Published in Volume
103, Issue 9
(May 1, 1999)J Clin Invest.
Copyright © 1999, American Society for Clinical Investigation
New protective antigen of group A streptococci
1Veterans Affairs Medical Center, Memphis, Tennessee 38104, USA2Department of Medicine, and3Department of Anatomy and Neurobiology, University of Tennessee, Memphis, Tennessee 38104, USA
Address correspondence to: James B. Dale, Veterans Affairs Medical Center (11A), 1030 Jefferson Avenue, Memphis, Tennessee 38104, USA. Phone: (901) 577-7207; Fax: (901) 448-8231; E-mail: firstname.lastname@example.org
Published May 1, 1999
Submitted: September 3,
1998; Accepted: March 23,
It is widely believed that the surface M protein of group A streptococci is the predominant surface protein of these organisms containing opsonic epitopes. In the present study, we identified a new surface protein, distinct from M protein, that evokes protective antibodies. A type 18 M-negative mutant was found to be both resistant to phagocytosis in human blood and virulent in mice. The wild-type strain, but not the M-negative mutant, was opsonized by antisera against purified recombinant M18 protein or a synthetic peptide copying the NH2-terminus of M18. However, antisera raised against a crude pepsin extract of the M-negative mutant opsonized both strains, indicating the presence of a protective antigen in addition to type 18 M protein. This antiserum was used to identify and purify a 24-kDa protein fragment (Spa, streptococcal protective antigen) that evoked antibodies that opsonized the M18 parent and the M-negative mutant. The results of passive mouse protection tests confirmed the presence of protective epitopes within Spa. The deduced amino acid sequence of a 636-bp 5′ fragment of the spa18 gene showed no homology with sequences in GenBank. These studies reveal the presence of a new protective antigen of certain strains of group A streptococci that may prove to be an important component of vaccines to prevent streptococcal infections.
Group A streptococcal infections cause a wide array of clinical syndromes; these range from uncomplicated pharyngitis and pyoderma to serious, invasive infections and toxic shock syndrome. Protection against infection is largely mediated by antibodies against the surface M protein of the organisms. The M protein is an α-helical, coiled-coil molecule that extends from the surface with its hypervariable NH2-terminus exposed to the outside and its conserved COOH-terminus buried in the cytoplasm (1). The NH2-terminus contains type-specific epitopes evoking bactericidal antibodies that correlate with protection against the homologous serotype (2). The emm gene is located in a regulon controlled by the upstream positive regulator mga (3). Depending on the serotype, the regulon may contain 1, 2, or 3 emm and emm-like genes (4). In serotypes containing only 1 emm gene, deletion (5) or interruption (6) of the emm gene results in an avirulent organism that can no longer resist phagocytosis. In serotypes expressing several emm-like genes, each may partially contribute to phagocytosis resistance (7), but among the many defined surface proteins of group A streptococci, only antibodies against the M protein have been shown to be opsonic.
In this study, we identify a new surface antigen, streptococcal protective antigen (Spa) of group A streptococci, that evokes protective antibodies. A fully virulent M-negative mutant of type 18 streptococci was opsonized by rabbit antiserum raised against a crude pepsin extract of the M-negative mutant. This antiserum was used to identify and purify the new protective antigen, Spa. Antiserum against the purified Spa protein was bactericidal against both the type 18 parent strain and, importantly, the M-negative mutant. Passive mouse protection tests using rabbit antisera raised against purified Spa confirmed that the new surface protein contained protective epitopes. In addition, the anti-Spa antibodies opsonized type 3 and type 28 streptococci, indicating that this new surface protein is a cross-protective antigen for certain strains of group A streptococci.
Bacterial strains. The parent type 18 streptococcal strain 87-282, has been the subject of previous studies in this laboratory (8) and others (9) and was obtained from Patrick Cleary (University of Minnesota, Minneapolis, Minnesota, USA). The M3 strain (3375) was provided by James Musser (Baylor College of Medicine, Houston, Texas, USA). The M28 strain (S2356) was from our laboratory collection.
Reagents. The preparation of rabbit antisera against the synthetic peptide SM18(1-30)C has been described previously (8). Antisera were also raised against recombinant M18 that was purified from periplasmic extracts of Escherichia coli, as previously described (8, 10). Rabbit antisera were similarly prepared against purified Spa (8). A peptide copying the NH2-terminal 23 amino acid residues of Spa [S-Spa18(1-23)C] was commercially synthesized (Research Genetics Inc., Huntsville, Alabama, USA) and chemically coupled to KLH by methods described previously (11). Antisera against S-Spa18(1-23)C-KLH were raised in rabbits according to our standard, published protocol (11).
Insertion of the Ω-interposon into emm18 and transformation of type 18 streptococci. Insertional inactivation of emm18 was accomplished essentially as described previously for type 24 streptococci (6). Briefly, emm18 was ligated into pKK223-3 and then cut with XhoI, which recognizes a single site between bases 136 and 141 of the emm18 gene. The ends were repaired with a Klenow fragment (12). pBR322-ΩKm2 was digested with SmaI, and the ΩKm2 fragment was purified from a 0.8% agarose gel and ligated into the cut emm18 gene (12). The resulting plasmid, pKKM18Ω, was electroporated into type 18 streptococci as described (6). One kanamycin-resistant colony, designated M18Ω, was selected for further study.
Southern blots. Streptococcal chromosomal DNA was digested with BsaHI and electrophoresed in a 1% agarose gel. The DNA fragments were transferred to a nylon membrane (Sigma Chemical Co., St. Louis, Missouri, USA) and probed with digoxigenin-labeled ΩKm2 fragment or emm18 according to instructions provided by the manufacturer (Boehringer Mannheim Biochemicals, Indianapolis, Indiana, USA).
Western blots. Western blots were performed using extracts of intact streptococci and purified proteins as described previously (13). In some experiments using purified recombinant Enn18 protein, the nitrocellulose membranes were first incubated in buffer containing 10% normal human serum to block nonspecific binding of immunoglobulins.
Cloning and sequencing enn18. The type 18 strain (87-282) used for these studies contains an M-like gene, enn18, downstream from the emm18 gene (4). Although Enn18 has been shown to be expressed only in low levels by this strain of group A streptococci (14), we cloned the enn18 gene and purified the recombinant protein from extracts of E. coli. The purified Enn18 was used in subsequent experiments to determine whether Spa antibodies cross-reacted with this additional surface protein. The enn18 gene was initially cloned by PCR using a forward primer that copied the 3′ end of the emm18 gene and a reverse primer that was specific for the SF3 subfamily of emm-like genes (4). The purified PCR product was ligated into pKK223-3, and the insert was sequenced by standard methods (12). The intact enn18 gene was later cloned using a reverse primer that copied the 5′ end of the scpA gene (15), and the purified PCR product was ligated into pQE-30 (QIAGEN Inc., Chatsworth, California, USA). The recombinant Enn18 protein was purified by affinity chromatography over a nickel column according to the instructions provided by QIAGEN Inc.
Opsonization and bactericidal assays. In vitro opsonization assays and bactericidal assays were performed using nonimmune whole human blood as described previously (16). Briefly, rotation mixtures contained 0.1 mL of test antiserum, a standard inoculum of streptococci, and 0.4 mL of heparinized (10 U/mL) human blood. For opsonization assays, the number of streptococcal CFUs per leukocyte in the mixture was approximately 10. The tubes were rotated end-over-end for 45 minutes at 37°C; smears were then made on glass slides and stained with Wright’s stain. Opsonization was quantitated by counting 50 neutrophils per slide and calculating the percentage with associated streptococci (percent opsonization). The bactericidal activity of test antisera was determined using a similar assay except that fewer streptococci were added to the mixture, which was rotated for 3 hours at 37°C. At the end of the rotation, 0.1-mL aliquots were added to sheep blood agar, and pour plates were made to quantitate viable bacteria.
Determination of LD50. A comparison of the mouse virulence of the M18 parent strain and the M18Ω mutant was made after intraperitoneal challenge infections of Swiss white mice. Four groups of 6 mice each were challenged with 10-fold increasing inocula, ranging from 2.7 × 104 to 2.7 × 107 CFU of either M18 or M18Ω. Deaths were recorded for 7 days after challenge infections. The LD50 was determined by the method of Reed and Muench (17).
Passive mouse protection tests. Eight-week-old female white Swiss mice were given 0.5 mL of rabbit antiserum via the intraperitoneal route (2). After 24 hours, they were challenged using the same route with the indicated number of streptococci that had been grown to early log-phase in Todd-Hewitt broth. Deaths were recorded twice daily for 7 days.
Extraction and purification of Spa from pepsin extracts of M18Ω. Extraction of peptide fragments from the surface of M18Ω was accomplished using dilute solutions of pepsin at suboptimal pH, as described previously for the extraction of pep M proteins (18). The extract was precipitated in 60% saturated ammonium sulfate, dialyzed extensively against distilled water, and then lyophilized. The mixture of surface proteins and peptides is referred to as crude pep M18Ω.
Spa was identified in the crude pepsin extract by opsonization inhibition assays. The crude pep M18Ω was separated by SDS-PAGE on a preparative 10% gel using reducing conditions. The entire gel was electroblotted to nitrocellulose paper, each end was cut vertically, and then these pieces were stained Coomassie brilliant blue to identify the protein bands. The center section of the nitrocellulose paper was cut into horizontal strips approximately 8–10 mm wide. Each strip was then used to absorb rabbit antiserum raised against the crude pep M18Ω for 2 hours at 37°C. Opsonization assays were performed using absorbed serum samples and unabsorbed serum, as described above. Once the opsonic inhibitory peptide had been identified in the crude pepsin extract, it was further purified by preparative gel electrophoresis (Prep Cell 491; Bio-Rad Laboratories Inc., Hercules, California, USA) by methods described previously (10).
Amino acid sequence analyses of Spa. The purified Spa protein was electrophoretically transferred to a polyvinylidene difluoride (PVDF) membrane and submitted to the Protein and Nucleic Acid Facility (Beckman Center, Stanford University Medical Center, Stanford, California, USA) for NH2-terminal sequencing by Edman degradation. The sequence of an internal peptide of Spa was also determined in the Stanford University Facility. The intact protein on PVDF was digested with LysC protease (0.25 μmol/mL) at 37°C overnight. The resultant peptides were purified by HPLC on a Vydac C18 column. Selected fractions were assessed for purity by mass spectroscopy, and 1 peptide with a mass of 1.249 kDa was selected for NH2-terminal sequencing.
PCR, cloning, and sequencing a fragment of the spa18 gene. A fragment of the spa gene was amplified by PCR using degenerate inosine-containing oligonucleotide primers. The forward set of primers derived from the NH2-terminal sequence of Spa from amino acid residues 7–11 contained the sequence GAR GTI GCI GAY CC. The reverse primers, from the NH2-terminal sequence of the internal peptide, contained the sequences RTG IGA YTC RCT RTC and RTG RCT YTC IGA RTC. PCR was performed as described previously (10) using chromosomal DNA from type 18 streptococci as the template. The forward primer in combination with the second reverse primer listed above resulted in a single PCR product of 336 bp, which was ligated into pCR2.1-TOPO (Invitrogen Corp., San Diego, California, USA). DNA sequencing was performed by automated techniques in the University of Tennessee Molecular Resources Center using primers from the 5′ and 3′ flanking ends of the plasmid. The identity of the spa sequence was confirmed by comparing the translated DNA sequence to the amino acid sequence of Spa that was not used to construct the degenerate PCR primers.
An additional sequence of spa18 was obtained by methods reported previously by Evans et al. (19). The 336-bp PCR product obtained above was purified, biotinylated, and then bound to streptavidin-coated magnetic beads. Chromosomal DNA from type 18 streptococci was partially digested with Sau3A and then ligated to synthetic oligonucleotide pairs (19) containing each half of the Sau3A sequence. The ligation mixture and bound PCR target were melted and mixed to capture spa-containing chromosomal sequences. The magnetic beads were washed, and then PCRs were performed using an internal spa primer and each of the 4 extension primers. PCR products were cloned into pCR2.1-TOPO, which was used to transform E. coli. Plasmids from positive transformants were screened for inserts and sequenced by automated techniques. One ∼600-bp PCR product yielded an overlapping sequence plus 300 bases of a unique downstream sequence.
Characterization of the M-negative mutant (M18Ω). The new streptococcal protective antigen was discovered during studies of an M-negative mutant of type 18 streptococci. The M-negative M18 strain was constructed by interrupting the emm18 gene with an Ω element. Southern blot analyses of chromosomal DNA from M18 and M18Ω strains were probed with ΩKm2 and emm18 fragments and PCR analyses using primers from the emm gene and the Ω element; these probes revealed that there was a single copy of the Ω element inserted ∼140 bp beyond the start codon of the emm18 gene. In addition, PCR analyses using primers from the emm18 gene revealed that there was not another copy of the emm18 gene in the M18Ω strain (data not shown).
To confirm that the emm18 gene was not expressed in the M18Ω strain, immunoblots were performed using extracts of whole bacteria and rabbit antisera evoked by purified recombinant M18 (rM18), a synthetic peptide of M18, SM18(1-30), and synthetic peptide SM5(265-291), which copies the C-repeat domain of type 5 M protein that is common to all M proteins (1) (Figure 1). Rabbit antiserum against intact, recombinant type 18 M protein (anti-rM18) reacted with a triplet of proteins extracted from the parent strain, the largest of which had a molecular mass of ∼43 kDa (Figure 1, lane a). The immunoreactive proteins with lower molecular mass are presumably degradation products of M18 that were present in the crude extracts of whole bacteria. There was no reaction of rM18 antiserum with the extract from the M18Ω mutant (Figure 1, lane b). Anti–SM18(1-30) reacted only with proteins in extracts from the parent strain (Figure 1, lane c), and the pattern was identical to that observed with antiserum against rM18. Interestingly, antiserum against the C-repeat peptide of M protein, SM5(265-291), reacted only with an extract from the parent strain (Figure 1, lane e) and not with an extract from the mutant (Figure, lane f). This suggests that the mutant did not express another M protein or M-like protein containing these shared repeating epitopes. These data, taken together with the results of the Southern blot analyses and PCR, indicate that the Ω element was inserted in the emm18 gene and that the M18 protein was not expressed by M18Ω.
Western blot analysis of whole cell extracts of M18 (lanes a, c, and e) and M18Ω (lanes b, d, and f) reacted with rabbit antisera against rM18 (lanes a and b), SM18(1-30) (lanes c and d), and SM5(265-291) (lanes e and f). Coomassie blue–stained multiple proteins in extracts from both strains (M18, lane g; and M18Ω, lane h).
Growth in blood and mouse virulence of M18 and M18Ω. . The virulence of the M-negative mutant of type 18 streptococci was first assessed by its ability to grow in nonimmune human blood (Table 1). The parent strain of M18 grew to just over 8 generations after a 3-hour rotation in blood. The M18Ω mutant grew to approximately 7.5 generations in the same assay. The most stringent laboratory assay for virulence is intraperitoneal challenge infections of nonimmune mice. The LD50 of M18 and M18Ω were calculated after intraperitoneal injections of 10-fold increasing doses of each organism. The LD50 of the parent M18 strain was 0.73 × 105, and the LD50 of the M18Ω was 1.26 × 105. Of the 24 mice challenged with each organism, 8 challenged with M18 died, and 7 challenged with M18Ω died. Organisms recovered from the spleens of mice that succumbed to challenge infections with M18Ω were kanamycin resistant and M18 negative, indicating that there was no reversion to the parent phenotype in vivo. These results demonstrated that the expression of M18 was not required for virulence of type 18 streptococci.
Growth of M18 and M18Ω streptococci in human blood
Opsonization of M18 and M18Ω. Interruption of emm18 expression could have resulted in a virulent mutant that did not express a protective antigen, or alternatively, one that expressed a second protective antigen on its surface. In vitro opsonization experiments were performed to determine whether M18Ω expressed a surface antigen that contained opsonic epitopes (Table 2). Only the M18 parent strain was opsonized by antisera against SM18(1-30) or recombinant M18. The association of M18Ω with polymorphonuclear neutrophils remained at baseline levels in the presence of both of these antisera. Antiserum raised in rabbits against a crude pepsin extract of the M18Ω strain, however, opsonized both the parent and the M-negative mutant (Table 2). This antiserum was used in subsequent experiments to identify and purify the putative new protective antigen of type 18 streptococci.
Opsonization of M18 and M18Ω by rabbit antisera against recombinant M18, S-M18(1-30), and crude pep M18
Identification and purification of Spa. To identify the surface antigen of M18Ω containing opsonic epitopes, we performed opsonization inhibition experiments using fractions of the crude pepsin extract of M18Ω that had been separated by SDS-PAGE and then transferred to nitrocellulose paper. One section of the nitrocellulose paper was reacted with pep M18Ω antiserum to identify the immunoreactive bands, and another section was cut into horizontal strips that were used to absorb opsonic antibodies in the pep M18Ω antiserum. The pep M18Ω extract contained a protein with a molecular mass of 24 kDa (Spa) that absorbed the majority of the opsonic antibodies in the pep M18Ω antiserum (Figure 2).
Identification of Spa in a crude pepsin extract of M18Ω. The extract was separated by preparative PAGE, and the entire gel was transferred to nitrocellulose paper. Each bar, labeled 1–12, indicates the gel fraction that was used to absorb opsonic antibodies from a rabbit antiserum raised against a crude pepsin extract of M18Ω (see Methods). The unfractionated pepsin-extracted proteins from M18 and M18Ω (labeled pep M18 and pep M18Ω, respectively) at a concentration of 1 mg/mL inhibited opsonization by 70–80%. Absorption of the antiserum with gel fraction number 10, which contained a protein of approximately 24 kDa, resulted in a similar level of inhibition of opsonization.
Spa was purified from the crude pepsin extract of M18Ω by ammonium sulfate precipitation and preparative gel electrophoresis. The purified Spa migrated as a single band, as assessed by SDS-PAGE, with a molecular mass of 24 kDa (Figure 3b). Edman degradation of the purified protein revealed the sequence of the first 23 amino acids to be DSVSGLEVADPSDSKKLIELGLA. In addition, an internal peptide purified from a LysC digest of the intact Spa ontained the NH2-terminal sequence YRLDSESHLK.
PAGE of the purified fragment of Spa (lanes a and b) and Western blot analysis of the native protein extracted by phage lysin (lanes c and d). The crude pep M18Ω (lane a) and the purified Spa (lane b) were electrophoresed in a 10% polyacrylamide gel under reducing conditions and stained with Coomassie blue. The purified Spa migrated with a molecular mass of 24 kDa. Western blot analysis was performed to identify the native Spa that was released from the cell wall of M18Ω by group C streptococcal phage-associated lysin (19). The crude lysin extract contained multiple proteins that stained with Coomassie blue (lane c). Antiserum against the synthetic peptide of Spa reacted with a single protein in the extract with a molecular mass of 50 kDa (lane d).
Bactericidal activity of Spa antibodies. To directly assess the presence of bactericidal epitopes in Spa, rabbits were immunized with 100 μg of the purified protein in CFA at 0, 4, and 8 weeks. Antiserum obtained 10 weeks after the first injection contained antibodies that opsonized the M18 parent and M18Ω (Table 3). None of the Spa antisera cross-reacted with purified recombinant M18 or purified recombinant Enn18, as determined by ELISA (data not shown). The results of the opsonization assays were confirmed by indirect bactericidal tests (Table 4). The Spa antiserum was bactericidal against both M18 and M18Ω. Antiserum against rM18 was bactericidal against only the parent M18 strain and not against M18Ω (Table 4). A mixture of Spa and rM18 antisera resulted in complete killing of the M18 parent strain, suggesting that the 2 antisera have an additive bactericidal effect (Table 4).
Opsonization of type 18 streptococci by rabbit antisera against purified Spa and a synthetic peptide of Spa
Bactericidal activity of rabbit antisera against rM18 and purified Spa against M18 and M18Ω
Passive mouse protection tests with anti-Spa and anti-rM18. Passive mouse protection tests were performed to demonstrate the presence of protective epitopes within Spa. In both experiments, groups of 10 mice each were pretreated with either anti-Spa, anti-rM18, or an equal mixture of the 2 antisera, and then later challenged with M18 or M18Ω (Table 5). Of the mice challenged with the parent M18 strain, only 3 out of 20 control animals survived. Those pretreated with anti-rM18 or anti-Spa had higher rates of survival, but these were not statistically different from the control group. Interestingly, the mice that received a mixture of anti-Spa and anti-rM18 were significantly protected from challenge infections with the parent M18 strain (Table 5). Of the mice challenged with M18Ω, those that were pretreated with anti-Spa alone were significantly protected from death when compared with the control animals receiving normal rabbit serum (Table 5). Taken together, these results directly demonstrate the presence of protective epitopes within Spa.
Passive protection of mice against challenge infections with M18 or M18Ω by rabbit antisera against rM18 and Spa
Identification of opsonic epitopes within the NH2-terminus of Spa. To confirm the presence of opsonic epitopes within the covalent structure of Spa, a peptide [S-Spa18(1-23)C] copying the NH2-terminal 23 amino acids of the purified 24-kDa fragment was chemically synthesized. The peptide was covalently linked to KLH (11), and 3 rabbits were immunized with 100-μg doses using the same schedule described above. Serum from all 3 rabbits opsonized both the parent and M18Ω strains of group A streptococci (Table 3), confirming the presence of opsonic epitopes within this limited region of Spa. The antiserum against the synthetic peptide of Spa was also used to identify the native protein in phage lysin extracts of M18Ω (20). The S-Spa18(1-23)C antiserum reacted in the lysin extract with a single protein with a molecular mass of 50 kDa (Figure 3d), suggesting that the pepsin-derived Spa was a fragment of the larger native protein.
Opsonization of heterologous serotypes of group A streptococci by Spa antiserum. To determine whether Spa evoked opsonic antibodies against serotypes of group A streptococci other than type 18, opsonization assays were performed with antisera against purified Spa or against the pep M18Ω and a series of selected streptococci (Table 6). Both antisera opsonized type 3 and type 28 streptococci, in addition to type 18 organisms. Serotypes tested that were not opsonized by either antiserum were M1, M2, M5, M6, M13, M14, M19, and M24, all of which were from our laboratory collection. The anti–S-Spa18(1-23)C antisera did not opsonize types 3 or 28 streptococci, indicating that this limited region of Spa does not contain cross-opsonic epitopes.
Opsonization of heterologous serotypes of group A streptococci by rabbit antisera against purified Spa and a crude pepsin extract of M18Ω
Cloning and sequencing a fragment of the spa18 gene. A fragment of the spa18 gene was cloned using PCR and degenerate primers based on the NH2-terminal peptide sequence and the sequence of an internal peptide purified from a LysC digest of Spa. The PCR product was 336 bp in length, and the translated amino acid sequence contained residues 12–23 of the sequence derived from the Spa protein (Figure 4), confirming that the DNA was a fragment of spa18. An additional sequence of spa18 was obtained from an overlapping Sau3A fragment of chromosomal DNA that was captured using the original PCR fragment (Figure 4). A search of the current entries in GenBank and the type 1 streptococcal genome sequence database (http://dna1.chem.ou.edu/strep.html) revealed that the 636-bp spa18 gene fragment did not share sequence homology with any known proteins, suggesting that Spa is a new protective antigen of group A streptococci.
Partial DNA sequence of the spa gene and the deduced amino acid sequence. The solid underline indicates the amino acid sequence of purified Spa determined by Edman degradation. The unmarked DNA sequence (bases 1–336) was determined from the PCR product obtained using chromosomal DNA and degenerate oligonucleotide primers based on the NH2-terminal and internal amino acid sequences. The dashed line indicates the sequence derived from an overlapping DNA fragment that was captured from a Sau3A digest of chromosomal M18 DNA using the original 336-bp PCR fragment. The sequence has been submitted to GenBank and has the accession no. AF086813.
Previous studies have shown that the surface M proteins of group A streptococci play a dual role in the pathogenesis of infection. M proteins are a major determinant of virulence and confer upon group A streptococci the ability to resist phagocytosis and killing by neutrophils (2). In addition, M proteins elicit bactericidal antibodies thought to protect the host against subsequent infection by the same serotype (2). The antibodies with the greatest bactericidal activity are directed against the NH2-terminal, hypervariable epitopes of M proteins (21–24). Among the many defined surface proteins of group A streptococci, only antibodies against the M protein have been shown to be opsonic. In the present study, however, we have identified a new protective antigen, distinct from M protein, that evokes bactericidal antibodies.
The identification of Spa was facilitated by the fact that the M-negative mutant of type 18 streptococci was fully virulent when compared with the parent strain. This finding is somewhat contradictory to the recent results of Moses et al. (14), who showed reduced virulence of an M-negative mutant of the same M18 strain (87-282). They concluded that the hyaluronate capsule and the surface M protein played significant but unequal roles in resistance to phagocytosis and in mouse virulence. The identification of Spa in this strain raises the possibility that it too may play a role in virulence. This has not yet been tested directly, but experiments are now underway to create a Spa-negative/M-negative mutant that will be assessed for mouse virulence and its ability to grow in human blood.
The concept that group A streptococci may express more than 1 surface protein that contains protective epitopes is not new, but to date there has been insufficient data to prove the hypothesis. As early as 1939, Hirst and Lancefield (25) reported cross-protective antisera raised against crude extracts of different serotypes of group A streptococci. Wiley and Wilson (26) later reported the occurrence of 2 M antigens on the surface of certain strains of type 14 organisms. Other similar studies reported cross-protective immune responses evoked by various antigen preparations (27, 28). We now recognize that some M proteins contain cross-protective and type-specific epitopes (29). However, none of these earlier studies provided definitive evidence that any of the organisms actually expressed 2 distinct surface proteins containing protective epitopes.
A more recent study of cross-protective immunity by Chappell and Stuart (30) may be directly related to the protective epitopes of Spa. In these experiments, mice that were parentally immunized with an M-negative mutant of type 28 streptococci were protected against intravenous challenge infections with the type 28 parent and types 3 and 18 streptococci. In addition to the apparent non-M protein–mediated protective immune responses, the investigators also observed that the type 28 M-negative mutant was as virulent in mice as the M-positive type 3 and type 18 strains (30). Our finding that rabbit antisera against Spa opsonized all 3 of these serotypes suggests that the cross-protective immune responses observed by these investigators may have indeed been evoked by a Spa-like protein on the surface of type 28 streptococci.
In the present studies, we have shown that opsonic epitopes reside on both the M18 protein and Spa18. Both proteins evoked antibodies in rabbits that were bactericidal against the parent M18 strain in human blood. Although a greater number of mice survived when treated with anti-M18 or anti-Spa18 than when treated with preimmune serum, neither antiserum alone provided statistically significant protection against challenge infections of mice with the parent strain. Only the mice that received a mixture of the 2 antisera were protected to a statistically significant degree. Although these results suggest that both antigens may be required to evoke a fully protective immune response, we have not ruled out the possibility that our findings simply reflect the absolute quantity of antibodies against rM18 or Spa in each of the immune rabbit sera. Studies are currently in progress to quantitate the relative role of each antibody population in bactericidal killing.
The identification of Spa raises many questions that remain to be answered. Experiments to determine the complete sequence of the spa gene and its position in the streptococcal chromosome are currently in progress, as are experiments to determine the frequency of the gene and its expression by other serotypes of group A streptococci. Other important questions relate not only to the prevalence of spa within different strains of type 18 streptococci but also to the regulation of its expression. Type 18 streptococci have been identified as an important serotype in the resurgence of acute rheumatic fever in Utah (31). In addition, types 3 and 18, and to a lesser extent type 28, streptococci have all been implicated in the recent resurgence of serious streptococcal diseases (32). Differential expression of spa could potentially result in new subtypes of the organism with different virulence characteristics that may be partly defined by the immune status of the host and the population as a whole. The ability of an individual M serotype to alter the protective antigens expressed on its surface could potentially impact the changing epidemiology of group A streptococcal infections.
The discovery of a new protective antigen of group A streptococci has obvious implications for the development of vaccines that would prevent these infections. Our current approach is to use limited NH2-terminal fragments of M proteins that are linked in tandem to evoke protective immune responses against each serotype represented in the vaccine (10, 23, 24). This has the advantage of limiting the total amount of protein contained in the vaccine. However, because each NH2-terminal M protein fragment is type specific, this will likely necessitate the development of relatively complex vaccines to prevent the majority of streptococcal infections in a given population or geographic region. If Spa and Spa-like proteins are found to elicit significant cross-protection, this could facilitate the development of broadly protective vaccines that are considerably less complex than those presently in development.
To our knowledge, this is the first study to show that group A streptococci may express 2 distinct surface proteins that contain opsonic epitopes. Whether Spa is involved directly in the pathogenesis of streptococcal infections is not yet known. We believe that continued studies of Spa and related proteins that may be expressed by other serotypes of group A streptococci will provide new insights into the pathogenesis of these infections and the protective immune responses that they elicit.
The authors thank Valerie Long and Yi Li for expert technical assistance. The protein sequencing was performed by Dick Winant and Al Smith in the Protein and Nucleic Acid Facility at the Beckman Center for Molecular and Genetic Medicine, Stanford University. The authors also thank Philip Potter (St. Jude Children’s Research Hospital Memphis, Tennessee, USA) for advice on the design of degenerate PCR primers. This work was supported by research funds from the US Department of Veterans Affairs (to J.B. Dale and D.L. Hasty), the National Institutes of Health (grant AI-10085 to J.B. Dale), and ID Vaccine Corporation.
Fischetti, VA. Streptococcal M protein: molecular design and biological behavior. Clin Microbiol Rev 1989. 2:285-314.
Lancefield, RC. Current knowledge of the type-specific M antigens of group A streptococci. J Immunol 1962. 89:307-313.
Caparon, MG, Scott, JR. Identification of a gene that regulates expression of M protein, the major virulence determinant of group A streptococci. Proc Natl Acad Sci USA 1987. 84:8677-8681.
Hollingshead, SK, Readdy, TL, Yung, DL, Bessen, DE. Structural heterogeneity of the emm gene cluster in group A streptococci. Mol Microbiol 1993. 8:707-717.
Perez-Casal, J, Caparon, MG, Scott, JR. Introduction of the emm6 gene into and emm-deleted strain of Streptococcus pyogenes restores its ability to resist phagocytosis. Res Microbiol 1992. 143:549-558.
Courtney, HS, Bronze, MS, Dale, JB, Hasty, DL. Analysis of the role of M24 protein in streptococcal adhesion and colonization by use of omega-interposon mutagenesis. Infect Immun 1994. 62:4868-4873.
Podbielski, A, Schnitzler, N, Beyhs, P, Boyle, MDP. M-related protein (Mrp) contributes to group A streptococcal resistance to phagocytosis by human granulocytes. Mol Microbiol 1996. 19:429-441.
Dale, JB, Washburn, RG, Marques, MB, Wessels, MR. Hyaluronate capsule and surface M protein in resistance to phagocytosis of group A streptococci. Infect Immun 1996. 64:1495-1501.
Wessels, MR, Moses, AE, Goldberg, JB, DiCesare, TJ. Hyaluronic acid capsule is a virulence factor for mucoid group A streptococci. Proc Natl Acad Sci USA 1991. 88:8317-8321.
Dale, JB, Chiang, EY, Lederer, JW. Recombinant tetravalent group A streptococcal M protein vaccine. J Immunol 1993. 151:2188-2194.
Dale, JB, Beachey, EH. Localization of protective epitopes of the amino terminus of type 5 streptococcal M protein. J Exp Med 1986. 163:1191-1202.
Sambrook, J., Fritsch, E.F., and Maniatis, T. 1989. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory Press. Cold Spring Harbor, NY.
Dale, JB, Beachey, EH. Multiple, heart cross-reactive epitopes of streptococcal M proteins. J Exp Med 1985. 161:113-122.
Moses, AL, et al. Relative contributions of hyaluronic acid capsule and M protein to virulence in a mucoid strain of the group A streptococcus. Infect Immun 1997. 65:64-71.
Chen, C, Cleary, PP. Complete nucleotide sequence of the C5a peptidase gene of Streptococcus pyogenes. J Biol Chem 1990. 265:3161-3167.
Beachey, EH, et al. Purification and properties of M protein extracted from group A streptococci with pepsin: covalent structure of the amino terminal region of the type 24 M antigen. J Exp Med 1977. 145:1469-1481.
Reed, LJ, Muench, H. A simple method for estimating fifty percent endpoints. Am J Hyg 1938. 27:493-499.
Beachey, EH, Campbell, GL, Ofek, I. Peptic digestion of streptococcal M protein. II. Extraction of M antigen from group A streptococci with pepsin. Infect Immun 1974. 9:891-896.
Evans, DJ, Evans, DG, Lampert, HC, Nakano, H. Identification of four new prokaryotic bacterioferritins from Helicobacter pylori, Anabaena variabilis, Bacillus subtilis and Treponema pallidum, by analysis of gene sequences. Gene 1995. 153:123-127.
Zabriskie, JB, Freimer, EH. An immunological relationship between the group A streptococcus and mammalian muscle. J Exp Med 1966. 124:661-678.
Dale, JB, Seyer, JM, Beachey, EH. Type-specific immunogenicity of a chemically synthesized peptide fragment of type 5 streptococcal M protein. J Exp Med 1983. 158:1727-1732.
Jones, KF, Fischetti, VA. The importance of the location of antibody binding on the M6 protein for opsonization and phagocytosis of group A M6 streptococci. J Exp Med 1988. 167:1114-1123.
Dale, JB, Simmons, M, Chiang, EC, Chiang, EY. Recombinant, octavalent group A streptococcal M protein vaccine. Vaccine 1996. 14:944-948.
Dale, JB. Multivalent group A streptococcal vaccine designed to optimize the immunogenicity of six tandem M protein fragments. Vaccine 1999. 17:193-200.
Hirst, GK, Lancefield, RC. Antigenic properties of the type-specific substances derived from group A hemolytic streptococci. J Exp Med 1939. 69:425-438.
Wiley, GG, Wilson, AT. The occurrence of two M antigens in certain group A streptococci related to type 14. J Exp Med 1960. 113:451-460.
Wiley, GG, Bruno, PN. Cross-reaction among group A streptococci. I. Precipitin and bactericidal cross-reactions among types 33, 41, 43, 52, and Ross. J Exp Med 1968. 128:959-972.
Bergner-Rabinowitz, S, Ofek, I, Moody, MD. Cross-protection among serotypes of group A streptococci. J Infect Dis 1972. 125:339-347.
Dale, JB, Beachey, EH. Unique and common protective epitopes among different serotypes of group A streptococcal M proteins defined with hybridoma antibodies. Infect Immun 1984. 46:267-269.
Chappell, JD, Stuart, JG. Demonstration of protection in mice from a lethal challenge of three M serotypes of Streptococcus pyogenes using an M-negative vaccine. Vaccine 1993. 11:643-649.
Veasy, LG, et al. Resurgence of acute rheumatic fever in the intermountain area of the United States. N Engl J Med 1987. 316:421-427.
Johnson, DR, Stevens, DL, Kaplan, EL. Epidemiologic analysis of group A streptococcal serotypes associated with severe systemic infections, rheumatic fever, or uncomplicated pharyngitis. J Infect Dis 1992. 166:374-382.