Structural Investigations and Identification of the Extracellular Bacteriolytic Endopeptidase L1 from Lysobacter sp. XL1

The N-terminal amino acid sequence (23 amino acid residues) and the amino acid composition of the extracellular bacteriolytic enzyme L1 of 21 kD from the bacterium Lysobacter sp. XL1 have been determined. The enzyme was hydrolyzed by trypsin, the resulting peptides were isolated, and their primary structures were determined. A high extent of homology (92%) of the N-terminal amino acid sequence and the primary structure of isolated peptides of the enzyme L1 (62 amino acid residues or 31% of protein sequence) to the corresponding sites of alpha-lytic proteinases (EC 3.4.21.12) of Lysobacter enzymogenes and Achromobacter lyticus was found. These data allowed identification of the endopeptidase L1 of Lysobacter sp. XL1 as alpha-lytic proteinase EC 3.4.21.12.     

Production of superoxide dismutases from Proteus mirabilis and Proteus vulgaris

Proteus mirabilis and Proteus vulgaris expressed a combination of superoxide dismutase (Sod) activities, which was assigned to FeSod1, FeSod2 and MnSod for P. mirabilis, and FeSod, MnSod and CuZnSod for P. vulgaris. Production of the Sod proteins was dependent on the availability of iron, whether cells were grown under anaerobiosis or aerobiosis and growth phase. Nalidixic acid and chloramphenicol inhibited cell growth and the iron- and dioxygen-dependent production of Sod. These results support the involvement of metal ions and redox status in the production of Proteus Sods.     

Bacteriophage Typing of Proteus mirabilis, Proteus vulgaris, and Proteus morganii

A bacteriphage typing scheme for differentiating Proteus isolated from clinical specimens was developed. Twenty-one distinct patterns of lysis were seen when 15 bacteriophages isolated on 8 Proteus mirabilis, 1 P. vulgaris, and 1 P. morganii were used to type 162 of 189 (85.7%) P. mirabilis and P. vulgaris isolates. Seven phages isolated on 3 P. morganii were used to type 13 of 19 (68.4%) P. morganii isolates. Overall, 84.1% of the 208 isolates were lysed by at least 1 phage at routine test dilution (RTD) or 1,000 x RTD. Fifty isolates, retyped several weeks after the initial testing, showed no changes in lytic patterns. The phages retained their titers after storage at 4 C for several months. A computer analysis of the data showed that there was no relationship between the source of the isolate and bacteriophage type. This bacteriophage typing system may provide epidemiological information on strains involved in human infections.     

Some biological features of Proteus bacilli. 2. Haemolytic activities of Proteus mirabilis and Proteus vulgaris strains

The haemolytic activities of Proteus mirabilis and P. vulgaris strains were studied under different conditions. No filterable alpha haemolysin could be detected in P. mirabilis uropathogens provided from patients with urinary tract infections. Together with the results presented in the accompanying paper, in which three clinical isolates with temporary ability to produce a soluble haemolysin were described, the occurrence of alpha haemolytic P. mirabilis isolates did not exceed 3%. Cell bound beta haemolysin is present in nearly 35% of P. mirabilis urinary strains. Another kind of haemolytic activity was observed when P. mirabilis and P. vulgaris strains were grown in liquid media supplemented with erythrocytes. During the logarithmic growth phase nearly 100% of P. mirabilis and P. vulgaris strains of various origin haemolyzed 100-50% of erythrocytes. Except for Serratia, the other representatives of Enterobacteriaceae did not demonstrate such activity in the same conditions. The preliminary characteristics of this phenomenon is given.     

Species-specific detection of Proteus vulgaris and Proteus mirabilis by the polymerase chain reaction

Sets of primers for the species-specific detection of P. mirabilis and P. vulgaris by the polymerase chain reaction (PCR) were developed. As targets for these primers beta-lactamase and 16S rRNA gene fragments were chosen on the basis of the multiple leveling of the sequences of the DNA of all known P. mirabilis and P. vulgaris isolates. For differential detection oligonucleotides were selected in such a way that primers, specific for P. vulgaris, contained the non-paired nucleotide for P. mirabilis isolate at the 3'-end, and all other nucleotides were complementary to the beta-lactamase gene fragment. Primers, specific for gene 16S rRNA of P. mirabilis, contained the non-paired nucleotide for P. vulgaris isolates at the 3'-end. Standard PCR was carried out for 6 P. mirabilis and P. vulgaris strains. The use of PCR species-specific primers to P. vulgaris DNA made it possible to amplify the DNA fragment of the expected length only for P. vulgaris isolates, while the result of PCR for P. mirabilis was negative. PCR with primers specific to P. mirabilis permitted the detection of amplicon sized 101 nucleotides pairs only for P. mirabilis strains. These primers were optimized so as to use them in the specific differentiation of closely related P. mirabilis and P. vulgaris species by multiplex PCR. Genus-specific primers permitted the detection of bacterial gyrB gene of the genus Proteus were developed also.     

Ultrastructural characteristics of Proteus vulgaris and Proteus mirabilis cells differing in the capacity for swarming

Specific differences in the structure of colonies and the location of microbial cells in colonies, characteristic for aggregating and nonaggregating genetically related pairs of P. vulgaris and P. mirabilis strains, have been demonstrated by means of transmission and scanning electron microscopy. In calculating the number of flagellae per 100 outlines of microbial bodies revealed in negatively stained preparations, the fact that both aggregating and nonaggregating bacteria possess practically the same number of flagellae, on the average 4-8 flagellae per microbial cell outline, has been established. This fact indicates that the presence of flagellae in microbial cells is unrelated to their capacity for swarming.     

Adhesion of Proteus mirabilis and Proteus vulgaris to uroepithelial cells following exposure to various antimicrobial agents.

The effect of subinhibitory concentrations of netilmicin, ceftriaxone, cefotaxime, aztreonam and piperacillin on the adherence of Proteus species to uroepithelial cells was examined. Bacterial adhesion to human uroepithelial cells, measured microscopically, was affected by all five antibiotics but to different extents. The most effective was netilmicin. There was a correlation between the decreased rate of bacterial attachment and morphological changes in the drug-exposed bacteria.     

Optimization of in vivo crosslinking technique for the study of AlpB-protein interactions in Lysobacter sp. XL1 cells

The bacterium Lysobacter species strain XL1 is known as a producer of extracellular lytic enzymes, which are capable of degrading cell wall components of other bacteria and simple eukaryotes. This ability determines the ecological, medical and agricultural relevance of Lysobacter sp. XL1. However, the molecular mechanism of secretion of lytic exoenzymes from Lysobacter cells is yet unknown, which in turn necessitates the search of protein–protein interactions that occur during exoenzyme secretion. The current paper is concerned with investigation of protein complexes that are likely formed during the secretion of AlpB lytic protease from the cells of Lysobacter sp. XL1. In this study, we have optimized the method of stabilization of protein complexes formed in the intact cells of Lysobacter sp. XL1 by using crosslinking reagent dithiobis(succinimidylpropionate) (DSP) and detected DSP-linked protein complexes by the monoclonal antibodies against AlpB propeptide.     

Effect of R-plasmid RP1 on surface hydrophobicity of Proteus mirabilis

The presence of R-plasmid RP1, as well as the conditions of growth, affected the surface hydrophobicity of a clinical isolate of Proteus mirabilis. However, results depended upon the method of assessment. Stationary phase plasmid-containing cells appeared to be less hydrophobic than plasmid-free cells when hydrophobicity was measured by the contact angle method, but more hydrophobic when measured by bacterial adherence to hydrocarbons or hydrophobic interaction chromatography. Cells growing in a chemostat differed in hydrophobicity from stationary phase cells and results varied with the growth rate. Plasmid-mediated effects were greatest in iron-depleted cells, and differences between plasmid-containing and plasmid-free cells were virtually eliminated by pre-treatment with antiserum.     

The effect of the extracellular bacteriolytic enzymes of Lysobacter sp. on gram-negative bacteria

The effect of the extracellular bacteriolytic enzymes of Lysobacter sp. on gram-negative bacteria was studied. These enzymes were found to be able to hydrolyze the peptidoglycan that was isolated from the gram-negative bacteria, the hydrolysis being completely inhibited by the cell wall lipopolysaccharide of these bacteria. The native cells of the gram-negative bacteria became susceptible to the bacteriolytic enzymes after the permeability of the outer membrane of the cells had been altered by treating them with polymyxin B.     

Secretion of bacteriolytic endopeptidase L5 of Lysobacter sp. XL1 into the medium by means of outer membrane vesicles

The Gram-negative bacterium Lysobacter sp. XL1 secretes various proteins, including bacteriolytic enzymes (L1-L5), into the culture medium. These proteins are able to degrade Gram-positive bacteria. The mechanism of secretion of extracellular proteins by Lysobacter sp. XL1 has not been studied hitherto. Electron microscopic investigations revealed the phenomenon of the formation of extracellular vesicles by Lysobacter sp. XL1. These vesicles contained components of the Lysobacter sp. XL1 outer membrane, and demonstrated bacteriolytic activity against Gram-positive and Gram-negative bacteria: Staphylococcus aureus 209-P and Erwinia marcescens EC1, respectively. Western blotting analysis with antibodies to homologous bacteriolytic endopeptidases L1 and L5 showed that endopeptidase L5 was secreted into the culture medium by means of vesicles, unlike its homolog, endopeptidase L1. When inside the vesicles, endopeptidase L5 actively lysed the Gram-negative bacterium Erwinia marcescens; outside the vesicles, it lost this ability. The secretion of bacteriolytic endopeptidase L5 through the outer membrane vesicles is of great biological significance: because of this ability, Lysobacter sp. XL1 can compete in nature with both Gram-positive and Gram-negative bacteria.