Search

The attenuated expression of virulence genes found in a group A streptococcal strain that is naturally pathogenic for mice was postulated to result from a defect in the strain's multigene regulator, Mga. The sequence of the mga gene reveals three amino acid changes in the gene product that might affect protein function. The defect in the mga gene was complemented by providing either the closely similar mga4 allele or a more divergent mga1 allele in trans. Complementation increased the amount of emm50 transcript and the quantity of surface-extractable M protein, restoring virulence function.

In group A streptococci, genetic markers for principal tissue reservoir are located within emm genes, which encode surface proteins that have a role in virulence. A worldwide collection of 160 isolates was evaluated for two traits: chromosomal emm gene markers for tissue tropism (designated patterns A-E), and bacteriophage-associated genes (speA and speC) encoding pyrogenic exotoxins. The speA and speC alleles of organisms harboring the emm marker for a pharyngeal reservoir (pattern A-C) differ from spe alleles that predominate in organisms with the emm marker for impetigo (pattern D). However, organisms that display the emm marker for both tissue sites (pattern E) are not intermediate for the distribution of either speA or speC alleles, but instead resemble pattern A-C isolates for speA and pattern D strains for speC. Statistically significant nonrandom associations between exotoxin alleles and emm patterns were observed but cannot be readily explained by niche separation alone.

To investigate the role of putative virulence factors of Streptococcus pyogenes (group A streptococcus; GAS) in causing disease, we introduced specific mutations in GAS strain B514, a natural mouse pathogen, and tested the mutant strains in two models of infection. To study late stages of disease, we used our previously described mouse model (C3HeB/FeJ mice) in which pneumonia and systemic spread of the streptococcus follow intratracheal inoculation. To study the early stages of disease, we report here a model of long-term (at least 21 days) throat colonization following intranasal inoculation of C57BL/10SnJ mice. When the three emm family genes of GAS strain B514-Sm were deleted, the mutant showed no significant difference from the wild type in induction of long-term throat colonization or pneumonia. We inactivated the scpA gene, which encodes a complement C5a peptidase, by insertion of a nonreplicative plasmid and found no significant difference from the wild type in the incidence of throat colonization. However, there was a small but statistically significant decrease in the incidence of pneumonia caused by the scpA mutant. Finally, we demonstrated a very important effect of the hyaluronic acid capsule in both models. Following intranasal inoculation of mice with a mutant in which a nonreplicative plasmid was inserted into the hasA gene, which encodes hyaluronate synthase, we found that all bacteria recovered from the throats of the mice were encapsulated revertants. Following intratracheal inoculation with the hasA mutant, the incidence of pneumonia within 72 h was significantly reduced from that of the control strain (P = 0.006). These results indicate that the hyaluronic acid capsule of S. pyogenes B514 confers an important selective advantage for survival of the bacteria in the upper respiratory tract and is also an important determinant in induction of pneumonia in our model system.

In summary, the emm chromosomal patterns distinguish between the two principal tissue site reservoirs of group A streptococci--the nasopharyngeal mucosa and impetigo lesion. Strains derived from normally sterile tissue sites are probably transmitted to new hosts by respiratory droplets, at least in the Connecticut population. The speA gene provides an example of how genetic exchange between different strains of group A streptococci may be limited to a single tissue site or to a subset of emm chromosomal patterns.

The strain B514, an M serotype 50 strain, is capable of causing a natural upper respiratory infection leading to death in mice, as reported by Hook et al. in 1960 (E. W. Hook, R. R. Wagner, and R. C. Lancefield, Am. J. Hyg. 72:111-119, 1960). Thus, this strain was of interest for use in developing an animal model for group A streptococcal colonization and disease. The emm gene cluster for this strain was examined by PCR mapping and found to contain three emm family genes and cluster pattern 5. PCR-generated fragments corresponding to the SF4 (mrp50), SF2 (emmL50), and SF3 (enn50) genes were cloned and the entire gene cluster was sequenced. The gene cluster has greater than 97% DNA identity to previously sequenced regions of the gene cluster of the M2 strain T2/44/RB4 if two small divergent regions that encode the mature amino terminus of the SF-2 and SF-3 gene products are not included. If expressed, the genes encode proteins which bind human immunoglobulin G (Mrp50 and EmmL50) or immunoglobulin A (Enn50). However, in isolates taken directly after passage in mice, the surface proteins arising from these genes were barely detectable. The transcription of each gene in the B514 strain was investigated by Northern (RNA) hybridization, and mRNA transcripts were detected and quantitated relative to those of the recA gene, a housekeeping gene. Transcription of all three emm family genes was found to be over 30-fold attenuated relative to transcription of the same genes in strain T2/44/RB4. This suggests that the positive regulator, Mga, either is not expressed in this strain or has a different requirement for activation; it also suggests that the capsule may be sufficient to inhibit phagocytosis under these circumstances.

The existence of discrete populations of throat and skin strains of group A streptococci has long been recognized; however, a molecular basis for this distinction is not known. The emm gene structure was analyzed for 105 isolates obtained from patients with well-defined group A streptococcal diseases: uncomplicated pharyngitis, impetigo, and acute rheumatic fever. Four emm gene sub-family forms, defined by nucleotide sequence differences in regions encoding the peptidoglycan-spanning domain of M and M-like surface proteins, were found to exist in five different chromosomal patterns among naturally occurring isolates. Strong correlations were made between disease and the number and arrangement of emm subfamily genes. These findings provide a genetic basis for the historical references to "throat," "skin," and "rheumatogenic" types.

To facilitate future genetic studies with Streptococcus pyogenes (Sp), a recA mutant (Rec11) was constructed using a streptococcal integration vector carrying a PCR-derived internal recA fragment. The insertion of the plasmid in the mutant chromosome was identified by Southern hybridization. Resistance to UV and the ability to accept linear DNA transformation by Rec11 were greatly decreased, confirming its RecA phenotype. Using the PCR-derived fragment as a probe, we cloned and sequenced the complete Sp recA gene, which is highly homologous to the recA of S. pneumoniae and Lactococcus lactis.

Group A streptococci have a virulence regulon containing a single emm locus or two or three distinct and adjacent loci of structurally related emm family genes. The products of the emm gene cluster consist of fibrillar surface proteins, at least some of which are known to contain determinants of type specificity located in their NH2-terminal regions, lying distal to the cell surface. The emm genes can be categorized into four major subfamilies (SFs), based on structural differences within their 3' regions encoding the peptidoglycan-spanning domain. In this study, we investigate the polymorphism within the 5' region of SF-4 and SF-3 emm genes (which occupy the first and last emm positions of the gene cluster, respectively) in 22 strains representing different serotypes. Our findings indicate that unlike the centrally positioned SF-1 or SF-2 genes, SF-3 and SF-4 genes each display only limited polymorphism in their 5' regions, suggesting that their gene products may not be major contributors to type specificity. Two forms of the SF-3 gene (SF3a, SF3b) and two forms of the SF-4 gene (SF4a, SF4b) are found to exist in all four possible combinations (SF3aSF4a, SF3aSF4b, SF3bSF4a, SF3bSF4b), strongly suggesting that horizontal gene spread has contributed to the evolution of emm genes and to the generation of emm gene diversity in group A streptococci.

The emm genes are members of a gene family in group A streptococci (GAS) that encode for antiphagocytic cell-surface proteins and/or immunoglobulin-binding proteins. Previously sequenced genes in this family have been named "emm," "fcrA," "enn," "arp," "protH," and "mrp"; herein they will be referred to as the "emm gene family." The genes in the emm family are located in a cluster occupying 3-6 kb between the genes mry and scpA on the chromosome of Streptococcus pyogenes. Most GAS strains contain one to three tandemly arranged copies of emm-family genes in the cluster, but the alleles within the cluster vary among different strains. Phylogenetic analysis of the conserved sequences at the 3' end of these genes differentiates all known members of this family into four evolutionarily distinct emm subfamilies. As a starting point to analyze how the different subfamilies are related evolutionarily, the structure of the emm chromosomal region was mapped in a number of diverse GAS strains by using subfamily-specific primers in the polymerase chain reaction. Nine distinct chromosomal patterns of the genes in the emm gene cluster were found. These nine chromosomal patterns support a model for the evolution of the emm gene family in which gene duplication followed by sequence divergence resulted in the generation of four major-gene subfamilies in this locus.

As the initial step in infection, group A streptococci (GAS) colonize either the nasopharyngeal mucosa or the skin of humans. A number of virulence factors have been implicated in the colonization phase of pathogenesis based upon their in vitro activities, but the in vivo data supporting their role in colonization of the host tissues is lacking. In this investigation, the potential requirement for M protein in pharyngeal colonization by GAS was explored by using near-isogenic strains in experimental animals studies. Fischer rats were infected by intranasal and oral inoculation with both M-positive and M-negative Streptococcus pyogenes strains. Colonization of the pharyngeal area by the streptococci was monitored at various time intervals. Both M-positive and M-negative strains colonized during the first week following infection, indicating that M protein was not necessary for this initial colonization. Two M-positive strains of S. pyogenes were recovered from the rats up to 23 weeks following inoculation, while the colonization levels for M-negative strains decreased rapidly in the second and third weeks, becoming negligible by the fourth week. This indicates a potential role for M protein in the persistence of colonization at this mucosal surface. Colonization of rats with either M-positive strain of S. pyogenes also resulted in the appearance of salivary and serum antibody responses. This in vivo model should allow further investigation into factors required for GAS disease, including the examination of the potential role of the host immune response both in modulation of the pharyngeal surface and in modulation of antigenic changes in M protein or other surface factors.

One or more distinct copies of emm genes lie within a gene cluster that is located downstream from a transcriptional regulatory gene (mry). Mry is a positive regulator for the genes in this cluster and for the downstream gene, scpA. The objective of this study is to examine the structure of this cluster and the distribution of specific alleles within the cluster among group A streptococcal isolates of 32 different serotypes. The peptidoglycan (PG)-spanning domain, which exists in four divergent forms, was used to identify specific alleles of the genes within the emm cluster. Gene content of the cluster was determined by Southern hybridization with allele-specific oligonucleotides. Five different chromosomal patterns for this cluster were observed. Sequence heterogeneity in the adjacent mry locus was demonstrated by the ability of some of the isolates to hybridize with a whole mry gene probe, but not with mry-based oligonucleotide probes. A PCR-based chromosomal mapping technique was used to examine further the gene order within the emm gene clusters. Structural heterogeneity of the emm gene cluster was found within class I isolates in this study, while class II isolates were relatively homogeneous at this chromosomal locus and distinct from class I.

The M protein of group A streptococci is considered one of the primary virulence determinants of this organism because it prevents phagocytosis. All of the 56 different M protein types of group A streptococci tested contain a DNA region that hybridizes with a labeled probe consisting of DNA encoding the type 6 M protein (emm6 gene). In addition, 2 of 3 functionally M- group A strains (strains that are successfully phagocytized) also contain this gene, but this DNA is not present in 1 M- strain. Of the other streptococcal groups, only groups C and G, which are closely related to group A, contain DNA that hybridizes with emm6. None of the non-streptococcal Gram-positive bacteria tested contains emm6-related DNA. In the strain from which emm6 was cloned, there appears to be only a single copy of this gene.

M protein, a major surface protein and virulence factor for the group A streptococcus, exhibits extraordinary size variation in strains of the same serotype (Fischetti et al. 1985). RNA sequence analysis of spontaneous M protein size variants shows that deletion mutations arise in a single strain by homologous recombination events between intragenic tandem repeats. Similar deletion and duplication events also occur in serial streptococcal isolates from a single patient and among related strains in a recent outbreak. We discuss how homologous recombination events can lead to the generation of antigenic variation.

DNA hybridization with probes from the structural gene for the M6 protein shows that the carboxy-terminal region (proximal to the cell surface) is conserved among strains of different M serotypes. The amino-terminal region, which is more available to immunological surveillance, exhibits greater variation.

In studies primarily confined to the amino-terminal region of the fibrillar group A streptococcal M protein, only limited immunological crossreactions have been observed among M serotypes. In this investigation, two monoclonal antibodies generated against nearly the entire M6 molecule (LysM6) were used to determine the extent of crossreactions among M serotyping strains and to localize their epitopes on the M molecule. Colony blot and immunoblot analyses revealed that an epitope responsible for crossreactions among 5 of the 56 strains of different M serotypes tested is located in the amino-terminal half of the molecule, distal to the cell surface. In contrast, a more common crossreactive epitope, reacting with 20 of the 56 strains, is located near the middle of the M molecule. These studies also reveal that the more conserved determinant, located more proximally to the cell surface, is accessible to the immune system, even on the whole organism, and, thus, may be useful in devising means to protect against infections by multiple group A streptococcal M serotypes.

Deletions of highly, but not entirely, homologous intragenic sequence repeats result in amino acid sequence and conformational changes in the M proteins of spontaneous M protein-size variants of group A streptococci. To determine if antigenic changes occurred as a result of these deletion mutations, monoclonal and polyclonal antibodies with defined epitopes were used in competition assays. Competing antigens were either purified pepsin-derived fragments (representing the amino-terminal half of the molecule) of parent and mutant M proteins or were intact bacterial cells. These assays showed that antigenic variation occurred at the site(s) of these deletions but not at adjacent or distant epitopes. Once cleaved from the bacterium by pepsin, the M molecules also underwent conformational changes, which were reflected in their ability to compete. A monoclonal antibody opsonic for M6 streptococci lost its ability to completely opsonize one of the size mutants in this study. Therefore, spontaneous intragenic events between repeats within emm-6, the structural gene for the M6 protein, do result in structural variations within the mutant M molecules. This variation alters the ability of certain antibodies, originally produced in response to sequences in the parental M molecule, to bind to the mutant M molecules or opsonize the mutant organisms. Group A streptococci have evolved a mechanism for generating antigenic diversity that differs from currently known mechanisms in other bacterial species.

Streptococcal M protein is a coiled-coil fibrillar structure extending about 60 nm from the cell wall. From DNA sequencing of the M6 protein gene, it has been determined that the C-terminal end contains a membrane anchor and an adjacent cell wall stabilization domain, both of which are similar to C-terminal regions of surface proteins in other gram-positive organisms. Studies with monoclonal antibodies revealed that the C-terminal half of the protein is conserved among M proteins of different serotypes, whereas the N-terminal half varies. The M protein contains tandem repeats, which, through homologous recombination, are responsible for the observed size variation of the M proteins from different streptococcal strains. M protein size mutants occur in a laboratory-grown culture at a frequency of 1/2,000 colony-forming units. DNA sequence analysis of the M gene in size mutants derived from a single strain suggests that sequence changes which result from homologous recombination may play a role in the antigenic variation of M protein.

The DNA sequence of the gene for type 6 M protein of Streptococcus pyogenes contains two extended tandem repeat regions and one nontandem repeat region. We suggest that the duplication and deletion of these repeats generates the observed diversity in size and sequence among the family of M proteins in the group A streptococci. In addition, the DNA sequence reveals the presence of a 42-amino-acid signal peptide, a region rich in proline that is thought to be located in the cell wall, and a membrane anchor sequence at the carboxyl-terminal end of the protein. Signals similar to the consensus sequences recognized for the initiation of transcription and translation in Gram-positive bacteria have been identified in the DNA sequence. Codon usage is similar to that of other Gram-positive bacteria and significantly different from that of Escherichia coli.

In group A streptococcal strains of 10 different serotypes, the sequence previously identified as homologous to the structural gene for type 6 M protein (emm6) was found to be transcribed. The conserved sequence, which shows greater than 95% homology among heterologous M proteins, was identified as the 3' third of the gene.