INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Acute respiration distress syndrome (ARDS) and pulmonary hemorrhage are frequently major complications in toxic shock-like syndrome (TSLS) caused by group A streptococci (GAS). About 60% of TSLS cases exhibit ARDS (15, 26, 28). ARDS is a result of pulmonary edema, which is caused by the permeabilization of pulmonary capillaries and ultimately leads to alveolar flooding in many cases. ARDS is one of the complications in systemic bacterial infections which have been recently classified as systemic inflammatory response syndrome; however, only a few bacterial toxins and cell components have been identified as causative substances for ARDS (23, 25).

Because of the worldwide prevalence of TSLS, a number of studies have been undertaken to identify the relevant virulence factors (6, 18, 29). The proposed virulence factors include exotoxins such as streptococcal pyrogenic exotoxins A (SpeA), B (SpeB), C (SpeC), and F (SpeF; previously referred to as mitogenic factor) and streptococcal superantigen (3, 5, 17, 32, 33). These exotoxins have superantigenic activities which induce massive T-cell proliferation and release large amounts of cytokines. It has therefore been postulated that some clinical symptoms such as multiorgan failure, high fever, and shock in TSLS are caused by the action of these superantigens (11, 14, 20). SpeB is a protease and activates interleukin-1 (10). The cytokines tumor necrosis factor alpha and interleukin-1 have toxic effects on endothelial cells (4, 27), and therefore extremely high levels of these cytokines in the circulation may injure pulmonary microvascular endothelial cells, resulting in permeabilization of pulmonary capillaries. Another hypothetical mechanism for the induction of ARDS in TSLS patients is that toxins produced by GAS immediately damage pulmonary endothelial cells. Lee et al. particularly noted that this hypothesis is of major importance in the development of TSLS (12).

An isolated perfused lung model used previously for the study of ARDS induced by bacterial toxins (23) has enabled us to measure the permeabilization effect of toxins on pulmonary capillaries. We found that crude preparations from the culture supernatant of GAS isolates of M types 1 and 3 are potent inducers of ARDS in the isolated perfused rat lung model. We then further purified and characterized the active substance.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial isolates. Streptococcus pyogenes isolates used in this study were isolated in Japan between 1992 and 1994 (13). The isolates used in the lung vascular permeability assays and in immunoblot analysis are listed in Tables 1 and 2, respectively. The case definition of TSLS was based on the formula developed by the U.S. Working Group on Severe Streptococcal Infections (34).

M typing. M antigens of S. pyogenes were extracted by the hot HCl method, and M types were determined by the capillary precipitation reaction, as described previously (19).

Purification of the active substance. S. pyogenes was grown in 50 ml of brain heart infusion broth (Difco Laboratories, Detroit, Mich.) at 37°C in a 5% CO₂ atmosphere. The culture supernatant was collected by centrifugation and filtered through a 0.22-µm-pore-size sterile membrane filter (Millipore Corp., Bedford, Mass.). The proteins in the culture supernatant were precipitated with a 50% saturated ammonium sulfate solution at 4°C for 2 days. The precipitate was collected and resolubilized in phosphate buffer (0.1 M sodium phosphate [pH 7.0]), and the solution was dialyzed into phosphate buffer at 4°C overnight. After dialysis, the volume of the crude preparation was adjusted to a concentration of 0.5 g of protein/ml and used for the lung vascular permeability assays.

Lung vascular permeability assay. Male rats Wistar weighing 250 to 300 g were anesthetized with an intraperitoneal administration of pentobarbital sodium (35 mg/kg of body weight). The lungs were isolated and perfused for the lung vascular permeability assay as described below.

1

Preparation of antisera. Rabbit antisera were prepared as described previously (8). Briefly, the purified antigen preparation or a synthetic peptide coupled with keyhole limpet hemocyanin at a concentration of 10 µg/ml was emulsified with an equal volume of Freund's complete adjuvant (Difco) and subcutaneously injected into a rabbit. The second antibody response was elicited by immunization with the antigen alone, and serum was obtained. The synthetic peptide, Val-Leu-Val-Tyr-Asn-Thr-Ala-Asn-Thr-Ile-Asn-Tyr-His-Asn-Gly-Thr-Pro-Thr-Gln-Lys, was derived from the published SpeF sequence (8) and was coupled with keyhole limpet hemocyanin for immunization.

Western blot analysis. Active fractions were separated by SDS-PAGE in 12% acrylamide and electroblotted onto a nitrocellulose filter membrane as described previously (31). After immunoreaction with rabbit antiserum, the membrane was secondarily stained with horseradish peroxidase-labeled goat anti-rabbit immunoglobulin (Biosource, Camarillo, Calif.).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Lung vascular permeabilization induced by crude preparations. We used the isolated perfused rat lung model to assay lung vascular permeability. The capillary filtration coefficient value was used as an index of permeabilization. This model utilizes an artificial solution instead of blood, enabling us to detect the direct effects of samples on pulmonary endothelial cells by minimizing the effects of blood components secreted from cells such as lymphocytes.

View larger version (27K): [in this window] [in a new window]   FIG. 1.   Lung vascular permeabilization induced by crude preparations. At time zero, 300 µl of each crude preparation was added to 50 ml of the perfusion fluid. The capillary filtration coefficient values for the preparations were measured at 20, 50, 80, and 110 min. The M type and source of each S. pyogenes isolate are described in Table 1.

Purification and characterization of the active substance. S. pyogenes isolate 1286 from a TSLS patient exhibited high activity in lung vascular permeability assays. The culture supernatant of the isolate was saturated with ammonium sulfate, and the protein fraction was precipitated.

View larger version (73K): [in this window] [in a new window]   FIG. 2.   The purified active substance in SDS-PAGE. The active substance was purified from the crude preparation of S. pyogenes 1286 as described in Materials and Methods and analyzed by SDS-PAGE. Lane 1, molecular weight markers; lane 2, crude preparation; lane 3, active fraction after DEAE column purification; lane 4, active fraction after phenyl-Sepharose column purification; lane 5, active fraction after Superdex 75 column purification.

Lung vascular permeabilization by purified SpeF and its inhibition by anti-SpeF serum 1. In the isolated perfused rat lung model, purified SpeF at 1.0 mg/ml in phosphate-buffered saline was added stepwise to 50 ml of the perfusion fluid to give concentrations of 10, 30, and 100 ng/ml at 0, 40, and 80 min, respectively. The capillary filtration coefficient values were then measured at 10, 30, 50, 70, 90, and 110 min (Fig. 3). As Fig. 3 shows, purified SpeF had a cumulative effect on lung vascular permeabilization, which was abolished by anti-SpeF antiserum prepared with immunization of the purified SpeF (anti-SpeF serum 1); in contrast, normal rabbit serum had no effect on the permeabilization activity of the purified SpeF.

View larger version (19K): [in this window] [in a new window]   FIG. 3.   Inhibition of lung vascular permeabilization by ant-SpeF serum 1. One milligram of the highly purified active substance (SpeF) in phosphate-buffered saline was mixed with an equivalent volume of anti-SpeF serum 1 or a normal rabbit serum, and the mixture was incubated for 30 min at room temperature. The mixtures were added to 50 ml of the perfusion fluid to reach a toxin concentration of 100 ng/ml. The capillary filtration coefficient values were then measured at each indicated time. , SpeF treated with a normal rabbit antiserum; , SpeF; , SpeF treated with anti-SpeF serum 1. Values are the means of four different experiments ± standard deviations (SpeF) and the means of two different experiments (SpeF + anti-SpeF serum 1).

Western blot analysis of the production of SpeF. The relationship between lung vascular permeabilization activity and SpeF production by each S. pyogenes isolate was examined. Immunoblot analysis was performed to evaluate the production of SpeF in the culture fluid. The cultures were preadjusted to give ~10⁸ CFU/ml, and aliquots (0.5 ml) of the culture supernatants were prepared by centrifugation and filtration. The proteins in the culture supernatants were then precipitated with 1.0 ml of ethanol. The precipitates were collected and resolubilized in 20 µl of distilled water, and 10-µl aliquots were analyzed by Western blotting using anti-SpeF serum 1. This antiserum detected only a single band on SDS-PAGE, at the position of the purified SpeF. This band was also stained with anti-SpeF serum provided by T. Yutsudo and the antiserum prepared by immunization with the synthetic peptide (anti-SpeF serum 2) (data not shown).

View larger version (29K): [in this window] [in a new window]   FIG. 4.   Immunoblot analysis of SpeF production from S. pyogenes isolates used in the lung vascular permeability assay. Anti-SpeF serum 1 was used as the first antibody.

View larger version (47K): [in this window] [in a new window]   FIG. 5.   Immunoblot analysis of SpeF production from various S. pyogenes clinical isolates belonging to M types 1 and 3 (A) and to other M types (B). Anti-SpeF serum 1 was used as the first antibody. The M type and source of each S. pyogenes isolate are described in Table 2.

Inhibition of lung vascular permeabilization activity of crude preparation by anti-SpeF serum 2. We suspected that the lung vascular permeabilization was also induced by other exoproteins in the crude preparations and therefore prepared anti-SpeF serum 2 by immunization with a synthetic peptide derived from the SpeF sequence. This new antiserum was highly specific to SpeF, but the antibody titer was slightly lower than that of anti-SpeF serum 1, as judged by Western blot reaction.

View larger version (19K): [in this window] [in a new window]   FIG. 6.   Inhibition of lung vascular permeabilization activity of crude preparation by anti-SpeF serum 2. Three hundred microliter of the crude preparation of 1269 (M1), 1281 (M3), or 1286 (M3) was mixed with threefold volumes of anti-SpeF serum 2, and the mixture was incubated for 30 min at room temperature. The mixtures were added to 50 ml of the perfusion fluid, and the capillary filtration coefficient values for each sample were measured at 20, 50, and 80 min, respectively. , crude preparation of 1269; , crude preparation of 1269 treated with anti-SpeF serum 2; , crude preparation of 1281; , crude preparation of 1281 treated with anti-SpeF serum 2; , crude preparation of 1286; , crude preparation of 1286 treated with anti-SpeF serum 2.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

SpeF produced by S. pyogenes is a multifunctional protein with mitogenic, superantigenic, and nuclease activities (9, 30, 32). In the present study, we found that SpeF had vascular permeabilization activity in a rat lung model. Induction of significant reactions in the rat lung model required approximately 100 ng of purified SpeF in 1 ml of perfusion solution, a dose somewhat higher than those used for the induction of in vitro mitogenic activity by SpeF (16). Seeger and coworkers reported that several bacterial toxins including Escherichia coli hemolysin, Pseudomonas aeruginosa cytotoxin, and Staphylococcus aureus -toxin induce lung vascular permeabilization in an isolated perfused rabbit lung model, very similar to the rat model used in this study; they showed that these toxins play important roles in the pathogenesis of ARDS caused by bacterial infections (1, 21, 24). SpeF may be a crucial factor for onset of ARDS during severe infection or TSLS caused by S. pyogenes.

It has been proposed that the mechanisms by which these toxins exert lung vascular damage in the isolated perfused rabbit lung model include transmembrane pore formation by toxins, calcium gating through the pores, calcium-mediated induction of phospholipolytic activities, and subsequent formation of cell specific arachidonic acid metabolites (22, 23).

There are a few possible mechanisms which would explain rat lung vascular permeabilization by SpeF. One is that mentioned by Seeger et al. (1, 21, 24): SpeF would form pores on lung vascular endothelial cells and then causes sequential events which trigger arachidonic acid cascades and consequently causes lung vascular permeabilization. However, this is unlikely, because so far there is no evidence that SpeF has pore-forming ability. Second, SpeF has potent superantigenic activity (30) and may activate T lymphocytes to produce excessive amounts of lymphokines and induce vascular permeabilization. However, in this isolated rat lung model, most T cells in the blood vessels were washed away with Krebs-Henseleit solution before SpeF application and permeabilization was induced within 30 min after the application, minimizing the possibility of lymphocyte involvement. Third, SpeF might immediately stimulate vascular endothelial cells to cause vascular permeabilization via an unidentified pathway. It has been demonstrated that S. pyogenes produces various extracellular toxic proteins, including pore-forming toxins and superantigenic Spe toxins (3, 5, 17, 25, 32, 33). Minor contaminants of these toxins in the purified SpeF preparation would induce permeabilization of lung blood vessels. However, anti-SpeF serum 2 prepared with a synthetic peptide based on the published SpeF sequence also inhibited the permeabilization activity. We therefore suggest that streptococcal SpeF is a major causative toxin for permeabilization of lung blood vessels. Since the inhibition by anti-SpeF serum 2 was not complete for some crude preparations, other toxins such as hemolysin in the crude preparations would also be responsible for induction of vascular permeabilization. It should be noted, however, that the anti-SpeF titer of serum 2 was lower than that of serum 1.

The amount of SpeF secreted in the culture varied among S. pyogenes isolates, but M type 1 and 3 isolates generally produced high amounts of SpeF. Isolates belonging to other M types produced varied amounts of SpeF. Isolates K5 and O3 expressed antigenically positive SpeF with an apparent molecular mass the same as that of the purified preparation, but the crude preparations of these isolates showed no positive permeabilization activity. The structure of the active domains of SpeF may be lost in K5 and O3, although sequence data for the inactive SpeF are not yet available. It is also likely that the antigenic determinant domains of SpeF and the domains relevant to the permeabilization activity are different.