John Elmerdahl Olsen

Host adapted serotypes of Salmonella enterica.

Salmonella constitutes a genus of zoonotic bacteria of worldwide economic and health importance. The current view of salmonella taxonomy assigns the members of this genus to two species: S. enterica and S. bongori. S. enterica itself is divided into six subspecies, enterica, salamae, arizonae, diarizonae, indica, and houtenae, also known as subspecies I, II, IIIa, IIIb, IV, and VI, respectively [1]. Members of Salmonella enterica subspecies enterica are mainly associated with warm-blooded vertebrates and are usually transmitted by ingestion of food or water contaminated by infected faeces. The pathogenicity of most of the distinct serotypes remains undefined, and even within the most common serotypes, many questions remain to be answered regarding the interactions between the organism and the infected host.Salmonellosis manifests itself in three major forms: enteritis, septicaemia, and abortion, each of which may be present singly or in combination, depending on both the serotype and the host involved. Although currently over 2300 serovars of Salmonella are recognized, only about 50 serotypes are isolated in any significant numbers as human or animal pathogens [2, 3] and they all belong to subspecies enterica. Of these, most cause acute gastroenteritis characterized by a short incubation period and a severe systemic disease in man or animals, characterized by septicaemia, fever and/or abortion, and such serotypes are often associated with one or few host species [4–6].It is the intention of this review to present a summary of current knowledge of these host-adapted serotypes of S. enterica. The taxonomic relationships between the serotypes will be discussed together with a comparison of the pathology and pathogenesis of the disease that they cause in their natural host(s). Since much of our knowledge on salmonellosis is based on the results of work on Typhimurium, this serotype will often be used as the baseline in discussion. It is hoped that an appreciation of the differences that exist in the way these serotypes interact with the host will lead to a greater understanding of the complex host–parasite relationship that characterizes salmonella infections.

Clonal lines of Salmonella enterica serotype enteritidis documented by IS200-, ribo-, pulsed-field gel electrophoresis and RFLP typing

Sixty-two selected strains of Salmonella serotype Enteritidis of 33 phage types (PTs), and one strain classified as RDNC, were characterised by four different chromosomally based typing methods to elucidate genetic relationships among strains of different phage types. Based on IS200-hybridisation patterns, two major groups, containing strains of the most commonly encountered phage types, and six minor groups (seven with the RDNC strain included) were observed. IS200 pattern was a stable epidemiological marker in strains of all phage types except PT 6a and 14b. Ribotyping separated strains of the phage types into one major and five minor groups; the pattern of the RDNC strain was not seen with other strains. More than one ribotype was observed among strains of Enteritidis PTs 6, 7, 14b and 21. By pulsed-field gel electrophoresis, strains of 21 of the 33 phage types formed one large cluster when bands > 125 kb were used as the criterion for separation. Among strains belonging to PTs 1, 6, 7 and 14b, more than one pattern was observed by this method. By probing with five random cloned fragments of the Enteritidis chromosome, strains from 27 of 31 phage types examined showed the same hybridisation pattern. With the combined use of four genotypic methods, two groups of strains, representing eight and seven of 33 Enteritidis phage types, were formed; these two groups may be considered as the main evolutionary lines of Enteritidis. Strains of the remaining phage types, and the RDNC strain, belonged to separate groups.

Listeria monocytogenes exists in at least three evolutionary lines: evidence from flagellin, invasive associated protein and listeriolysin O genes.

Regions of the genes encoding flagellin (flaA), the invasive associated protein (iap), listeriolysin O (hly) and 23S rRNA were sequenced for a range of Listeria monocytogenes isolates of different origin and serotypes. Several nucleotide sequence variations were found in the flaA, iap and hly genes. No differences were found for the rRNA genes, but our approach does not exclude the existence of differences between single copies of these genes. Based on the sequence differences, the L. monocytogenes strains can be divided into three distinct sequence types. Further, the presence of only a small number of sequence differences within each group indicates a strong degree of conservation within the groups. There was a complete correspondence among the groups of strains formed according to the analysis of the flaA, iap and hly genes, and the grouping correlates with serotype, pulsed field gel electrophoretic and multilocus enzyme electrophoretic data. Analysis of the region encoding the threonine-asparagine repeat units in the iap gene revealed some striking features. Sequence type 1 strains were found to have 16-17 repeats, sequence type 2 strains had 16-20 repeats whereas the two sequence type 3 strains analysed had only 11 repeats. Furthermore, within a 19 bp segment there was a 37% difference between the sequences of type 1 and 2 strains and that segment was absent in type 3 strains. Within the threonine-asparagine repeat region the nucleotide differences gave rise to four amino acid changes; however, all were changes among the three amino acids present in the repeat structure indicating a strong selective pressure on the composition of this region.

Salmonella identification by the polymerase chain reaction

Polymerase chain reaction (PCR) primers for genus specific detection of Salmonella have been selected from a Salmonella-specific fragment of 2.3 kilobases (kb). Due to interserovar sequence diversity within this fragment, primer selection was based on DNA sequence alignment of sequences from 20 different Salmonella serovars. The specific PCR product of 429 base pairs (bp) was formed from 144 of 146 salmonella strains tested (116 of 118 serovars). The two false-negative strains belonged to two different serovars of the rarely isolated subspecies IIIa (monophasic S. arizonae). No product was produced in any of 86 non-Salmonella Enterobacteriacea strains tested, covering 41 species from 21 genera.

DNA-DNA hybridization determined in micro-wells using covalent attachment of DNA.

The present study was aimed at reducing the time and labour used to perform DNA-DNA hybridizations for classification of bacteria at the species level. A micro-well-format DNA hybridization method was developed and validated. DNA extractions were performed by a small-scale method and DNA was sheared mechanically into fragments of between 400 and 700 bases. The hybridization conditions were calibrated according to DNA similarities obtained by the spectrophotometric method using strains within the family Pasteurellaceae. Optimal conditions were obtained with 300 ng DNA added per well and bound by covalent attachment to NucleoLink. Hybridization was performed with 500 ng DNA, 5% (w/w) of which was labelled with photo-activatable biotin (competitive hybridization) for 2.5 h at 65 degrees C in 2 x SSC followed by stringent washing with 2 x SSC at the same temperature. The criteria for acceptance of results were a maximum of 15% standard deviation, calculated as a percentage of the mean for four replicate micro-wells, and that DNA similarities were not significantly different in at least two independent experiments. The relationship between DNA similarities obtained by the micro-well method (y) and by the spectrophotometric method (x) was y = 0.534x+30.6, when these criteria had been applied to 23 pairs of strains of Actinobacillus species, avian [Pasteurella] haemolytica-like bacteria and Mannheimia species. The correlation (Pearson) between DNA similarities obtained by interchange of strains used for covalent binding and hybridization was 0.794. Significantly lower DNA similarities were observed by the spectrophotometric compared with the micro-well method for three pairs of hybridizations. After removal of these data, the relationship between DNA similarities obtained by the micro-well and spectrophotometric methods improved to y = 0.855x + 11.0. It was found that the accuracy and precision of the micro-well method was at the same level as that of the spectrophotometric method, but the labour and analysis time were reduced significantly. The use of hybridization in the micro-well format will allow DNA-DNA hybridizations to be carried out between all strains selected for a particular taxonomic study, in order to construct complete data matrices and improve species definition.

Distribution and in-vitro transfer of tetracycline resistance determinants in clinical and aquatic Acinetobacter strains.

Following characterisation by phenotypic tests and amplified ribosomal DNA restriction analysis (ARDRA), 50 tetracycline-resistant (MIC > or = 16 mg/L) Acinetobacter strains from clinical (n = 35) and aquatic (n = 15) samples were analysed by PCR for tetracycline resistance (Tet) determinants of classes A-E. All the clinical strains were A. baumannii; most (33 of 35) had Tet A (n = 16) or B (n = 17) determinants, and only two did not yield amplicons with primers for any of the five tetracycline resistance determinants. The aquatic strains belonged to genomic species other than A. baumannii, and most (12 of 15) did not contain determinants Tet A-E. Strains negative for Tet A-E were also negative for Tet G and M; further analysis of two aquatic strains with specific primers for Tet O and Tet Y and degenerate primers for Tet M-S-O-P(B)-Q also showed negative results. Transfer of tetracycline resistance was tested for 20 strains with three aquatic Acinetobacter strains and Escherichia coli K-12 as recipients. Transfer of resistance was demonstrated between aquatic strains from distinct ecological niches, but not from clinical to aquatic strains, nor from any Acinetobacter strain to E. coli K-12. Most transconjugants acquired multiple relatively small plasmids (<36 kb). Transfer did not occur when DNA from the donor strains was added to the recipient cultures and was not affected by deoxyribonuclease I, suggesting a conjugative mechanism. It is concluded that Tet A and B are widespread among tetracycline-resistant A. baumannii strains of clinical origin, but unknown genetic determinants are responsible for most tetracycline resistance among aquatic Acinetobacter spp. These differences, together with the inability of clinical strains to transfer tetracycline resistance in vitro to aquatic strains, contra-indicate any important flow of tetracycline resistance genes between clinical and aquatic acinetobacter populations.

Rapid Quantification of Viable Campylobacter Bacteria on Chicken Carcasses, Using Real-Time PCR and Propidium Monoazide Treatment, as a Tool for Quantitative Risk Assessment

ABSTRACT A number of intervention strategies against Campylobacter-contaminated poultry focus on postslaughter reduction of the number of cells, emphasizing the need for rapid and reliable quantitative detection of only viable Campylobacter bacteria. We present a new and rapid quantitative approach to the enumeration of food-borne Campylobacter bacteria that combines real-time quantitative PCR (Q-PCR) with simple propidium monoazide (PMA) sample treatment. In less than 3 h, this method generates a signal from only viable and viable but nonculturable (VBNC) Campylobacter bacteria with an intact membrane. The methods performance was evaluated by assessing the contributions to variability by individual chicken carcass rinse matrices, species of Campylobacter, and differences in efficiency of DNA extraction with differing cell inputs. The method was compared with culture-based enumeration on 50 naturally infected chickens. The cell contents correlated with cycle threshold (CT) values (R2 = 0.993), with a quantification range of 1 × 102 to 1 × 107 CFU/ml. The correlation between the Campylobacter counts obtained by PMA-PCR and culture on naturally contaminated chickens was high (R2 = 0.844). The amplification efficiency of the Q-PCR method was not affected by the chicken rinse matrix or by the species of Campylobacter. No Q-PCR signals were obtained from artificially inoculated chicken rinse when PMA sample treatment was applied. In conclusion, this study presents a rapid tool for producing reliable quantitative data on viable Campylobacter bacteria in chicken carcass rinse. The proposed method does not detect DNA from dead Campylobacter bacteria but recognizes the infectious potential of the VBNC state and is thereby able to assess the effect of control strategies and provide trustworthy data for risk assessment.

Role of Flagellin and the Two-Component CheA/CheY System of Listeria monocytogenes in Host Cell Invasion and Virulence

ABSTRACT The flagellum protein flagellin of Listeria monocytogenes is encoded by the flaA gene. Immediately downstream of flaA, two genes, cheY and cheA, encoding products with homology to chemotaxis proteins of other bacteria, are located. In this study we constructed deletion mutants with mutations in flaA. cheY, and cheA to elucidate their role in the biology of infection with L. monocytogenes. The ΔcheY, ΔcheA, and double-mutant ΔcheYA mutants, but not ΔflaA mutant, were motile in liquid media. However, the ΔcheA mutant had impaired swarming and the ΔcheY and ΔcheYA mutants were unable to swarm on soft agar plates, suggesting that cheY and cheA genes encode proteins involved in chemotaxis. The ΔflaA, ΔcheY, ΔcheA, and ΔcheYA mutants (grown at 24°C) showed reduced association with and invasion of Caco-2 cells compared to the wild-type strain. However, spleens from intragastrically infected BALB/c and C57BL/6 mice showed larger and similar numbers of the ΔflaA and ΔcheYA mutants, respectively, compared to the wild-type controls. Such a discrepancy could be explained by the fact that tumor necrosis factor receptor p55 deficient mice showed dramatically exacerbated susceptibility to the wild-type but unchanged or only slightly increased levels of the ΔflaA or ΔcheYA mutant. In summary, we show that listerial flaA. cheY, and cheA gene products facilitate the initial contact with epithelial cells and contribute to effective invasion but that flaA could also be involved in the triggering of immune responses.

Cloning and characterization of a gene encoding flagellin of Listeria monocytogenes

The gene, flaA, encoding the flagellin protein of Listeria monocytogenes (strain 12067) has been isolated from an expression library in Escherichia coli using a flagellin‐specific monoclonal antibody. DNA sequence analysis of a positive clone revealed the presence of an open reading frame of 287 amino acid residues with a calculated molecular mass of 30.4 kDa. Comparison of this sequence with flagellins from other bacteria showed a significant degree of homology in both the N‐ and C‐terminal parts of the protein. The flagellin mRNA was determined to be 1 kb in size, which is the expected size for a monocistronic mRNA, and the temperature‐dependent expression of flagellin was found to be regulated at the transcriptional level. Southern blot analysis, using the flagellin gene as probe, indicated that L. monocytogenes can be divided into two groups. These groups correspond to the flagellar antigens AB and ABC, respectively, as well as to the two types of L. monocytogenes based on the DNA sequence of the listeriolysin gene.

Phylogenetic relationships of Salmonella based on rRNA sequences

To establish the phylogenetic relationships between the subspecies of Salmonella enterica (official name Salmonella choleraesuis), Salmonella bongori and related members of Enterobacteriaceae, sequence comparison of rRNA was performed by maximum-likelihood analysis. The two Salmonella species were separated by 16S rRNA analysis and found to be closely related to the Escherichia coli and Shigella complex by both 16S and 23S rRNA analyses. The diphasic serotypes S. enterica subspp. I and VI were separated from the monophasic serotypes subspp. IIIa and IV, including S. bongori, by 23S rRNA sequence comparison.