Characterization of the Bacteroides fragilis pathogenicity island in human blood culture isolates.

Bacteroides fragilis is an important anaerobic pathogen accounting for up to 10% of bacteremias in adult patients. Enterotoxin producing B. fragilis (ETBF) strains have been identified as enteric pathogens of children and adults. In order to further characterize the B. fragilis pathogenicity island (BfPAI) and using PCR assays for bft- and mpII-metalloprotease genes, we determined the frequency of B. fragilis strains with pattern I (containing the BfPAI and its flanking region), pattern II (lacking both the BfPAI and the flanking region), and pattern III (lacking the BfPAI but containing the flanking region) in 63 blood culture isolates. The results were compared to 197 B. fragilis isolates from different clinical sources. We found 19% of blood culture isolates were pattern I (ETBF), 43% were pattern II (NTBF) and 38% were pattern III (NTBF). Comparatively, B. fragilis isolates from other clinical sources were 10% pattern I, 47% pattern II and 43% pattern III. This suggests that the pathogenicity island and the flanking elements may be general virulence factors of B. fragilis.

Nosocomial neonatal outbreak of Serratia marcescens--analysis of pathogens by pulsed field gel electrophoresis and polymerase chain reaction.

BACKGROUND: We investigated an outbreak of Serratia marcescens in the neonatal intensive care unit (NICU) and the pediatric intensive care unit (ICU) of the University Children's Hospital Leipzig, Germany. PATIENTS AND METHODS: From September to November 1998 15 patients were infected or colonized by S. marcescens. During the outbreak swabs from eye, blood, throat and nose were taken from every patient hospitalized in the ICUs. RESULTS: In 15 cases (14 from the NICU and one from the pediatric ICU) the cultures yielded S. marcescens. All strains were investigated by pulsed field gel electrophoresis (PFGE) as well as by polymerase chain reaction (PCR) fingerprinting. Both molecular typing methods revealed corresponding fingerprint patterns in all of the 15 isolates. Typing results of the outbreak-related isolates demonstrated that two epidemic strains of distinct genotypes were associated with cross-infections of a group of five and a group of ten patients, respectively. The three invasive and seven of the colonizing isolates were related genotypically. CONCLUSION: This survey shows that PCR and PFGE are comparable in respect to the discrimination and reproducibility for epidemiological studies of S. marcescens strains in nosocomial outbreaks. Genotypic fingerprinting of bacterial isolates is useful and important to limit nosocomial infections. Fingerprinting sources of nosocomial infections can be traced both by PFGE and PCR. All patients infected recovered completely and the nosocomial outbreak could be stopped rapidly.

Pasteurella multocida subsp. multocida and P. multocida subsp. septica differentiation by PCR fingerprinting and alpha-glucosidase activity.

Pasteurella multocida is composed of three subspecies that are often differentiated by fermentation of sorbitol and dulcitol. We studied 35 dulcitol-negative P. multocida isolates from infected dog and cat bite wounds, 16 of which yielded weak and/or conflicting fermentation reactions in Andrades sorbitol, thus making it difficult to distinguish between the two dulcitol-negative subspecies of P. multocida, i.e., P. multocida subsp. multocida and P. multocida subsp. septica. All isolates and two control strains were further analyzed using a PCR fingerprinting technique with a single primer (M13 core) and assessed for alpha-glucosidase (alpha-Glu) activity. Although the PCR fingerprint patterns and alpha-Glu activity did not correlate well with the sorbitol fermentation reactions, they did correlate well with each other. All strains identified as P. multocida subsp. septica were positive for alpha-Glu activity and exhibited the group I PCR fingerprint profile. All strains categorized as P. multocida subsp. multocida displayed either the group II or group III PCR fingerprint profile; 9 of 11 of these isolates were alpha-Glu negative. These data suggest that both PCR fingerprinting and alpha-Glu activity provide reliable means for differentiating P. multocida subsp. multocida from P. multocida subsp. septica, particularly in strains that produce weak and/or discrepant sorbitol fermentation reactions.

In vitro activities of fourteen antimicrobial agents against obligately anaerobic bacteria.

The in vitro activities of fourteen antimicrobial agents were tested against 292 clinical isolates of obligately anaerobic bacteria using the broth microdilution technique. Taking all strains as a group the MIC(50/90) (mg/l) values were metronidazole and imipenem 0.25/1, meropenem 0.25/0.5, trovafloxacin 0.25/1, gatifloxacin and moxifloxacin 0.5/2, levofloxacin 2/16, ciprofloxacin 4/32, clindamycin 0.5/8, amoxycillin/clavulanate 1/4, doxycycline and chloramphenicol 2/4, erythromycin 4/>32 and penicillin G 16/>32.

In vitro activity of telithromycin (HMR 3647) and seven other antimicrobial agents against anaerobic bacteria.

We assessed the in vitro activity of telithromycin (HMR 3647) and seven other antimicrobials against 292 strains of obligately anaerobic bacteria. MICs were determined with the microdilution technique and Wilkins-Chalgren broth according to DIN 58940-83. MIC50/MIC90s (mg/L) for telithromycin were 4/4 for Bacteroides fragilis, Bacteroides ovatus and Bacteroides thetaiotaomicron, 2/4 for Fusobacterium spp. and Bilophila wadsworthia, 2/2 for Bacteroides caccae, 1/4 for Bacteroides vulgatus, 0.25/4 for Prevotella spp., > or =0.03/0.5 for Clostridium spp. and 0.125/4 for Peptostreptococcus spp.

Comparative activity of moxifloxacin in vitro against obligately anaerobic bacteria.

The antimicrobial activity of moxifloxacin and seven other antibiotics (four of them quinolones) against 292 strains of obligately anaerobic bacteria was assessed employing a broth microdilution technique performed in Wilkens-Chalgren broth. MIC50/MIC90 values (mg/l) for moxifloxacin were as follows: Bacteroides fragilis (n = 62) 0.25/2, Bacteroides ovatus (n = 70) 1/4, Bacteroides vulgatus (n = 29) 0.25/1, Bacteroides thetaiotaomicron (n = 17) 2/2, Bacteroides caccae (n = 11) 1/2, Prevotella spp. (n = 11) 0.25/2, Fusobacterium spp. (n = 17) 1/4, Bilophila wadsworthia (n = 29) 0.5/1, and Clostridium spp. (n = 29) 0.125/0.5, respectively. MIC50 values (mg/l) for Bacteroides distasonis (n = 8) and Peptostreptococcus spp. (n = 9) were 0.25. The results indicated that moxifloxacin was almost as active as trovafloxacin, as active as gatifloxacin, and more active than levofloxacin and ciprofloxacin against the anaerobes tested.

Occurrence of Bacteroides fragilis enterotoxin gene-carrying strains in Germany and the United States.

Ninety-three Bacteroides fragilis isolates from different geographic locations were analyzed for the presence of an enterotoxin-encoding gene. It was shown that blood culture isolates were more likely to carry this gene than strains from other sources. All enterotoxin-positive strains belonged to the PCR fingerprint group I.

Epidemiologic characterization of Pseudomonas aeruginosa in patients with cystic fibrosis.

OBJECTIVE: To determine persistence and variability of colonization with Pseudomonas aeruginosa in cystic fibrosis patients over long time periods, and to look for possible cross-colonization. METHODS: In total, 469 Pseudomonas aeruginosa isolates were obtained from 30 patients during the period from April 1994 to April 1996. The sources were mainly sputum and a few deep throat swabs. All grown strains dissimilar in macromorphology were processed separately. Typing with PFGE was carried out by contour-clamped homogeneous electric field electrophoresis. Genomic DNA was subjected to the rare-cutting restriction enzyme SpeI. For pyocin typing, the procedure described by Fyfe was applied. RESULTS: After typing with PFGE, we observed 40 restriction profiles. Eighteen different pyocin types were found. The most frequent pyocin type was type 3, followed by types 1 and 5. Twenty-two patients were persistently colonized by one clone specific and different for each patient, and four were co-colonized by a second clone also different for each of these patients. Cross-colonization had apparently been rare in the cystic fibrosis center of Leipzig. CONCLUSIONS: Typing with PFGE is well suited for detailed investigations of colonization with Pseudomonas aeruginosa in cystic fibrosis patients. Pyocin typing can provide additional information for epidemiologic purposes.

In vitro activities of gatifloxacin, two other quinolones, and five nonquinolone antimicrobials against obligately anaerobic bacteria.

The activity of the new fluoroquinolone gatifloxacin was compared with those of other quinolones and antimicrobial agents of other classes against 294 anaerobes by the broth microdilution technique. For all strains tested, gatifloxacin MICs at which 50 and 90% of the isolates were inhibited were 0.5 and 2 mg/liter, respectively, and were 3 to 4 dilution steps lower than, e.g., ciprofloxacin.

Bilophila wadsworthia clinical isolates compared by polymerase chain reaction fingerprinting.

Bilophila wadsworthia isolates recovered from a right-ear cholesteatoma and brain abscess of the same patient were analyzed by means of polymerase chain reaction (PCR) fingerprinting with single primers (T3B and M13 core) to ascertain if they originated from the same clone. Their PCR fingerprint profiles were compared with those of three additional B. wadsworthia clinical isolates and the type strain (ATCC 49260). The two isolates from the same patient produced PCR fingerprint profiles identical to each other, regardless of which primer was used. All isolates' PCR fingerprint profiles, with use of either the T3B or M13 core primer, shared some major and minor bands. However, differences in additional major and minor bands distinguished each of the additional isolates, suggesting that there are different subgroups of B. wadsworthia.

Use of polymerase chain reaction fingerprinting to compare clinical isolates of Bacteroides fragilis and Bacteroides thetaiotaomicron from Germany and the United States.

Accurate identification of Bacteroides species is often problematic. Therefore, we used a polymerase chain reaction (PCR) fingerprinting technique with either a single nonspecific primer derived from tDNA intergenic spacer or a single primer that anneals to mini- and microsatellite DNA sequences to compare 34 clinical isolates of B. fragilis and 21 clinical isolates of B. thetaiotaomicron from Southern California with 32 clinical isolates of B. fragilis and 10 isolates of B. thetaiotaomicron from Germany. All German B. fragilis isolates (32 of 32) formed one PCR fingerprint group that matched the PCR profile of the B, fragilis reference strain ATCC (American Type Culture Collection) 25285, representative of DNA homology group I. In contrast, the isolates from Southern California formed two PCR fingerprint groups. Although most of these strains (29 of 34) also matched B. fragilis ATCC 25285, some strains (4 of 34) matched the DNA homology group II reference strain VPI (Virginia Polytechnic Institute) 2393. One of the 34 strains showed a unique profile. German B. thetaiotaomicron strains (10 of 10) formed one PCR fingerprint group, matching the reference strain B. thetaiotaomicron ATCC 29742, whereas the B. thetaiotaomicron isolates from Southern California showed heterogenous profiles.

Characterization of saccharolytic Bacteroides and Prevotella isolates from infected dog and cat bite wounds in humans.

Saccharolytic, nonpigmented, anaerobic gram-negative rods isolated from infected dog and cat bite wounds in humans have been poorly characterized, and most are not included in the databases of kits used for anaerobic identification; thus, they are problematic for clinical laboratories to identify. Fifty strains isolated from such wounds were characterized with commercial kits for preformed-enzyme detection, carbohydrate fermentation, and other biochemical tests. PCR fingerprinting was performed on these strains to further characterize subgroups within these species. Bacteroides tectum is a frequent isolate in bite wounds and resembles Prevotella bivia in colony morphology and saccharolytic activity, except that it grows in 20% bile and hydrolyzes esculin. Profile numbers generated by various kits associate B. tectum with P. bivia, Prevotella oralis group, or Prevotella melaninogenica. PCR fingerprinting identified at least four subgroups and confirmed the heterogeneous nature of this species. Prevotella heparinolytica was also frequently isolated from these bite wounds. It produces indole and generates a profile number in preformed-enzyme kits that is usually associated with Bacteroides uniformis. However, it is bile sensitive and quite distinct from the Bacteroides fragilis group of anaerobes. The PCR fingerprint profiles generated by strains of P. heparinolytica were very similar to that of the type strain and to each other. Prevotella zoogleoformans, occasionally isolated from dog and cat bite wounds in humans, resembles P. heparinolytica except for a negative indole test. Clinical laboratories should be aware of the characteristics of these animal species when identifying isolates from animal bite wounds in humans.

Characterization of indole-negative Bacteroides fragilis group species with use of polymerase chain reaction fingerprinting and resistance profiles.

Biochemical tests alone do not adequately differentiate the various Bacteroides species, groups, and antimicrobial-resistant variants. Consequently, we used a polymerase chain reaction (PCR) fingerprinting technique, with either a single nonspecific primer derived from the t-DNA intergenic spacer region (T3B) or a single primer that anneals to minisatellite DNA sequences (M13 core), to identify and characterize 58 clinical isolates of Bacteroides fragilis group species (B. fragilis, B. distasonis, and B. caccae). In addition to species- and subspecies-specific differences, 4 strains of B. fragilis, 1 of B. distasonis, and 3 of B. caccae that showed increased resistance to imipenem, ampicillin, and ampicillin/sulbactam also produced unique PCR fingerprint profiles. Analysis by the clinical source of isolation (i.e. blood or intraabdominal, skin, or soft-tissue infection) indicated that no particular PCR fingerprint type was associated with greater pathogenicity of any individual clinical source. The PCR fingerprinting technique proves to be a useful tool for species identification and taxonomic studies, as well as for epidemiological studies of Bacteroides species.

Frequency of isolation of Porphyromonas species from infected dog and cat bite wounds in humans and their characterization by biochemical tests and arbitrarily primed-polymerase chain reaction fingerprinting.

We isolated 40 strains of Porphyromonas (formerly Bacteroides) species from 29 of 102 cat and dog bite wounds in humans. P. salivosa, P. gingivalis, and P. canoris were the most frequent isolates. A comparison of the RapID ANA II system (Innovative Diagnostic Systems, Norcross, GA), An-IDENT panels (bioMérieux, St. Louis), and API ZYM strips (bioMérieux) showed that the latter kit best characterized these isolates because it included tests for trypsin and chymotrypsin activity; however, the tests for glycosidase activity in this kit were less sensitive than were those in the other kits. None of the biochemical systems was able to identify all species. Arbitrarily primed-polymerase chain reaction fingerprinting with a nonspecific single primer, T3B, yielded distinct profiles for type strains and for the clinical isolates, suggesting that some of the isolates represented previously undescribed species. Growth of these species took > or = 5 days; therefore, laboratories should incubate anaerobic plates from bite wound cultures for > or = 7 days to assure isolation of these common pathogens.

Comparison of Etest to broth microdilution method for testing Streptococcus pneumoniae susceptibility to levofloxacin and three macrolides.

When the Etest was compared to broth microdilution for susceptibility testing of Streptococcus pneumoniae, levofloxacin, erythromycin, and penicillin results correlated for both methods; azithromycin and clarithromycin showed discrepancies of > or = 2 dilutions for 95.8% and 31.5% of the isolates, respectively. Levofloxacin was active against 141 of 142 isolates (< or = 2.0 micrograms/ml), making it a potentially useful new fluoroquinolone.

United States National Hospital Survey of anaerobic culture and susceptibility methods, II.

To assess the status of clinical anaerobic bacteriology in the United States, we surveyed (by means of a questionnaire) 120 hospitals selected at random with bed capacities of 200-1000, and we received responses from 78 (65%), all of which performed some degree of clinical anaerobic microbiology. Separate anaerobic blood culture bottles were used by 73 labs (94%) (median, 450 specimens/mo): 56% used Bactec 7, 27 or 37; 15% used 'BacT-Alert'; 11% used Columbia broth; 5% used thioglycolate and 'lytic'; 3% each used, Dupont Isolator, Supplemented peptone or other media. Selective media was used for primary anaerobe isolation by 89% labs which included: LKV, 76%; PEA, 53%; BBE, 31%; CNA, 28%; 'CDC', 12%. Sixty labs (78%) stored anaerobes after isolation (median 7 days), most using blood agar plates (31%), chopped meat (26%) or thioglycolate broth (27%) either for further identification (30 out of 78) or susceptibility testing (33 out of 78), if clinically indicated. Only 23% performed routine anaerobic susceptibility testing of clinical isolates. Of the 77% that do not perform susceptibility studies, 59% would not even perform them upon physician request; 30% relied on published surveys; 68% did not publish results of anaerobic susceptibility in annual summaries. When susceptibility testing was performed, the test agents selected were related to availability on a commercial system (21), NCCLS recommendation (20), hospital formulary (15) or hospital committee input (20). Nine of 78 labs (12%) had discussed stopping or decreasing the performance of both anaerobic bacteriology and susceptibility testing. Despite educational and published guidelines, clinical anaerobic bacteriology is not uniformly practiced and could be improved. In addition, an educational effort must be made in order to stress the relevance and increase performance of anaerobic bacteriology.

Survival of anaerobic bacteria in various thioglycolate and chopped meat broth formulations.

Three commercially available formulations of thioglycolate broth and of chopped meat broth were evaluated for their abilities to maintain the viabilities of 32 strains of anaerobic bacteria during a period of 8 weeks. While thioglycolate broth supported the initial (48-h) growth of all strains tested, approximately half of the strains died off within 4 weeks. Chopped meat broths maintained the viabilities of almost all cultures during the test period.