Clonal spread of group A streptococcus with the new type of erythromycin resistance. Finnish Study Group for Antimicrobial Resistance.

In 1990, a new type of erythromycin resistance phenotype, designated NR, was found in group A streptococcus (GAS) in Finland. In the present study, the distribution of GAS isolates with this and other erythromycin-resistance phenotypes was surveyed in Finland, and the clonality of the isolates was explored. Of 4179 GAS isolates collected from all over Finland, 695 (17%) were resistant to erythromycin, and 82% of these had the NR phenotype. Of a group of 96 isolates with the NR phenotype from different areas, 91% was T4 serotype, opacity factor-positive. The majority of these isolates were studied further: All were M4 serotype and 88% were of one clonal origin in genetic analyses. Thus, one single clone predominates among erythromycin-resistant GAS in Finland. This clone is of T4M4 serotype and mediates the new type of erythromycin resistance, characterized by the NR phenotype.

The effect of changes in the consumption of macrolide antibiotics on erythromycin resistance in group A streptococci in Finland. Finnish Study Group for Antimicrobial Resistance.

BACKGROUND: In the early 1990s there was an increase in erythromycin resistance among group A streptococci in Finland. In response, policies regarding outpatient antibiotic therapy were changed, and nationwide recommendations were issued that called for reductions in the use of macrolide antibiotics for respiratory and skin infections in outpatients. We studied the effect of this policy on the pattern of erythromycin resistance throughout Finland. METHODS: From 1991 through 1996, a total of 39,247 group A streptococcal isolates from throat swabs (82 percent of the isolates) and pus samples (18 percent) and 290 isolates from blood cultures were studied in regional microbiology laboratories. The susceptibility of the isolates to erythromycin was tested by the disk-diffusion or the screening-plate method. RESULTS: Consumption of macrolide antibiotics decreased from 2.40 defined daily doses per 1000 inhabitants per day in 1991 to 1.38 in 1992 (P=0.007) and remained near the lower level during the study period. The change in consumption was followed by a steady decrease in the frequency of erythromycin resistance among group A streptococcal isolates from throat swabs and pus samples, from 16.5 percent in 1992 to 8.6 percent in 1996 (odds ratio for 1996 as compared with 1992, 0.5; 95 percent confidence interval, 0.4 to 0.5). CONCLUSIONS: In Finland, after nationwide reductions in the use of macrolide antibiotics for outpatient therapy, there was a significant decline in the frequency of erythromycin resistance among group A streptococci isolated from throat swabs and pus samples.

Molecular comparison of group A streptococci of T1M1 serotype from invasive and noninvasive infections in Finland.

A total of 98 isolates of group A streptococci of T1M1 serotype isolated in Finland during 1988-1995 from bacteremic, pharyngeal, and pyogenic infections were studied by molecular means. The typing techniques used included restriction endonuclease analysis, ribotyping, and random amplified polymorphic DNA analysis. Representatives of each observed genotype were also examined for carriage of the speA gene encoding for pyrogenic exotoxin A. All of the serotype T1M1 isolates studied were considered to be of a single clonal origin, and 66% were identical by all three genotyping techniques used. Among the remaining 34% of closely related isolates, 13 different genotypes were detected. All isolates studied carried the speA gene. No evidence was found of correlation of the genomic type to the severity of the infection or the time of isolation.

Group A streptococcal outbreak of perianal infection in a day-care centre.

Within 9 days in a day-care centre of 20 children 3 cases and 2 suspected cases of perianal infection (PAI) were revealed. The causative agent for PAI was confirmed to be beta-haemolytic group A streptococcus (GAS). The isolates were further characterized by T- and M-serotyping, conventional restriction endonuclease analysis (REA) and pulsed field gel electrophoresis (PFGE). The typing methods revealed that all outbreak isolates were of same T28R28-serotype and genetically identical. Unrelated, control GAS isolates of the same serotype differed genetically from the outbreak isolates indicating that a single clone had caused the perianal infections.

Role of gamma interferon in late stages of murine salmonellosis.

Gamma interferon (IFN-gamma) is an important mediator of natural resistance of mice to Salmonella species during the first week of infection, when it restricts the rate of intracellular growth of the bacteria but does not lead to their killing (A. Muotiala and P. H. Mäkelä, Microb. Pathog. 8:135-141, 1990). We used the experimental mouse salmonellosis model to investigate the role of IFN-gamma in the later stages of a sublethal infection and the ensuing specific immunity. When anti-IFN-gamma was administered starting 6 days after challenge, it did not prevent the cessation of intracellular bacterial growth and the formation of the plateau stage in the second week of infection. In addition, anti-IFN-gamma given 14 and 16 days after challenge did not alter the elimination of the bacteria in the clearance stage in the third week of infection. When mice immunized 2 months previously with live vaccine were infected with virulent salmonellae, depletion of IFN-gamma enhanced the early growth of the bacteria in the same manner as that seen in naive mice. However, when the immunized mice were infected with attenuated aroA bacteria, their clearance started immediately and was unaffected by IFN-gamma depletion, demonstrating that IFN-gamma is not required for the clearance. We conclude that IFN-gamma restricts the rate of intracellular bacterial multiplication in the first week of Salmonella infection in both naive and immune mice but is not a mediator of bacterial clearance in either naive or immunized mice.

Conformation of Escherichia coli outer membrane protein OmpA produced in Bacillus subtilis: influence of lipopolysaccharide.

The conformation of the outer membrane protein OmpA of Escherichia coli produced in Bacillus subtilis and solubilized in Sarkosyl was studied by measuring its ability to bind OmpA-specific phage K3 and to inhibit F-mediated conjugation. The partially purified protein was inactive in both of these assays. Refolding of the protein in the presence of lipopolysaccharide resulted in preparations with full phage-binding and conjugation-inhibiting capacity, indicating the formation of surface-exposed loops of OmpA of native conformation. The finding is of importance for the potential use of outer membrane proteins of Gram-negative bacteria as vaccines.

Low biological activity of Helicobacter pylori lipopolysaccharide.

Lipopolysaccharide from the gastroduodenal pathogen Helicobacter pylori was tested for its ability to induce mitogenicity in mouse spleen cells, pyrogenicity in rabbits, and toxic lethality in galactosamine-sensitized mice. Compared with those for enterobacterial lipopolysaccharide, mitogenicity and pyrogenicity were a thousand-fold lower and lethal toxicity was 500-fold lower. We suggest that the phosphorylation pattern and acylation in lipid A are responsible for the low biological activity.

Anti-IFN-gamma-treated mice--a model for testing safety of live Salmonella vaccines.

Salmonella enterica serovars Typhimurium and Enteritidis are facultative intracellular pathogens which cause in mice a disease similar to human typhoid fever caused by serovar Typhi. An essential phase in the infection process is bacterial replication inside cells of the liver and spleen; the rate of replication is restricted by interferon-gamma (IFN-gamma) produced early in the infection. In the present study the effect of IFN-gamma was neutralized in vivo with monoclonal antibody and the fate of bacteria in the liver and spleen in these mice compared with that in control mice after intravenous challenge. The depletion of IFN-gamma remarkably sensitized the mice to Typhimurium infection. Five different Typhimurium and Enteritidis candidate vaccine strains were tested. Only one of them, the aroA mutant SL3261, was avirulent also in anti-IFN-gamma treated mice. This finding may have important implications for the safety of live attenuated Salmonella vaccines since immunosuppression is likely to cause a state of reduced production of IFN-gamma.

Immunogenicity and protective efficacy of pertussis toxin subunit S1 produced by Bacillus subtilis.

Pertussis toxin (PT) subunit S1 was produced in Bacillus subtilis as a secretory protein designated BacS1. BacS1 was partially purified and used to immunize mice. The sera were tested for PT-neutralizing antibodies and for protective capacity in a mouse model. Unlike previous findings with recombinant S1 from Escherichia coli, the recombinant BacS1 protein induced antibodies that were both neutralizing and protective. An adjuvant was necessary for efficient immunization with BacS1 but not with PT. Of the four adjuvants tested, aluminium phosphate gel was insufficient whereas Freund's incomplete adjuvant, Klebsiella lipopolysaccharide and Ribi's monophosphoryl lipid A-trehalose dimycolate emulsion all resulted in protective antibody production in NIH mice.

A strong antibody response to the periplasmic C-terminal domain of the OmpA protein of Escherichia coli is produced by immunization with purified OmpA or with whole E. coli or Salmonella typhimurium bacteria.

We produced in Bacillus subtilis the complete, as well as the N-terminal two-thirds, OmpA protein of Escherichia coli (called here Bac-OmpA and Bac-OmpA-dN, respectively). These Bac-OmpA proteins were used to examine the immunological properties of different parts of OmpA, free of lipopolysaccharide and other components of the outer membrane. The full-length Bac-OmpA was indistinguishable from the authentic protein isolated from E. coli (Coli-OmpA) both as immunogen and as antigen in enzyme immunoassay (EIA). The N-terminal Bac-OmpA-dN was a poor immunogen which gave rise to significantly lower titers of anti-OmpA antibody than did the full-length OmpA preparations. When used as an antigen in EIA, the Bac-OmpA-dN detected anti-OmpA antibody in serum samples from animals immunized with the full-length OmpA much less efficiently than did either Bac-OmpA or Coli-OmpA. The periplasmic C-terminal domain therefore appears to be an immunodominant epitope of the purified OmpA protein. Also, when rabbits and mice were immunized with intact, live or dead E. coli, the antibody response detected by EIA with the full-length protein, Bac-OmpA, was much stronger than that detected with the N-terminal two-thirds, Bac-OmpA-dN. Similar results were obtained with the OmpA of Salmonella typhimurium. Because the ompA gene of enterobacteria is highly conserved, the Bac-OmpA might be useful as a group-specific EIA antigen to diagnose diseases caused by members of the family Enterobacteriaceae.

The role of IFN-gamma in murine Salmonella typhimurium infection.

Influence of both recombinant interferon gamma (r-IFN-gamma) and monoclonal anti-IFN-gamma on murine Salmonella typhimurium infection was studied in vivo. As a challenge we used either a virulent or an avirulent strain of S. typhimurium. The avirulent strain is unable to replicate in the mouse spleen. The effect of IFN-gamma or anti-IFN-gamma treatment on infection was assayed by counting the number of bacteria in the spleens 1, 2, 4 or 7 days after infection. Exogenously supplied IFN-gamma was found to decrease the number of both the virulent and the avirulent bacteria in the spleens 1 day after challenge but thereafter had no detectable effect. Anti-IFN-gamma-treated mice were unable to clear a sub-lethal dose of the virulent bacteria. The enhanced growth of the bacteria in the spleen was seen after 2 days of infection. In the spleens of anti-IFN-gamma-treated mice and control mice the fate of the non-replicating bacteria was similar. We conclude that the mice produce IFN-gamma during the early phase of Salmonella infection and that this causes a reduction in the apparent growth rate of the virulent bacteria. Because the avirulent bacteria, which do not multiply, were not affected by anti-IFN-gamma treatment, the effect of IFN-gamma is primarily bacteriostatic rather than bactericidal. After challenge, the production of IFN-gamma seems to start with a lag of approximately 2 days.

Serum antibody response to B. pertussis Tn5 mutants, purified PT and FHA in two different mouse strains and passive protection in the murine intranasal infection model.

The protective capacities of antibodies to pertussis toxin (PT) were compared with antibodies to several other pertussis antigens in an experimental murine model of intranasal infection with Bordetella pertussis. Protection from lethal challenge was achieved by passive immunization with mouse antisera to whole cells of the virulent B. pertussis BP338(Vir+) and its Tn5-generated mutants, BP353(Fha-), BP348(Adc-Hly-) and to a lesser extent of BP347(Vir-). The immune sera were produced in two different mouse strains, a good PT antibody responder (NIH) and a poor responder (F1 of CBA x C57BL/6). The antisera produced in the F1 mice contained no detectable neutralizing antibodies to PT as measured by the CHO cell assay. In spite of this the anti-BP353(Fha-) and BP348(Adc-Hly-) sera of the F1 mice seemed as protective as those of the NIH mice. A strong dependence between PT neutralizing antibody and protection was seen only when comparing sera of NIH and F1 mice immunized with purified active PT. The protective capacity of sera of both mouse strains immunized with purified filamentous hemagglutinin (FHA) correlated with their anti-FHA titers measured by enzyme immunoassay. The data thus confirm the protective capacity of anti-PT and anti-FHA, but also show that antibodies of other specificities can confer protection.

Antibody production to a meningococcal outer membrane protein cloned into liv Salmonella typhimurium aroA vaccine strain.

We cloned a 28 kDa outer membrane protein (OMP) of Neisseria meningitidis group B into a live Salmonella typhimurium aroA vaccine strain SL3261. The cloned 28 kDa protein was produced in large amounts in the S. typhimurium transformant SH8182 and located in the outer membrane. A mouse-passaged derivative of SH8182 was used as a live vaccine to immunize mice; with antibiotic pressure the strain survived in the mice as well as the parent strain SL3261 and maintained the plasmid carrying the gene encoding the 28 kDa OMP. The mice produced a high titer of antibodies to the 28 kDa OMP, showing that it had been effectively presented to the immune system. The hyperimmune mouse serum bound in an enzyme immunoassay to whole cells of E. coli and group B meningococci expressing the 28 kDa OMP, but its bactericidal activity towards the meningococci was marginal. In a passive protection study, the antiserum did not protect infant rats from meningococcal infection. The results indicate that the antibodies elicited did not bind to intact meningococcal cells, possibly because of inaccessibility of the 28 kDa OMP.

Protective immunity in mouse salmonellosis: comparison of smooth and rough live and killed vaccines.

Protective immunity against Salmonella infection was studied in a mouse model. To study specificity of protection we used smooth (O-4,5,12) and rough vaccines; live and killed vaccines of both types were compared. The protection was assessed by enumerating the number of bacteria in the livers, and by following survival of the mice after intravenous challenge with smooth O-4,5,12 bacteria. Passively transferred antibodies induced by the smooth vaccines had a small protective effect and those induced by the rough vaccines no protective effect in this model. Both live vaccines induced long-lasting protective immunity which was much stronger than that mediated by antibodies. Since the live rough vaccine induced protective immunity and contained no O antigen we conclude that the protective immunity induced by it was mainly cell-mediated and directed to other antigens than the O antigen, the target of protective antibodies. Both killed vaccines also induced protective immunity, but this was weaker than that induced by the corresponding live vaccine.