点击显示 收起
Division of Infectious Diseases, Children's Hospital Boston, and Channing Laboratory, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts
The hemolytic zone that surrounds colonies on blood agar is a characteristic feature that is used in the preliminary identification of group A streptococcus (Streptococcus pyogenes or GAS) in clinical samples. In the 1930s, E. W. Todd distinguished 2 discrete toxins that accounted for the ability of GAS to lyse mammalian erythrocytes [1]. Streptolysin O (SLO) was the designation assigned to the oxygen-dependent labile hemolysin found in cell-free culture fluids during the growth of GAS in liquid cultures. A second cell-associated hemolysin was named streptolysin S (SLS), because its activity was stable to atmospheric oxygen. This latter toxin is responsible for the hemolytic zone around GAS colonies grown under routine aerobic conditions.
The first of these hemolysins, SLO, has been the subject of detailed investigation during the past 80 years. Its characterization was aided by the fact that GAS infection is associated with a brisk antibody response to SLOantibodies that could be quantified in relation to their capacity to neutralize the hemolytic activity of SLO. The measurement of anti-SLO titers (and, to a lesser extent, antibodies to other GAS antigens) became a mainstay of the serodiagnosis of the postinfectious syndromes acute rheumatic fever and poststreptococcal glomerulonephritis. The molecular characterization of SLO and of related toxins produced by other gram-positive bacteria led to the definition of a large family of proteins now known as cholesterol cytolysins [2]. Other well-known members of this family include perfringolysin O (produced by Clostridium perfringens), pneumolysin (produced by Streptococcus pneumoniae), and listeriolysin O (produced by Listeria monocytogenes). These toxins are secreted proteins that bind to cholesterol-containing cell membranes and oligomerize to create large pores. In addition to its direct membrane-damaging activity, SLO serves as a binding partner or conduit to deliver an associated toxin, NAD-glycohydrolase, to the cytoplasm of host cells [3]. The latter toxin enhances SLO-mediated cytotoxicity by inducing host-cell apoptosis and by blocking the internalization and killing of GAS bound to the cell surface [4].
A detailed understanding of SLS has been more elusive. Multiple attempts at purification were frustrated by the fact that the hemolytic activity of SLS is rapidly lost in the absence of bacterial cells or other stabilizers, such as albumin or the RNAase-resistant fraction of yeast RNA (RNA core). In contrast to SLO, SLS does not evoke an antibody response during natural infection that can be detected by the neutralization of hemolysis. The characterization of partially purified preparations has suggested that SLS is a protein or peptide of 2.8 kD and that it has potent toxic effects on experimental animals in vivo and on leukocytes in vitro [5, 6].
The molecular identity of SLS was revealed only recently through molecular genetic studies by De Azavedo, Nizet, and other researchers [7, 8]. Those investigators discovered a 9-gene operon in GAS, designated sag (for SLS-associated genes), that was necessary and sufficient for SLS production. The first gene, sagA, encodes a 53-aa prepropeptide with a Gly-Gly proteolytic cleavage site that has been predicted to release a 30-aa propeptide from the 23-aa leader sequence. A synthetic peptide corresponding to the propeptide sequence evoked antibodies that neutralized the hemolytic activity of SLS, a finding that strongly supports identification of sagA as the structural gene for SLS [9, 10]. The remaining genes in the operon have features consistent with export functions, posttranslational modification of the SLS peptide, and a possible immunity protein; this gene organization is typical of that for the bacteriocin family of antimicrobial peptides. The molecular size of the SLS peptide and its amino acid composition also resemble those of bacteriocins produced by other bacterial species. The relationship of SLS to bacteriocins suggests that its toxic effects on mammalian cells may represent collateral damage from a system that evolved to control competing bacterial flora.
SLS is produced by the vast majority of GAS strains, although nonhemolytic (SLS-negative) strains have been occasionally associated with human infection, which suggests that SLS is not essential for GAS virulence [11]. Several groups have investigated the contribution of SLS to GAS virulence in experimental infection models. Transposon Tn916 insertion in the sag operon promoter produced SLS-negative mutant strains in the background of M type 1 or 18 GAS strains. The mutants had reduced virulence in a soft-tissue infection model in mice, compared with their respective wild-type parent strains [7]. However, Sierig et al. [12] found that the deletion of sagA in an M type 3 strain background had no effect on virulence in a similar animal model. Similarly, a nonpolar deletion mutant of sagA in an M type 5 strain was fully virulent in an intraperitoneal challenge model in mice and only modestly attenuated in the subcutaneous soft-tissue infection model [13].
Another chapter in the evolving SLS story appears in this issue of The Journal of Infectious Diseases [14]. Miyoshi-Akiyama et al. compared 2 M3 strains of GAS, one of which produces a large amount of SLS in vitro and is highly virulent in mice; the other produces a smaller amount of SLS and is relatively avirulent. The injection of the live avirulent strain or heat-killed organisms of the virulent strain elicited neutrophil infiltration into the peritoneal cavity, whereas injection of the live virulent strain did not. Furthermore, coinjection of the virulent strain could suppress the neutrophil infiltration induced by thioglycolate. Because the suppressive effect was not conferred by culture supernatants, the investigators suspected that it was mediated by cell-associated SLS that actively destroyed neutrophils recruited to the site of infection. Inactivation of the sagA locus in either strain increased the influx of neutrophils, an effect that is partially reversed by genetic complementation in the avirulent strain background. Unfortunately, the sagA mutation was not complemented in the virulent strain background that had a more pronounced effect on neutrophil infiltration. The exposure of mouse neutrophils to the various GAS strains in vitro supported a cytotoxic and/or apoptotic effect of SLS on these cells.
The study by Miyoshi-Akiyama et al. adds to the considerable evidence that SLS damages host cells and contributes to the pathogenicity of GAS infection. It highlights in particular that the destruction of neutrophils may be a specific virulence mechanism that effectively inactivates the phagocytic cells that are primarily responsible for the ingestion and killing of GAS. A recent study of histopathologic findings in necrotizing fasciitis suggested that a paucity of polymorphonuclear leukocytes in the involved tissues is an unfavorable prognostic sign [15]. SLS is not the only GAS product that blocks the neutrophil response to infection: SLO is also cytotoxic in neutrophils [12], C5a peptidase inactivates the chemotactic complement protein C5a [16], and, similarly, an unidentified GAS protease degrades the chemokine interleukin 8, abrogating its neutrophil-recruiting activity [17]. Elaboration of all these factors may destroy neutrophils or suppress their recruitment, thereby enhancing the survival and proliferation of GAS at the infected site.
Despite the demonstrated destructive effect of SLS in neutrophils, the inactivation of sagA had a relatively small effect on lethality in mice, as was the case in most of the studies of the SLS mutants discussed above. These results serve to emphasize both the utility and the limitations of studying single-gene knockouts in small-animal models of GAS infection. Such studies provide important clues to pathogenic mechanisms but seldom reveal a clear picture of the relative importance in human disease pathogenesis of one of a multitude of GAS virulence determinants. The pathogenesis of GAS infection involves the multifaceted interplay between a large number of bacterial products and the human host. The fact that most human encounters with GAS result in asymptomatic or self-limited infections is testimony to the delicate balance between the pathogenic effects of these bacterial virulence factors and host defenses. Fortunately, it is a relatively rare event when the balance is seriously upset and disease progresses to the potentially catastrophic consequences of invasive infection.
References
1. Todd EW. The differentiation of two distinct serologic varieties of streptolysin, streptolysin O and streptolysin S. J Pathol Bacteriol 1938; 47:42345. First citation in article
2. Tweten RK, Parker MW, Johnson AE. The cholesterol-dependent cytolysins. Curr Top Microbiol Immunol 2001; 257:1533. First citation in article
3. Madden JC, Ruiz N, Caparon M. Cytolysin-mediated translocation (CMT): a functional equivalent of type III secretion in gram-positive bacteria. Cell 2001; 104:14352. First citation in article
4. Bricker AL, Cywes C, Ashbaugh CD, Wessels MR. NAD+-glycohydrolase acts as an intracellular toxin to enhance the extracellular survival of group A streptococci. Mol Microbiol 2002; 44:25769. First citation in article
5. Bernheimer AW. Physical behavior of streptolysin S. J Bacteriol 1967; 93:20245. First citation in article
6. Ofek I, Bergner-Rabinowitz S, Ginsberg I. Oxygen-stable hemolysins of group A streptococci. 8. Leukotoxic and antiphagocytic effects of streptolysins S and O. Infect Immun 1972; 6:45964. First citation in article
7. Betschel SD, Borgia SM, Barg NL, Low DE, De Azavedo JC. Reduced virulence of group A streptococcal Tn916 mutants that do not produce streptolysin S. Infect Immun 1998; 66:16719. First citation in article
8. Nizet V, Beall B, Bast DJ, et al. Genetic locus for streptolysin S production by group A Streptococcus. Infect Immun 2000; 68:424554. First citation in article
9. Carr A, Sledjeski DD, Podbielski A, Boyle MD, Kreikemeyer B. Similarities between complement-mediated and streptolysin S-mediated hemolysis. J Biol Chem 2001; 276:417906. First citation in article
10. Dale JB, Chiang EY, Hasty DL, Courtney HS. Antibodies against a synthetic peptide of SagA neutralize the cytolytic activity of streptolysin S from group A streptococci. Infect Immun 2002; 70:216670. First citation in article
11. James L, McFarland RB. An epidemic of pharyngitis due to a nonhemolytic group A Streptococcus at Lowry Air Force Base. N Engl J Med 1971; 284:7502. First citation in article
12. Sierig G, Cywes C, Wessels MR, Ashbaugh CD. Cytotoxic effects of streptolysin O and streptolysin S enhance the virulence of poorly encapsulated group A streptococci. Infect Immun 2003; 71:44655. First citation in article
13. Fontaine MC, Lee JJ, Kehoe MA. Combined contributions of streptolysin O and streptolysin S to virulence of serotype M5 Streptococcus pyogenes strain Manfredo. Infect Immun 2003; 71:385765. First citation in article
14. Miyoshi-Akiyama T, Takamatsu D, Koyanagi M, Zhao J, Imanishi K, Uchiyama T. Cytocidal effect of Streptococcus pyogenes on mouse neutrophils in vivo and the critical role of streptolysin S. J Infect Dis 2005; 192:10716 (in this issue). First citation in article
15. Bakleh M, Wold LE, Mandrekar JN, Harmsen WS, Dimashkieh HH, Baddour LM. Correlation of histopathologic findings with clinical outcome in necrotizing fasciitis. Clin Infect Dis 2005; 40:4104. First citation in article
16. Wexler DE, Chenoweth ED, Cleary PP. Mechanism of action of the group A streptococcal C5a inactivator. Proc Natl Acad Sci USA 1985; 82:81448. First citation in article
17. Hidalgo-Grass C, Dan-Goor M, Maly A, et al. Effect of a bacterial pheromone peptide on host chemokine degradation in group A streptococcal necrotising soft-tissue infections. Lancet 2004; 363:696703. First citation in article