Previous Article | Next Article ![]()
Journal of Clinical Microbiology, October 2004, p. 4843-4845, Vol. 42, No. 10
0095-1137/04/$08.00+0 DOI: 10.1128/JCM.42.10.4843-4845.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Laboratory of Enteric Pathogens,1 Virus Reference Division, Specialist and Reference Microbiology Division, Health Protection Agency, London, United Kingdom2
Received 16 March 2004/ Returned for modification 22 April 2004/ Accepted 8 June 2004
|
|
|---|
|
|
|---|
While pulsed-field gel electrophoresis (PGFE) has shown some diversity among MR DT104 isolates (6, 7), use of the endonuclease XbaI has shown that the majority (
90%) have a single PFGE profile, xtm 1 (4, 5). Supplementary methods such as resistance typing and plasmid analysis generally provide little additional discrimination (9).
Fluorescent amplified fragment length polymorphism (FAFLP) analysis is a highly discriminatory and reproducible tool for subtyping genetically homogeneous genomes and identifying outbreak genotypes within bacterial genera (1, 2). In this report, the genotypic variation among MR DT104 isolates with the predominant PFGE profile xtm 1 was analyzed by FAFLP.
MR DT104 isolates, all with PFGE profile xtm 1, collected from humans and animals in England and Wales were examined. These comprised randomly selected examples from four distinct outbreaks (n = 45), sporadic isolates (n = 34), and "historic" isolates (n = 6) (Table 1). PFGE, resistance typing, and plasmid analysis were performed as previously described (5).
|
View this table: [in a new window] |
TABLE 1. Isolates of MR S. enterica serovar Typhimurium DT104a used in this study
|
![]() View larger version (16K): [in a new window] |
FIG. 1. Dendrogram showing variability of FAFLP patterns among MR S. enterica serovar Typhimurium DT104 isolates from FAFLP data obtained with the HindIII+0 and HhaI+C primer pair. The dendrogram was derived by UPGMA as previously described (1, 2). Each branch represents a unique FAFLP profile (profile numbers indicate different strains; Table 1). Scale bar, 10% divergence.
|
Outbreak 2 involved delegates who became ill following a conference. Epidemiologically this appeared to be a point source outbreak. However, FAFLP analysis showed three different profiles among the five isolates. Profiles F7 and F8 differed from each other by four AFs and from profile F3 (shared by three isolates) by two AFs each. This suggests that the cases with unique FAFLP profiles may have acquired their infections elsewhere or that multiple strains were present at the onset of the outbreak.
Outbreak 3 was a major milk-borne outbreak in northwest Lancashire, involving more than 86 cases. The isolates were characterized by additional reduced susceptibility to ciprofloxacin (CpL), caused by a point mutation at codon 87 of gyrA (10). Of 10 isolates analyzed by FAFLP, 3 were from cattle, 1 was from a milk filter, and 6 were from human cases. All shared profile F3, including one cow isolate that possessed an additional 40-MDa plasmid not seen among the other outbreak 3 isolates.
Outbreak 4 involved over 360 laboratory-confirmed cases nationwide (3); the isolates were additionally characterized by the presence of an unusual plasmid "fingerprint" (60 and 2 MDa), usually present in
5% of MR DT104 isolates (3). For FAFLP analysis 26 human isolates were selected, and all but one shared profile F3. The remaining isolate showed a closely related profile, F13, which differed from profile F3 by one AF.
The 34 contemporaneous isolates with PFGE profile xtm 1 were included for comparison with the outbreak-associated isolates; 28 (82%) of these had the common profile F3. Eight had resistance type ACSSuT, with a 60-MDa plasmid; seven of these had profile F3, and one had profile F5 (differing by six AFs from F3). A further seven isolates had resistance type ACSSuTCpL (also associated with outbreak 3); six of these had profile F3, and one had a closely related profile, F6 (differing from profile F3 by a single AF). The 60- and 2-MDa plasmid fingerprint characteristic of outbreak 4 was present in 20 sporadic isolates; 16 had profile F3, and 4 had distinct profiles: profiles F9 and F10 were from animal isolates, whereas F11 and F12 were from humans.
The six historic isolates were indistinguishable by PFGE, resistance type, and plasmid fingerprint. However, four FAFLP profiles were detected, with only three (50%) isolates sharing the predominant profile F3. The two earliest isolates, from 1984, showed profiles F1 and F2 (differing from F3 and each other by one AF). Two isolates from 1985, one from a human and one from a cow, had F3 and F4 respectively. Profile F4 was the most distinct in this study, differing from the other profiles in approximately two-thirds of the 118 AFs constituting its profile. The remaining isolates, from 1986 and 1988, had profile F3. It is possible that differences observed in FAFLP profiles in the historic isolates occurred due to degradation during storage, but this seems unlikely as serotype, phage type, PFGE profile, plasmids, and resistance type remained unaffected. Reproducibility studies confirm the stability of FAFLP profiles over prolonged storage, subculture, and use of three repeat DNA preparations from the same isolate (results not shown).
In summary, the predominant FAFLP genotype, F3, was present in one-half of the earliest (1980s) MR DT104 isolates from England and Wales. Eighteen years later, the same FAFLP genotype remains prevalent, having caused outbreaks of human illness throughout the 1990s and into the present decade. A small proportion of MR DT104 isolates appear to be genetically heterogeneous, exhibiting unique FAFLP genotypes. Clearly, these FAFLP genotypes possess pathogenic potential and are associated with human enteric disease, but, in this small survey, they were found only as unique types and not in an outbreak context.
In the present study all isolates were identical when hierarchically characterized by serology, phage type, and PFGE. This is in contrast to an earlier study where the PFGE and FAFLP profiles of S. enterica serovar Typhimurium isolates of unknown phage type were compared (8). It is important to note that the predominant PFGE profile, xtm 1, occurs in phage types other than DT104 (4, 5). While the investigation demonstrates the value of subtyping methods, such as resistance typing and plasmid analysis, in the identification of outbreaks caused by common phage types, it also emphasizes the need for high-resolution genotyping methodologies, such as FAFLP, to provide further insight into the microepidemiology of MR DT104 and other salmonellas.
Special thanks to Linda Ward and our colleagues in the Laboratory of Enteric Pathogens for the serotype, phage type, and resistance type data.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»