Search

Variability in the lipopolysaccharide (LPS) of the two most prevalent Salmonella serotypes causing food-borne salmonellosis was assessed using gas chromatography analysis of neutral sugars from 43 Salmonella enterica serovar Enteritidis (S. Enteritidis) and 20 Salmonella enterica serovar Typhimurium (S. Typhimurium) isolates. Four substantially different types of O-chain chemotypes were detected using cluster analysis of sugar compositions; these were low-molecular-mass (LMM) LPS, glucosylated LMM LPS, high-molecular-mass (HMM) LPS and glucosylated HMM LPS. Nineteen out of 20 S. Typhimurium isolates yielded glucosylated LMM. In contrast, S. Enteritidis produced a more diverse structure, which varied according to the source and history of the isolate: 45.5% of egg isolates yielded glucosylated HMM LPS; 100% of stored strains lacked glucosylation but retained chain length in some cases; and 83.3% of fresh isolates from the naturally infected house mouse Mus musculus produced glucosylated LMM LPS. A chain length determinant (wzz) mutant of S. Enteritidis produced a structure similar to that of S. Typhimurium and was used to define what constituted significant differences in structure using cluster analysis. Fine mapping of the S. Enteritidis chromosome by means of a two-restriction enzyme-ribotyping technique suggested that mouse isolates producing glucosylated LMM LPS were closely related to orally invasive strains obtained from eggs, and that stored strains were accumulating genetic changes that correlated with suppression of LPS O-chain glucosylation. These results suggest that the determination of LPS chemotype is a useful tool for epidemiological monitoring of S. Enteritidis, which displays an unusual degree of diversity in its LPS O-chain.

S. typhimurium isolates obtained during a large outbreak of human salmonellosis associated with smoked mackerels in the Czech Republic as well as strains of S. typhimurium isolated from black headed gull (Larus ridibundus) were examined following biotyping, phage typing, plasmid profiling and restriction endonuclease analysis (Eco RI, Hind III and Bam HI) of plasmid DNA. The epidemic strain of S. typhimurium and two isolates from environment of nesting colony black-headed gull were meso-inositol and L-rhamnose negative, phage type 141. The isolates from human and environment of nesting colony were found to share the same plasmid profile and REA.

A L-rhamnose transport-negative strain of Escherichia coli was generated by Mu d(ApR,lac)I mutagenesis. This strain was used to isolate a clone of Salmonella typhimurium DNA that encoded L-rhamnose-H+ transport activity, the gene for which, rhaT, was sequenced. The rhaT gene was mapped on the E. coli chromosome between rhaR and sodA at 87.9 min, initially by Southern blot analysis and then by the isolation, expression, and sequencing of the rhaT gene. Both rhaT genes encoded a hydrophobic protein of 344 amino acids (91% identical) that contained 10 putative transmembrane regions. The RhaT protein represents a novel class of sugar transport protein.

A Salmonella typhimurium LT2 cosmid library was constructed, and a 46-kb recombinant plasmid was isolated that could complement an Escherichia coli rha mutant. Subsequent subcloning generated a 13.6-kb ClaI restriction fragment that contained a functional regulatory element. By complementation analyses using different subclones, the approximate physical locations of rhaT, rhaC1, rhaC2, rhaB, rhaA, and rhaD were determined. The nucleotide sequence spanning the rhaB and rhaC2 genes was determined.

L-Rhamnose is degraded by strains of Salmonella typhimurium by isomerisation to L- rhamnulose , phosphorylation to L- rhamnulose -1-phosphate and cleavage to lactaldehyde and dihydroxyacetone phosphate. The enzymes involved are, respectively, rhamnose isomerase ( RhaI ), rhamnulokinase ( RhuK ) and an aldolase (Ald). Strains able to grow rapidly on L-rhamnose contained a high-affinity uptake system for 3H-L-rhamnose that was induced by L-rhamnose and repressed by D-glucose. The synthesis of RhaI and RhuK was also induced by L-rhamnose but was not repressed by D-glucose. The synthesis of Ald was constitutive. Data are presented on some strains which grow very slowly on L-rhamnose and on others which do not utilise it.

When grown anaerobically on L-rhamnose, Salmonella typhimurium excreted 1,2-propanediol as a fermentation product. Upon exhaustion of the methyl pentose, 1,2-propanediol was recaptured and further metabolized, provided the culture was kept under anaerobic conditions. n-Propanol and propionate were found in the medium as end products of this process at concentrations one-half that of 1,2-propanediol. As in Klebsiella pneumoniae (T. Toraya, S. Honda, and S. Fukui, J. Bacteriol. 139:39-47, 1979), a diol dehydratase which transforms 1,2-propanediol to propionaldehyde and the enzymes involved in a dismutation that converts propionaldehyde to n-propanol and propionate were induced in S. typhimurium cultures able to transform 1,2-propanediol anaerobically.

The effect of hydrofluoric acid (aqueous 48% HF) upon different lipopolysaccharides (LPS) was studied, employing conditions (48 h at +4 degrees C) that are commonly used to dephosphorylate LPS. From the LPS of Salmonella typhimurium having the O antigen 4,5,12 almost all of the O-antigenic sugars (Abe, Gal, Glc, Man, Rha) were liberated in dialysable form, whereas the saccharide chains of Salmonella LPS with O antigen 6,7 (Man, Glc, GlcNAc) were resistant to HF. The lability towards HF was shown to be due to the presence of the deoxysugar L-rhamnose in the saccharide backbone of the O antigen 4,5,12, since only Rha was found as the terminal sugar in the corresponding dialysable material. Hydrofluoric acid can thus be used to specifically cleave Rha-containing polysaccharides.

A new, two-tier system for biotyping Salmonella typhimurium gives a finer and more reliable differentiation of strains than the Kristensen scheme and is capable of future extension by the addition of new types and new tests. Strains are allocated to a primary type (1-32) by their reactions in five primary tests with Bitter's xylose medium, meso-inositol, L-rhamnose, d-tartrate and m-tartrate. Subtypes are distinguished within the primary types by reactions in ten secondary tests, which include observations for flagella and type-1 (haemagglutinating) fimbriae. Full biotypes are designated by letters indicating the subtype reactions appended to the primary-type numbers. A series of 2030 strains of S. typhimurium collected from many different sources and countries during 53 years was classified into 19 of the 32 potential primary biotypes and into 144 full biotypes. Of the series, 14% (275) were non-fimbriate inositol-nonfermenting rhamnose-nonfermenting (FIRN) strain in primary biotypes 29-32. Most other strains were fimbriate and rhamnose fermenting. Observations on several series of cultures isolated from different human or animal sources in the same epidemic showed that the biotype characters of a strain were generally stable during its growth in the natural environment and in the unselective media used for isolation and storage. Most non-fermenting strains gave rise to fermenting mutants on prolonged incubation in the substrate-containing--and therefore selective--test medium, and false-positive results from this cause were avoided by making the definitive readings of tests after a short, carefully chosen period of incubation. A genealogical tree has been drawn to show how eighteen observed primary biotypes may have evolved from a presumed archetypal ancestor of biotype 1.

Lysogenization of Salmonella typhimurium with either of the bacteriophages A3 and A4 results in O-acetylation of the L-rhamnose residues of the O-polysaccharide chain of the lipopolysaccharide of the bacterial cell envelope. The O-acetyl group is found on both O-2 and O-3 of the L-rhamnosyl residues. This lysogenic conversion prevents the adsorption of the A3 and A4 phages and also greatly reduces the rate of adsorption of phage P22 to the O-polysaccharide chain as measured by binding studies with whole bacteria. Isolated lipopolysaccharide from A3- and A4-lysogenized bacteria was also inefficient in inactivating these phages: the concentration required for 50% inactivation was 10,000-fold higher than that for lipopolysaccharide from S. typhimurium not lysogenized by any A phage. Binding of phages A3 and A4 is accompanied by hydrolysis of the alpha-1,3 linkage between rhamnose and galactose in the tetrasaccharide repeating unit of the O-polysaccharide. Phage hydrolysis generates saccharides of various lengths, the majority being dodecasaccharides, i.e., equivalent to three repeating units. It is surmised that O-acetylation of the rhamnosyl residue interferes with phage A3, A4, and P22 infection by preventing binding to and hydrolysis of the O-polysaccharide chain, the initial step in the phage infection cycle. The new O-acetyl-rhamnose entities did not elicit specific antibodies in rabbits in accordance with earlier experiences. The O-acetylation of O-2 and O-3 of rhamnose is a new, hitherto unknown, modification of the O-polysaccharide chain of S. typhimurium.

An equimolar amount of 1,2-propanediol was detected in the medium when Salmonella typhimurium or Klebsiella pneumoniae fermented L-fucose or L-rhamnose. These metabolic conditions induced a propanediol oxidoreductase that converted the lactaldehyde formed in the dissimilation of either sugar into the diol. The enzyme was further identified by cross-reaction with antibodies against Escherichia coli propanediol oxidoreductase. This indicates that L-fucose and L-rhamnose fermentation takes place in these species by 1,2-propanediol production and excretion.

A total of 28 L-rhamnose-negative mutants in Salmonella typhimurium LT2 were all linked by P22 transduction and were classified into five groups on the basis of genetic and biochemical experiments. Deletion mapping showed that the gene order was rhaD rhaA rhaB rhaC rhaT . rhaA mutants lacked an inducible L-rhamnose isomerase, rhaB mutants lacked an inducible L- rhamnulokinase , and rhaD mutants were probably defective in L- rhamnulose -1-phosphate aldolase. Mutants that were unable to accumulate L-[14C]rhamnose but could grow on 1% L-rhamnose were designated rhaT to indicate a defect in L-rhamnose transport. Genetic evidence supports the hypothesis that the rhaC gene is a positive regulator of rha gene expression. (i) Pleiotropically negative mutants in the rhaC gene were isolated at a high frequency. (ii) Mutants containing an insertion or deletion within the rhaC gene had a pleiotropically negative phenotype. (iii) Complementation tests indicated that rhaC + was dominant to rhaC -. (iv) Rha+ revertants of deletion and Tn10 insertion mutations in the rhaC gene were isolated.

For each of the isomers of tartaric acid, meso- or d- or l-, a pair of strains of Salmonella typhimurium was obtained, the one, a naturally occurring, non-fermenting strain and the other a spontaneous, tartrate-fermenting mutant derived from it. For each of the pairs of strains, competitive mixed cultures were grown from inocula of the tartrate-fermenting and tartrate-non-fermenting strains in peptone medium without or with the appropriate tartrate isomer. In each experiment, small numbers of tartrate-fermenting bacteria outgrew small or large numbers of tartrate-non-fermenting bacteria in 24 hours in tartrate-containing but not in tartrate-free peptone medium. The results of the experiments are discussed with reference to the choice of the definitive time of reading for tartrate-utilisation tests in the biotyping of S. typhimurium.

Salmonella typhimurium excreted the rhamnose fermentation product 1,2-propanediol not only under anaerobic conditions, but also under aerobic conditions. The absence of an aldehyde dehydrogenase enzymatic activity that oxidizes to lactate the lactaldehyde formed in the dissimilation of rhamnose raised the intracellular concentration of the aldehyde which was alternatively reduced to the excretable 1,2-propanediol by a residual propanediol oxidoreductase activity.

An efficient technique for biotyping, using disposable trays instead of glass tubes and agar instead of liquid media, is described.

A total of 1537 strains of Salmonella typhimurium belonging to seven prevalent phage types were examined on solid media for their ability to ferment rhamnose, xylose and inositol, for colicine production and for nutritional requirements. Most of the strains in each phage type were almost completely homogeneous, especially in their sugar fermentation reactions. However, strains of phage type 1 a/2 were not homogeneous, but could be assigned to one of four subgroups on the basis of ability to ferment inositol, inhibition of growth by meso-tartrate and auxotrophy for nicotinic acid. The subdivision proved to have epidemiological value. The inhibition of growth by meso-tartrate was observed on a defined medium containing citrate as the energy source. Inhibition did not occur if glucose, casein hydrolysate or aspartic acid were added to the medium.

Salmonella typhimurium strains from natural sources either ferment or do not ferment meso-inositol in peptone water in 24 hr at 37 C. Ninety-five percent of the strains that are designated inositol-nonfermenting on the basis of their phenotype at 37 C ferment inositol when incubated at 25 C. Two classes of temperature-sensitive mutants were detected among the 712 strains of S. typhimurium examined. The occurrence of low-temperature fermentation of inositol among wild-type strains of S. typhimurium from biotypes 10 through 13 and FIRN suggested a genealogical relationship between these two groups, and that FIRN strains (fim(-)inl(ts)rha(-)) might have descended from ancestral types like biotype 10 through 13 strains (fim(+)inl(ts)rha(+)).