- Volume 37, Issue 4, 1992

Eighty-six strains encompassing 11 species of coagulase-negative staphylococci were examined for the production of fatty acid modifying enzyme (FAME) and lipase. Staphylococcus schleiferi and S. saprophyticus most closely resembled S. aureus in that 80% of the strains produced both enzymes. In contrast, no strains of S. lugdunensis and S. haemolyticus tested produced these enzymes. S. simulans was unusual in that eight of 10 strains produced FAME, but only one produced lipase. Among the other species the proportion of strains producing both enzymes ranged from 10 to 60%. Generally there was a strong correlation between FAME and lipase production.

Fifty-five randomly selected Staphylococcus aureus strains were examined for fatty acid modifying enzyme (FAME) production. Of these, 20.4% did not elaborate the enzyme. Amongst the remaining strains, the lowest level produced in culture was 0.1 unit/109 cocci and the maximum was 2.01 U/109 cocci; the median level was 0.4 U/109 cocci. In a series of straight-chain saturated fatty acids with 11–24 carbons, all could be esterified by FAME. However, those with 15–19 carbons were generally better substrates than the others. For a particular chain length, the unsaturated forms were better substrates than the saturated form. Triglycerides with unsaturated fatty acid side chains were potent inhibitors of FAME. Diglycerides were almost as active as triglycerides, but monoglycerides were much less inhibitory. FAME was purified by gel filtration followed by hydrophobic interaction chromatography on hexyl agarose. FAME and lipase may have a role in determining the survival of S. aureus in lesions.

The treatment of abscess homogenates with calcium ionophores stimulated the production of a bactericidal lipid with properties indistinguishable from those of a previously unidentified bactericidal lipid that had been detected in staphylococcal abscesses. The lipid was identified as a monoglyceride by thin layer chromatography. It resembled the unidentified lipid in that it had a high specific activity, exhibited differential activity, was inhibited by Staphylococcus aureus δ toxin, lecithin and Ca++, and its activity was reduced by oxidation. Stimulation of monoglyceride production by calcium ionophore requires the joint presence of components from the sedimented and supernatant fractions of abscess homogenates, and was not produced if boiled homogenate was used. The addition of verapamil interfered with the production of monoglyceride in homogenates treated with calcium ionophore. Monoglyceride was produced only in abscess homogenates and not in homogenates of other normal tissues or tissues taken from mice infected with S. aureus. Calcium ionophore could be replaced by inositol triphosphate, suggesting that monoglyceride production involved the release of calcium from intracellular stores. The 2-monoglyceride was the form originally produced in abscess homogenates, but this spontaneously isomerised to the 1-monoglyceride. The fatty-acid moiety of the monoglyceride consisted primarily of 16:0 and 16:1 fatty acids.

The Phene Plate (PhP) system of biochemical fingerprinting of bacteria is a computerised typing system, based on quantitative measurements of the kinetics of several biochemical reactions of bacteria grown in liquid medium in microtitration plates. For each isolate tested, it yields a biochemical fingerprint comprising several kinds of quantitative data which are useful for establishing similarities among strains with a personal-computer program. In this study, a set of 16 specific substrates was chosen to differentiate strains of Salmonella of serotype Typhimurium. The system was evaluated for its typability, reproducibility and discriminatory power in tests with a collection of 100 epidemiologically unrelated Typhimurium strains and results were compared with those obtained by phage typing. At an identity level of 0.980, strains were assigned by this method to 51 biochemical phenotypes (BPTs), giving a diversity index of 0.963 and a resolution index of 0.210. In contrast, 24 phage types (PTs) were identified among these isolates (a diversity index of 0.901). The combined use of biochemical fingerprinting by the PhP system and phage typing discriminated 82 phenotypes (a diversity index of 0.994). Stability of markers in each of the methods was also evaluated after subculture of 20 strains for 21 consecutive days. Only nine biochemical reactions were found that were subject to small, but measurable, changes for at least one isolate. These changes slightly decreased the mean similarity coefficients among strains but the overall BPTs of the strains showed changes in four strains (20%). In contrast, eight strains (40%) showed changes in their PTs after this treatment. It is concluded that the PhP system is a highly discriminatory and reproducible method for typing Typhimurium strains. It is easy to perform, and may be used alone or in combination with phage typing in epidemiological studies of Typhimurium strains.

A collection of 86 strains of Salmonella of serotype Typhimurium isolated from children with gastroenteritis in Tehran, Iran was examined for biochemical phenotype, phage type and antibiotic-resistance pattern. Twenty-seven biochemical phenotypes (BPTs), 14 discrete phage types (PTs) and 18 resistotypes (RTs) were identified. Fifty-three strains (62%) belonged to two major and probably related BPTs, whereas the other 33 isolates belonged to less common BPTs. The two predominant BPTs contained 26 strains of the same PT and 23 strains of the same RT. Different PTs and RTs of strains with similar BPT were sometimes observed, possibly reflecting antibiotic pressures in Iran. These results suggest that two major “clones” of Typhimurium strains are particularly common in Iran and, although each method alone adequately detected these and other less common “clones”, biochemical fingerprinting provided additional information about relationships among strains.

Bacteriophages for Acinetobacter baumanii, Pseudomonas aeruginosa and Staphylococcus aureus were tested in experimental infections of mice to investigate their potential for the treatment of infections of man. As few as 102 particles of an acinetobacter phage protected mice against 5 LD50 (1 × 108) of a virulent strain of A. baumanii, and phage was demonstrated to have multiplied in the mice. A pseudomonas phage protected mice against 5 LD50 of a virulent strain of P. aeruginosa, with a PD50 of 1.2 × 107 particles. A staphylococcal phage failed to protect mice infected with a strain of S. aureus. These studies support the view that bacteriophages could be useful in the treatment of human infections caused by antibiotic-resistant strains of bacteria.

A total of 147 clinical and environmental isolates of Aeromonas that included 14 A. hydrophila, 60 A. sobria and 73 A. caviae strains was tested for haemolysin production and its correlation with enterotoxicity; 108 isolates produced β-haemolysis. For A. hydrophila and A. sobria, titres of haemolysin were 16–128 HU/ml and for A. caviae, 16–64 HU/ml. In the ileal loop test, 82 (55.8%) strains of Aeromonas spp. produced enterotoxin. Of the β-haemolytic strains, 72.7% of A. hydrophila, 58.6% of A. sobria and 68.6% of A. caviae isolates caused fluid accumulation in rabbit ileal loops. One strain each of α-haemolytic A. sobria and A. caviae, one of non-haemolytic A. sobria and nine of non-haemolytic A. caviae also caused a secretory response. The β-haemolytic strains caused significantly more (p < 0.05) fluid accumulation than the α-and non-haemolytic isolates regardless of their species designation. The remaining 65 (44.2%) isolates belonging to the three species included α-, β-and non-haemolytic strains: they failed to cause fluid accumulation in the initial experiments but did so after one to three consecutive passages through rabbit ileal loops. Two α-and 13 non-haemolytic strains switched to production of β-haemolysis when they showed positive ileal loop reactions. However, on repeated subcultures or on storage in the laboratory, all of them reverted to their original haemolytic character and no longer produced enterotoxic activity.

The association of Haemophilus ducreyi with epithelial cell cultures was studied by light microscopy, electronmicroscopy and viable counts. Associated organisms were engulfed by epithelial cells and sequestered from the cell-surface environment. Large numbers of organisms within epithelial cells appeared to induce cell lysis and release of H. ducreyi. Such a mechanism occurring in vivo may assist H. ducreyi to evade the bactericidal action of polymorphonuclear leucocytes and may explain some of the tissue damage seen in genital ulcers caused by H. ducreyi.

Collection strains (21) and non-pigmented clinical isolates (96) provisionally identified as Prevotella spp. were classified numerically on the basis of pyrolysis mass spectrometry (PMS) data and reaction patterns in conventional tests (CTRPs) for volatile and non-volatile fatty acids, pre-formed enzymes and biochemical activity. PMS and CTRP classifications were compared with a previous classification based on visual analysis of SDS-PAGE patterns. Although the order of clusters differed, cross-tabulation of cluster membership revealed strong correlations between classifications. Cluster membership in the PMS classification correlated particularly well with SDS-PAGE results. CTRP clusters corresponded largely to the recognised species of Prevotella, but PMS and SDS-PAGE divided two species into sub-groups: two in P. buccae and five in P. veroralis. The latter subgroups could be discriminated by small but consistent differences in CTRPs. An undesignated, well differentiated cluster of strains appeared closest to the main group of P. buccae strains in PMS and CTRPs. B. (P.) capillus could not be distinguished from P. buccae; these species are regarded as synonymous. Strains of P. zoogleoformans and B. (P.) pentosaceus were well separated from other strains in PMS. A complex comprising clusters of P. disiens, P. oralis, P. veroralis, P. loescheii and a further undesignated group similar to P. melaninogenica was well differentiated from P. buccae and P. oris in PMS; clusters corresponding to P. bivia, P. corporis, P. intermedia and P. denticola formed another complex.

The total blood lymphocyte proliferation assay (TLP) was evaluated as a screening test for infection with Mycobacterium tuberculosis and was compared with the tuberculin (Mantoux) skin test. The results of TLP assays performed on 33 patients with tuberculosis and 37 non-tuberculous subjects were compared with results of skin tests performed in the previous year. There was a high correlation between skin test responses and TLP responses to PPD which was statistically significant. The sensitivity, specificity and the predictive value of a positive test were also similar for the skin test and TLP test. These findings suggest that the TLP test is as effective in screening for M. tuberculosis infection as tuberculin skin testing. Future research leading to further simplification of the TLP method may lead to it replacing intradermal skin testing.

A system was developed for the identification of mycobacteria such as Mycobacterium tuberculosis and M. avium, by thin layer chromatography of 55Fe-labelled mycobactin. Approximately 2 × 103 mycobacteria were detected within 24 h and little operator time or skill was required. M. avium, M. intracellulare and M. scrofulaceum were found to have lower requirements for iron than other mycobacteria and this may influence their growth in host organisms.

The rat was evaluated as an experimental model for disseminated candidosis by quantitating blood clearance and initial organ localisation of 3H-leucine-labelled Candida albicans after intravenous injection into the tail or portal vein. Viable or formalin-killed blastoconidia or viable blastoconidia with germ tubes were injected into experimental animals. Blood and tissue samples were obtained up to 24 h after injection and processed for liquid scintillation counting (to determine the distribution of labelled yeasts) and quantitation of viable organisms. Yeasts were cleared rapidly after intravenous (i.v.) injection by either route, i.e., < 5% of the radioactivity was detected in the blood after 5 min. The liver and lung were the major organs that sequestered blood-borne yeasts 1 h after tail vein injection (42.5 SD 15% and 41.4 SD 6.4% of labelled yeasts injected, respectively). However, injections via the portal vein resulted in trapping of the yeasts predominantly by the liver. Recovery of radioactivity and viable yeasts from all organs except the kidneys decreased with time. Overall, the results indicated that the rat might serve as a reliable model for short-term studies on organ distribution and thus contribute to our understanding of tissue trophism in candidosis.