Cytotoxic Effects of Streptolysin O and Streptolysin S Enhance the Virulence of Poorly Encapsulated Group A Streptococci
Channing Laboratory and Division of Infectious Diseases, Brigham and Women's Hospital,1 Division of Infectious Diseases, Children's Hospital Boston, Harvard Medical School, Boston, Massachusetts 021152z]|@?5, 百拇医药
Received 4 June 2002/ Returned for modification 8 August 2002/ Accepted 15 October 2002z]|@?5, 百拇医药
ABSTRACTz]|@?5, 百拇医药
Although the toxicity of streptolysin O (SLO) and streptolysin S (SLS) in purified group A streptococci (GAS) has been established, the effect of these molecules in natural infection is not well understood. To identify whether biologically relevant concentrations of SLO and SLS were cytotoxic to epithelial and phagocytic cells that the bacteria would typically encounter during human infection and to characterize the influence of cell injury on bacterial pathogenesis, we derived GAS strains deficient in SLO or SLS in the background of an invasive GAS M3 isolate and determined their virulence in in vitro and in vivo models of human disease. Whereas bacterial production of SLO resulted in lysis of both human keratinocytes and polymorphonuclear leukocytes, GAS expression of SLS was associated only with keratinocyte injury. Expression of SLO but not SLS impaired polymorphonuclear leukocyte killing of GAS in vitro, but this effect could only be demonstrated in the background of acapsular organisms. In mouse invasive soft-tissue infection, neither SLO or SLS expression significantly influenced mouse survival. By contrast, in a mouse model of bacterial sepsis after intraperitoneal inoculation of GAS, SLO expression enhanced the virulence of both encapsulated and acapsular GAS, whereas SLS expression increased the virulence only of acapsular GAS. We conclude that the cytotoxic effects of SLO protect GAS from phagocytic killing and enhance bacterial virulence, particularly of strains that may be relatively deficient in hyaluronic acid capsule. Compared to SLO, SLS in this strain background has a more modest influence on GAS pathogenicity and the effect does not appear to involve bacterial resistance to phagocytosis.
INTRODUCTIONf-cq, 百拇医药
Group A streptococci (GAS) are important human pathogens. Millions of children each year develop GAS pharyngitis. Although throat infection is typically benign, it can be complicated by the development of acute and chronic rheumatic heart disease. Invasive GAS infection occurs less frequently than pharyngitis but causes significantly more morbidity. The incidence of severe invasive GAS disease associated with shock and organ failure appears to be increasing (14, 38, 40).f-cq, 百拇医药
GAS produce two hemolysins that may contribute to pathogenesis. Streptolysin O (SLO) is a well-characterized oxygen-labile prototype of a cholesterol-binding bacterial exotoxin. When cultured in broth, GAS express SLO during exponential-phase and early stationary-phase growth (1). Streptolysin S (SLS) is an oxygen-stable oligopeptide primarily responsible for the characteristic zone of beta-hemolysis surrounding GAS colonies grown on blood-agar medium (17). SLS production occurs when cells are in stationary-phase growth conditions in broth culture (9).
Early studies of purified SLO and SLS demonstrated that these hemolysins were toxic to a variety of human cells in vitro and in vivo (3, 10, 16, 19, 20). In rabbits, injection of SLO caused blood vessel contraction, increased capillary permeability, massive intravascular thrombosis, dermal necrosis, cardiotoxicity, and death (1). Intravenous injection of SLS in rabbits resulted in massive intravascular hemolysis and death (15). The severity of injury in these studies suggests that the injected concentrations of SLO and SLS may have exceeded those achieved during natural infection.pu]-*, 百拇医药
More recently, investigators have attempted to assess the effects of biologically relevant concentrations of SLO and SLS in GAS pathogenesis by determining the virulence of hemolysin-deficient mutants in animal models of human infection. Limbago and colleagues reported that SLO-deficient GAS injected subcutaneously in mice were attenuated (24). Similarly, Betschel et al. noted that SLS-deficient GAS subcutaneously injected in mice were less virulent than the wild-type parent strain (11).
To characterize further the pathogenic effects associated with bacterial expression of SLO and SLS, we derived GAS deficient in the production of SLO or SLS in the background of an invasive GAS M3 isolate and used the strains in models of human infection. Our results demonstrate that GAS expression of SLO and, to a lesser extent, of SLS is associated with human keratinocyte and polymorphonuclear leukocyte (PMN) injury in vitro and, particularly in the background of acapsular GAS strains, enhances bacterial virulence in murine models of invasive infection.k)elgux, 百拇医药
MATERIALS AND METHODSk)elgux, 百拇医药
Bacterial strains and growth conditions. GAS strain 950771 is a moderately encapsulated M-type 3 invasive isolate from a child with necrotizing fasciitis and sepsis. Strain 950771Sm is a spontaneous streptomycin-resistant variant of 950771. GAS strain 950771hasA is an acapsular derivative of 950771 (6). GAS strain 950771SLO is an SLO-deficient derivative of 950771Sm (13). GAS strains were grown in Todd-Hewitt broth (Difco Laboratories, Detroit, Mich.) supplemented with yeast extract (Difco) to a final concentration of 2.5% (wt/vol) (THY broth) or, for SLS-hemolysis assays, in brain-heart infusion (BHI) broth (Difco) supplemented with 2% (wt/vol) sodium bicarbonate and 1% (wt/vol) maltose. GAS were grown on either Trypticase-soy blood agar medium (Becton Dickinson Microbiological Systems, Franklin Lakes, N.J.) (TSA-blood) or THY agar supplemented with 5% (vol/vol) defibrinated sheep blood (PML Microbiologicals, Inc., Wilsonville, Oreg.) (THY-blood). When required, THY-blood medium was supplemented with 1.0 µg of erythromycin per ml. GAS were cultured at 37°C in the presence of 5% CO2.k)elgux, 百拇医药
DNA manipulation. Restriction endonuclease digests, DNA ligations, transformation of Escherichia coli, agarose gel electrophoresis, and Southern hybridization analysis were performed according to standard techniques (8). Restriction enzymes and DNA ligase were purchased commercially (Invitrogen, Carlsbad, Calif.). Plasmid DNA was recovered from E. coli strain DH5(Gabriele Sierig,Colette Cywes Michael R. Wessels and Cameron D. Ashbaugh)
Received 4 June 2002/ Returned for modification 8 August 2002/ Accepted 15 October 2002z]|@?5, 百拇医药
ABSTRACTz]|@?5, 百拇医药
Although the toxicity of streptolysin O (SLO) and streptolysin S (SLS) in purified group A streptococci (GAS) has been established, the effect of these molecules in natural infection is not well understood. To identify whether biologically relevant concentrations of SLO and SLS were cytotoxic to epithelial and phagocytic cells that the bacteria would typically encounter during human infection and to characterize the influence of cell injury on bacterial pathogenesis, we derived GAS strains deficient in SLO or SLS in the background of an invasive GAS M3 isolate and determined their virulence in in vitro and in vivo models of human disease. Whereas bacterial production of SLO resulted in lysis of both human keratinocytes and polymorphonuclear leukocytes, GAS expression of SLS was associated only with keratinocyte injury. Expression of SLO but not SLS impaired polymorphonuclear leukocyte killing of GAS in vitro, but this effect could only be demonstrated in the background of acapsular organisms. In mouse invasive soft-tissue infection, neither SLO or SLS expression significantly influenced mouse survival. By contrast, in a mouse model of bacterial sepsis after intraperitoneal inoculation of GAS, SLO expression enhanced the virulence of both encapsulated and acapsular GAS, whereas SLS expression increased the virulence only of acapsular GAS. We conclude that the cytotoxic effects of SLO protect GAS from phagocytic killing and enhance bacterial virulence, particularly of strains that may be relatively deficient in hyaluronic acid capsule. Compared to SLO, SLS in this strain background has a more modest influence on GAS pathogenicity and the effect does not appear to involve bacterial resistance to phagocytosis.
INTRODUCTIONf-cq, 百拇医药
Group A streptococci (GAS) are important human pathogens. Millions of children each year develop GAS pharyngitis. Although throat infection is typically benign, it can be complicated by the development of acute and chronic rheumatic heart disease. Invasive GAS infection occurs less frequently than pharyngitis but causes significantly more morbidity. The incidence of severe invasive GAS disease associated with shock and organ failure appears to be increasing (14, 38, 40).f-cq, 百拇医药
GAS produce two hemolysins that may contribute to pathogenesis. Streptolysin O (SLO) is a well-characterized oxygen-labile prototype of a cholesterol-binding bacterial exotoxin. When cultured in broth, GAS express SLO during exponential-phase and early stationary-phase growth (1). Streptolysin S (SLS) is an oxygen-stable oligopeptide primarily responsible for the characteristic zone of beta-hemolysis surrounding GAS colonies grown on blood-agar medium (17). SLS production occurs when cells are in stationary-phase growth conditions in broth culture (9).
Early studies of purified SLO and SLS demonstrated that these hemolysins were toxic to a variety of human cells in vitro and in vivo (3, 10, 16, 19, 20). In rabbits, injection of SLO caused blood vessel contraction, increased capillary permeability, massive intravascular thrombosis, dermal necrosis, cardiotoxicity, and death (1). Intravenous injection of SLS in rabbits resulted in massive intravascular hemolysis and death (15). The severity of injury in these studies suggests that the injected concentrations of SLO and SLS may have exceeded those achieved during natural infection.pu]-*, 百拇医药
More recently, investigators have attempted to assess the effects of biologically relevant concentrations of SLO and SLS in GAS pathogenesis by determining the virulence of hemolysin-deficient mutants in animal models of human infection. Limbago and colleagues reported that SLO-deficient GAS injected subcutaneously in mice were attenuated (24). Similarly, Betschel et al. noted that SLS-deficient GAS subcutaneously injected in mice were less virulent than the wild-type parent strain (11).
To characterize further the pathogenic effects associated with bacterial expression of SLO and SLS, we derived GAS deficient in the production of SLO or SLS in the background of an invasive GAS M3 isolate and used the strains in models of human infection. Our results demonstrate that GAS expression of SLO and, to a lesser extent, of SLS is associated with human keratinocyte and polymorphonuclear leukocyte (PMN) injury in vitro and, particularly in the background of acapsular GAS strains, enhances bacterial virulence in murine models of invasive infection.k)elgux, 百拇医药
MATERIALS AND METHODSk)elgux, 百拇医药
Bacterial strains and growth conditions. GAS strain 950771 is a moderately encapsulated M-type 3 invasive isolate from a child with necrotizing fasciitis and sepsis. Strain 950771Sm is a spontaneous streptomycin-resistant variant of 950771. GAS strain 950771hasA is an acapsular derivative of 950771 (6). GAS strain 950771SLO is an SLO-deficient derivative of 950771Sm (13). GAS strains were grown in Todd-Hewitt broth (Difco Laboratories, Detroit, Mich.) supplemented with yeast extract (Difco) to a final concentration of 2.5% (wt/vol) (THY broth) or, for SLS-hemolysis assays, in brain-heart infusion (BHI) broth (Difco) supplemented with 2% (wt/vol) sodium bicarbonate and 1% (wt/vol) maltose. GAS were grown on either Trypticase-soy blood agar medium (Becton Dickinson Microbiological Systems, Franklin Lakes, N.J.) (TSA-blood) or THY agar supplemented with 5% (vol/vol) defibrinated sheep blood (PML Microbiologicals, Inc., Wilsonville, Oreg.) (THY-blood). When required, THY-blood medium was supplemented with 1.0 µg of erythromycin per ml. GAS were cultured at 37°C in the presence of 5% CO2.k)elgux, 百拇医药
DNA manipulation. Restriction endonuclease digests, DNA ligations, transformation of Escherichia coli, agarose gel electrophoresis, and Southern hybridization analysis were performed according to standard techniques (8). Restriction enzymes and DNA ligase were purchased commercially (Invitrogen, Carlsbad, Calif.). Plasmid DNA was recovered from E. coli strain DH5(Gabriele Sierig,Colette Cywes Michael R. Wessels and Cameron D. Ashbaugh)