[Production and partial purification of the main hemolysin (pneumolysin) of Streptococcus pneumoniae]

The capacity of 36 Cuban strains of Streptococcus pneumoniae to produce pneumolysin was studied, and 94.4% of them were determined to be producers of that enzyme. One of the best producers was cultured at a great scale and the pneumolysin found in the supernatan was partially purified through an ion-exchange chromatography in mono-Q column. This method made it possible to recover the enzyme whose purity level increased by 4.39 with 100% output.

A comparison of the early inflammatory effects of an agr-/sar- versus a wild type strain of Staphylococcus aureus in a rat model of endophthalmitis

PURPOSE

We examined the ability of a wild type and an isogenic mutant strain of Staphylococcus aureus, deficient in the production of hemolysins and lipase (agr (-)/sar (-)), to induce endophthalmitis and inflammatory cell infiltration into the eye at 6, 24 and 48 hours after injection in a rat model of endophthalmitis.

METHODS

Rat eyes were injected with 25 microl of viable S. aureus or sterile saline. Eyes were graded for clinical signs of inflammation daily, removed and processed for standard histologic analysis 6, 24 and 48 hours after injections. Comparisons of clinical scores and mean inflammatory cell numbers were made between S. aureus and control injected eyes.

RESULTS

Both experimental groups developed clinical signs of endophthalmitis and demonstrated infiltration of inflammatory cells at 24 and 48 hours. Clinical inflammation in the Mutant I group was less than the wild type group at these times and significantly less at 48 hours (p<0.05). No statistically significant difference in the number of inflammatory cells was detected between the wild type and Mutant I injected eyes at 24 hours. At 48 hours, inflammatory cells increased by 75.0% in the wild type group and decreased by 19.0% in the Mutant I group and a statistically significant difference was seen between these two groups (p<0.05). At all times, the majority of inflammatory cells were neutrophils. By 48 hours, an increase in monocytes-macrophages was noted.

CONCLUSION

Both strains of S. aureus induced clinical signs of inflammation and inflammatory cell infiltration. Clinical inflammation and inflammatory cell numbers were less in rats injected with the Mutant I strain. These results suggest that hemolysins and lipase may be important in the early induction phase of the inflammatory response.

Unique synthetic peptides stimulating streptolysin S production in streptococci

A peptide has been isolated from pronase digest of bovine serum albumin as the stimulatory factor of streptolysin S (SLS) production by Streptococcus pyogenes, and its primary structure has been deduced [Akao et al. (1992) Infect. Immun. 60, 4777-4780]. To determine the essential structure for the stimulation, a peptide (P-1) having the deduced structure, in which three peptide fragments are linked by two disulfide bonds, and shorter analogs (P-2 to P-4) of peptide P-1 were chemically synthesized. Another peptide (P-5), in which Ala is inserted between the two Cys residues in the middle peptide chain of P-1, was also synthesized. These synthetic peptides were identified by mass spectrometry and analysis of amino acid compositions. The synthetic P-1 stimulated SLS production in a dose-dependent manner. Other peptide analogs also showed remarkable stimulation of SLS production. Treatment of P-1 with performic acid resulted in loss of its stimulatory activity, indicating that disulfide bridges of the peptides are necessary for their activity on SLS production. These results suggest that the unique primary structure of three peptide chains linked by two disulfide bridges is requisite for the stimulatory effect on SLS production.

Characterization of the haemolytic activity of Streptococcus equi

The haemolytic activity of Streptococcus equi, the cause of equine strangles, was characterized. Production of haemolysin in Todd Hewitt broth was dependent on an equine serum supplement and the logarithmic phase of growth after which activity declined sharply. RNA core also induced haemolysin production from cells harvested at the end of the logarithmic phase of growth. Haemolysis was not affected by cholesterol, was only slightly increased in reducing conditions and was completely inactivated by trypan blue, identifying the haemolytic activity as streptolysin S-like (SLS-like). Purification by hydroxyapatite and Sephacryl column chromatography yielded proteins of molecular weights of approximately 6000 and 17 000-22 000 Da with a 64-fold increase in specific activity. Low molecular weight proteins from the RNA core were still present in the purified toxin. Two non-haemolytic mutants were derived by conjugation with an Enterococcus faecalis-carrying transposon Tn916. Southern blots of HindIII digests of DNA revealed that one of the mutants contained three transposon insertions and the other just one. A lambda phage library of S. equi contained plaques whose haemolytic activity was enhanced by reducing conditions and inhibited by cholesterol, suggesting a streptolysin O-like (SLO-like) activity. However, haemolysin in culture sonicates of host E. coli in which the lambda phage insert was subcloned into plasmid (pUC18), was not affected by these conditions. Seven isolates of S. equi in medium without SLS-like inducers showed no SLO-like activity and no evidence for an SLO-like toxin could be found by immunoblotting with pneumolysin antiserum and monoclonal antibodies or by polymerase chain reaction with primers derived from sequences conserved between the SLO genes of Lancefield group A, C and G streptococci. S. equi does not appear to possess a streptolysin O but does make a streptolysin S-like toxin whose production can be interrupted at just one genetic locus.

Effects of synthetic hydroxy isothiocyanates on microbial systems

Hydroxy isothiocyanates (ITCs), including some new derivatives of naturally occurring compounds, were synthesized and their minimum inhibitory, minimum fungicidal, and minimum bactericidal concentrations for Aspergillus niger, Aspergillus fumigatus, Staphylococcus aureus, and Escherichia coli were estimated. These compounds were strongly antimicrobial; for example, 2-(4-hydroxyphenyl) ethyl ITC inhibited growth of all strains examined at concentrations of 7.8 to 15.6 micrograms/ml. The ATP concentration in E. coli was markedly reduced when cells were treated with 2-(4-hydroxyphenyl)ethyl ITC. Inhibition of the growth of E. coli by 2-(4-hydroxyphenyl)ethyl ITC was decreased in the presence of cysteine. Streptolysin S production in washed cells of Streptococcus equisimilis was extremely sensitive to this ITC derivative and this inhibition also was counteracted by cysteine. The results showed that the ITC compounds had antimicrobial effects by blocking sulfhydryl groups.

Temporal production of streptococcal erythrogenic toxin B (streptococcal cysteine proteinase) in response to nutrient depletion

The effects of various growth conditions on the production of streptococcal erythrogenic toxin B (streptococcal pyrogenic exotoxin B [SPE B]) by Streptococcus pyogenes were analyzed. SPE B was detected in broth culture supernatant fluid only during the stationary phase of growth when glucose and other potential carbon sources were depleted from the medium. Additionally, SPE B production was inhibited when the concentration of glucose in the medium was maintained. These results suggest that SPE B is secreted under conditions of starvation and may be involved in nutrient acquisition.

Differences in virulence for mice among Streptococcus pneumoniae strains of capsular types 2, 3, 4, 5, and 6 are not attributable to differences in pneumolysin production

We observed that differences in the in vivo growth kinetics of pneumococcal strains of capsular types 3, 4, 5, and 6 were reminiscent of differences that we had previously reported for type 2 strain D39 and its pneumolysin-deficient mutant, PLN. Capsular type 2 Streptococcus pneumoniae D39 exhibits exponential growth in the blood of XID mice until the death of the mice at 24 to 36 h. In contrast, PLN reaches a plateau in growth that is maintained for several days. Capsular type 3 and 5 strains exhibited exponential growth and caused rapid death of XID mice following intravenous challenge, similar to the observation with D39. Strains of capsular types 4 and 6 exhibited growth kinetics reminiscent of PLN. Since the observed differences in the pathogenesis of types 3 and 5 compared to 4 and 6 were reminiscent of the effects of pneumolysin deficiency in type 2, we examined the levels of in vitro pneumolysin production for the entire panel of strains. The onset of pneumolysin production in most strains was rapid and occurred near the end of log-phase growth. Differences in in vivo growth patterns of capsular type 2, 3, 4, 5, and 6 strains were not found to be associated with differences in the levels of pneumolysin.

Expression of active streptolysin O in Escherichia coli as a maltose-binding-protein--streptolysin-O fusion protein. The N-terminal 70 amino acids are not required for hemolytic activity

Streptolysin 0 (SLO) is the prototype of a family of cytolysins that consists of proteins which bind to cholesterol and form very large transmembrane pores. Structure/function studies on the pore-forming cytolysin SLO have been complicated by the proteolytic inactivation of a substantial portion of recombinant SLO (rSLO) expressed in Escherichia coli. To overcome this problem, translational fusions between the E. coli maltose-binding protein (MBP) gene and SLO were constructed, using the vectors pMAL-p2 and pMAL-c2. MBP-SLO fusion proteins were degraded if secreted into the E. coli periplasm, but intact, soluble MBP-SLO fusion proteins were produced at high levels in the cytoplasm. Active SLO with the expected N-terminus was separated from the MBP carrier by cleavage with factor Xa. Cleavage with plasmin or trypsin also yielded active, but slightly smaller forms of SLO. Surprisingly, uncleaved MBP-SLO was also hemolytic and cytotoxic to human fibroblasts and keratinocytes. The MBP-SLO fusion protein displayed equal activities to SLO. Sucrose density gradient analyses showed that the fusion protein assembled into polymers, and no difference in structure was discerned compared with polymers formed by native SLO. These studies show that the N-terminal 70 residues of mature (secreted) SLO are not required for pore formation and that the N-terminus of the molecule is probably not inserted into the bilayer. In addition, they provide a simple means for producing mutants for structure/function studies and highly purified SLO for use as a permeabilising reagent in cell biology research.

The role of pneumolysin and autolysin in the pathology of pneumonia and septicemia in mice infected with a type 2 pneumococcus

Mice were infected intranasally with a serotype 2 pneumococcus, a pneumolysin-negative derivative (PLN-A), or an autolysin-negative derivative (AL-2). Numbers of wild type pneumococci were seen in the lung from approximately 12 h after infection and were first detected in the blood around this time. Immunofluorescent staining of lung sections showed that pneumolysin was produced in vivo. Pneumococcal infection resulted in alteration of the composition of the blood but not the bone marrow. Some of the hematologic changes did not occur after PLN-A. PLN-A had a slower growth rate in the lung and bacteremia was delayed. AL-2 was rapidly cleared from the lungs and was not detected in the blood. These events paralleled the pattern of histology in the lung, with the severity of inflammation reduced with PLN-A and no inflammation or hematologic changes with AL-2.

Expression of streptolysin O gene in Bacillus subtilis

Streptolysin O (SLO) is a hemolytic, extracellular protein produced by Streptococcus pyogenes. A hybrid gene consisting of the promoter and signal sequence fused to the region encoding the mature sequence of the slo gene was constructed to secrete SLO in Bacillus subtilis. To increase secretion of SLO into the culture supernatant, several SLO expression vectors containing various combinations of promoters and pre-pro sequences were constructed. B. subtilis harboring pPA consisting of the P-43 promoter and the coding sequence of the pre region of the alkaline protease gene that was fused to the pro-mature region of the slo gene secreted SLO into media. The degree of hemolytic activity was found to be about 40-fold higher in the geneticaly engineered B. subtilis strain than that of S. pyogenes. Recombinant SLO was reacted with patients' sera infected by group A streptococci.

Postantibiotic effect of roxithromycin on streptolysin O production, hydrophobicity, and bactericidal activity of PMNL by Streptococcus pyogenes

Exposure of Streptococcus pyogenes to 5 x minimum inhibitory concentration of roxithromycin for 1 h produced a significant postantibiotic effect. More than 2.5 h was necessary for roxithromycin-treated bacteria to increase by 1 log10 in colony-forming units after drug removal, compared with the unexposed cells. After exposure to and removal of the drug, treated cells failed to exhibit normal hemolytic activity for at least 4 h. The inhibitory effect persisted for 20 h after drug removal, although the extent of growth for treated and untreated cells was almost the same. Hydrophobicity of treated cells, studied throughout the logarithmic growth phase with a water-hexadecan two-phase system, was markedly decreased by 40%, compared with untreated cells 4 h after drug removal. Cells that had been treated with roxithromycin became more susceptible to the bactericidal activity of human PMNL than untreated bacteria. The data indicate that some of the metabolic activity that contributes to the virulence of S. pyogenes is affected by postexposure to roxithromycin, and its minimum inhibitory concentration and serum level might not be the best indicators of efficacy in this class of drugs.

Purification and characterization of a peptide essential for formation of streptolysin S by Streptococcus pyogenes

Peptides in a pronase digest of bovine serum albumin were required for streptolysin S formation by Streptococcus pyogenes besides maltose and a carrier (the oligonucleotide fraction obtained by treatment of Saccharomyces cerevisiae RNA with RNase A). A peptide essential for streptolysin S formation was purified to homogeneity from a pronase digest of bovine serum albumin by Sephadex G-25 column chromatography, and anion-exchange, reverse-phase, and gel filtration high-performance liquid chromatography. The purified peptide was divided into more than two peptides by HCOOOH oxidation and was composed of four residues of cysteine, three of leucine, and one each of aspartic acid and glutamic acid. Leucine and cysteine were detected as amino-terminal residues, and leucine and glutamic acid were detected as carboxyl-terminal residues, suggesting that two or three peptides are linked by a disulfide bond(s). A disulfide bond structure in the peptide seemed to be required for streptolysin S formation.

The effect of roxithromycin on the virulence of gram-positive cocci

Antibiotics whose recognized mode of action comprises inhibition of bacterial protein biosynthesis are also recognized to modulate the expression of bacterial virulence factors when incorporated into culture media at sub-MIC levels. In this respect, one of the new macrolides, roxithromycin, has been examined for its effect on toxin/enzyme production by strains of Staphylococcus aureus, Streptococcus pyogenes, and Streptococcus pneumoniae. Biosynthesis of staphylococcal coagulase and DNase could be potentiated, whereas that of staphylococcal alpha-hemolysin, streptolysins O and S, and pneumolysin were unaltered. Expression of one structural virulence factor, pneumococcal polysaccharide, was repressed in the drug's presence, resulting in potentiation of phagocytic ingestion of the drug-exposed bacteria. The drug failed to have any effect on ingestion of Staph. aureus or Strep. pyogenes. These studies provide evidence that roxithromycin may exhibit "added value" as an antibiotic in its ability to potentiate the susceptibility of Strep. pneumoniae to host defenses such as phagocytosis.

The effect of Streptococcus pneumoniae pneumolysin on human respiratory epithelium in vitro

Streptococcus pneumoniae culture filtrates and pneumolysin both slow human ciliary beating and damage respiratory epithelium in vitro. A polyclonal pneumolysin antibody bound to sepharose beads removed pneumolysin from culture filtrates and showed that pneumolysin alone was responsible for the effects on epithelium. In a 48-h organ culture pneumolysin caused ciliary slowing and epithelial disruption in a dose-dependent manner down to 5 ng/ml. Comparison of the ciliary slowing activity and pneumolysin concentration in filtrates in a continuous broth culture showed a maximal effect at 16 h (pneumolysin 7.5 micrograms/ml). Later the activity decreased while the pneumolysin concentration increased (8.8 micrograms/ml). This loss of activity was prevented by neutralisation of the acid pH of the culture medium. Eight different culture filtrates produced significant (P less than 0.05) ciliary slowing which correlated (r = 0.95) with simultaneously measured haemolytic (pneumolysin) activity. Substitution of tryptophan (position 433) by phenylalanine reduced the haemolytic and ciliary slowing activity of pneumolysin, but did not affect its ability to activate complement. There was no correlation between the ciliary slowing produced by the culture filtrate and that produced by the autolysate of a particular strain, nor between ciliary slowing and the extent of autolysis or the serotype of the strain.

Effect of Streptococcus pneumoniae on human respiratory epithelium in vitro

A total of 11 of 15 Streptococcus pneumoniae culture filtrates and all five bacterial autolysates produced by cell death in the stationary phase caused slowed ciliary beating and disruption of the surface integrity of human respiratory epithelium in organ culture. This effect was inhibited by cholesterol and was heat labile and reduced by standing at room temperature but was stable at -40 degrees C. The activity was detected at the late stationary phase of culture and was associated with the presence of hemolytic activity. Gel filtration of a concentrated culture filtrate and autolysate both yielded a single fraction of approximately 50 kilodaltons which slowed ciliary beating and were the only fractions with hemolytic activity. Rabbit antiserum to pneumolysin, a sulfhydryl-activated hemolytic cytotoxin released by S. pneumoniae during autolysis, neutralized the effect of the culture filtrate on respiratory epithelium. Both native and recombinant pneumolysin caused ciliary slowing and epithelial disruption. Electron microscopy showed a toxic effect of pneumolysin on epithelial cells: cytoplasmic blebs, mitochondrial swelling, cellular extrusion, and cell death, but no change in ciliary ultrastructure. Recombinant pneumolysin (10 micrograms/ml) caused ciliary slowing in the absence of changes in cell ultrastructure. Release of pneumolysin in the respiratory tract during infection may perturb host defenses, allowing bacterial proliferation and spread.

Production of pneumolysin, a pneumococcal toxin, in Bacillus subtilis

We have expressed the pneumolysin gene of Streptococcus pneumoniae in Bacillus subtilis, both from its own promoter and as a fusion protein. The level of expression of pneumolysin from its own promoter was low. The protein produced was hemolytically active. A higher level of expression (about 10 micrograms/ml of culture) was achieved when either one of two C-terminal fragments (corresponding to amino acids 265-471 and 55-471, respectively) or the entire coding part of the pneumolysin gene were fused to the promoter and signal sequence-coding region of the alpha-amylase gene of Bacillus amyloliquefaciens. The C-terminal fusion peptides reacted with anti-pneumolysin serum, but were not hemolytically active. In both cases most of the peptide remained cell-associated. When the entire pneumolysin gene was fused to the signal sequence, a hemolytically active form of pneumolysin could be detected, and most of the product was found in a processed form in the culture supernatant. The full-length pneumolysin secreted from B. subtilis was partially purified and used as antigen in an enzyme immunoassay with rabbit anti-pneumolysin serum.

The role of iron in the growth and hemolysin (Streptolysin S) production in Streptococcus pyogenes

Four strains of Streptococcus pyogenes were propagated at 37 degrees C in a reduced iron medium supplemented with Fe3+-citrate to give concentrations of 1 through 11 micrograms per milliliter, in order to observe the effects of iron on growth and on the vitro production of Streptolysin S. Both growth and hemolysin production were observed to be influenced by medium iron concentration of which 1.2 micrograms per ml of iron was critical. Hemolysin was produced during the exponential phase of the growth cycle with maximum yield as the organism entered the stationary phase. Hemolytic activity (which was accepted as the ability of the hemolysin to lyse sheep erythrocytes) fell below detectable levels as the organisms entered fully into the stationary phase (9-10 hours post incubation). Serum (bovine, human, chicken) was observed to have a high stabilizing effect on the hemolysin.