times the intraperitoneal LD,, of D39 in BALB/c mice. intraperitoneally with 100 ~1 serum broth containing about S. pneumoniae strain PLN-A. The subsequent survival time
We thank Julie Goldfinch and Kylie Maddison helpful discussions. This work was supported Research Council of Australia.
of sera were serum dilution
The
determined causing
with Pd-A were Anti-pneumolysin per ml serum.
tested
as 50%
for titres
pneumolysin-negative S. mutagenesis of the virulent
were injected intraperitoneally with D39. This dose represents about 20 Alternatively, mice were challenged IO’ cfu of the pneumolysin-negative of each mouse was recorded.
for technical by a grant from
assistance the National
and Anne Berry for Health and Medical
Autolysin
and pneumolysin
as pneumococcal
immunogens
143
References 1.
2. 3. 4. 5.
6. 7. 8
9 10 11 12 13 14
La Force FM, Eickhoff TC. Pneumococcal vaccine: An emerging consensus, Ann Intern Med 1988; 108: 757-9. Davies AJ, Kumaratne DS. The continuing problem of pneumococcal infection. J Antimicrob Chemother 1988; 21: 387-91. Berry AM, Yother J, Briles DE, Hansman D, Paton JC. Reduced virulence of a defined pneumolysinnegative mutant of Streptococcus pneumoniae. Infect lmmun 1989; 57: 2037-42. Paton JC, Lock RA, Hansman DJ. Effect of immunization with pneumolysin on survival time of mice challenged with Streptococcus pneumoniae. Infect lmmun 1983; 40: 548-52. Saunders FK, Mitchell TJ, Walker JA, Andrew PW, Boulnois GJ. Pneumolysin, the thiol-activated toxin of Streptococcus pneumoniae, does not require a throl group for in vitro activity. Infect lmmun 1989; 57: 2547-52. Berry AM, Lock RA, Hansman D, Paton JC. Contribution of autolysin to virulence of Streptococcus pneumoniae. Infect lmmun 1989; 57: 2324-30. Briese T, Hakenbeck R. Interaction of the pneumococcal amidase with lipoteichoic acid and choline. Eur J Biochem 1985; 146: 417-27. Lock RA, Paton JC, Hansman D. Comparative efficacy of pneumococcal neuraminidase and pneumolysin as immunogens protective against Streptococcus pneumoniae. Microb Pathogen 1988; 5: 461-7. Chetty C, Kreger A. Characterization of pneumococcal purpura-producing principle. Infect lmmun 1980; 29: 158-64. Chetty C, Kreger A. Role of autolysin in generating pneumococcal purpura-producing principle. Infect lmmun 1981; 31: 339-44. Tuomanen E, Rich R. Zak 0. Induction of pulmonary inflammation by components of the pneumococcal cell surface. Am Rev Respir Dis 1987; 135: 869-74. Garcia P, Garcia JL, Garcia E, Lopez R. Nucleotide sequence and expression of the pneumococcal autolysin gene from its own promoter in Escherichia co/i. Gene 1986; 43: 265-72. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protern utilizing the principle of protein-dye binding. Anal Biochem 1976; 72: 248-54. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970; 227: 680-5.
