HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Supplementary Figure Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Hopkins, K. L. Articles by Threlfall, E. J. Search for Related Content PubMed PubMed Citation Articles by Hopkins, K. L. Articles by Threlfall, E. J. Agricola Articles by Hopkins, K. L. Articles by Threlfall, E. J.

Frequency and polymorphism of sopE in isolates of Salmonella enterica belonging to the ten most prevalent serotypes in England and Wales

Antimicrobial Resistance and Molecular Epidemiology Unit, Laboratory of Enteric Pathogens, Specialist and Reference Microbiology Division, Health Protection Agency, 61 Colindale Avenue, London NW9 5HT, UK

Correspondence Katie L. Hopkins Katie.Hopkins{at}HPA.org.uk

Received October 17, 2003
Accepted February 10, 2004

Translocated effector protein, SopE, leads to actin cytoskeletal rearrangements and membrane ruffling. Only a subset of Salmonella enterica serotypes possess sopE, with the majority of sopE-carrying S. enterica serotype Typhimurium associated with epidemics. Using real-time PCR and sequencing, sopE was investigated in the ten most prevalent serotypes of S. enterica in England and Wales in 2001. sopE was identified in S. Typhimurium definitive phage types 29, 44, 49, 204b and 204c, all of which either have been involved in major epidemics or are precursors of epidemic strains. The presence of sopE varied in the remaining nine serotypes, but was more common in the top four (Enteritidis, Virchow, Hadar and Newport). Nucleotide changes were detected throughout sopE and may result in altered specificity for certain signal transduction pathways. Since acquisition of sopE may play a key role in emergence of epidemic strains, detection of sopE could aid identification of those Salmonella strains with the potential for epidemic spread.

The GenBank accession numbers for the sopE sequences reported in this study are AY167929 (S. Enteritidis), AY167930 (S. Java), AY167931 (S. Infantis), AY168873 (S. Newport), AY168874 (S. Hadar), AY168876 (S. Virchow V1) and AY168875 (S. Virchow V2).

A sopE sequence alignment is available as supplementary material in JMM Online.

	Introduction

TOP Introduction Methods Results and Discussion ACKNOWLEDGEMENTS References

 
Non-typhoidal Salmonella spp. are the second most common cause of bacterial gastroenteritis in developed countries. Given the pattern of emergence of different multidrug-resistant (MDR) strains over the last three decades, concern has been expressed that ‘new’ MDR strains of Salmonella enterica will continue to emerge or old ones may re-emerge (Advisory Committee on the Microbiological Safety of Food, 1999). However, although the virulence factors influencing enteric infection in food production animals have been characterized, their contribution to the emergence and persistence of the strains and their involvement in human infection are poorly understood.

The majority of virulence genes are encoded in Salmonella pathogenicity islands (SPIs), which also encode type-III secretion systems used to deliver translocated effector proteins to host cells. Although the type-III secretion system itself is well conserved among Gram-negative bacteria, the effector proteins appear to be more diverse and may even vary between closely related strains (Galán & Bliska, 1996). SopE is an SPI-1-dependent translocated protein that belongs to a novel class of bacterial toxins that modulate host-cell RhoGTPase function via a non-covalent interaction (Rudolph et al., 1999). It is encoded by a P2-like cryptic bacteriophage (SopE) (Hardt et al., 1998) in S. enterica serotypes Typhimurium and Typhi, but is located on a cryptic -like phage in serotypes Enteritidis, Dublin, Hadar and Gallinarum (Mirold et al., 2001). Microinjection of SopE into cultured cells leads to actin cytoskeletal rearrangements and membrane ruffling that resembles the changes induced by Salmonella infection (Galán & Zhou, 2000). To date, the majority of S. Typhimurium isolates that have been found to carry sopE have belonged to epidemic strains that have persisted in humans and food animals for a long period of time (Mirold et al., 1999).

In this present study, strains from the ten most prevalent serotypes of S. enterica in England and Wales were screened for sopE using a real-time PCR assay developed for the LightCycler. Polymorphisms within the sopE amplicons were investigated by sequencing.

	Methods

TOP Introduction Methods Results and Discussion ACKNOWLEDGEMENTS References

 
Bacterial strains.

The 158 strains of S. Typhimurium represent antimicrobial-sensitive and antimicrobial-resistant phage types that have been epidemic in food-producing animals and humans since the 1960s and phage types that do not appear to have the capacity for epidemic spread. In addition, 20 strains from each of the remaining nine most prevalent serotypes (Enteritidis, Virchow, Hadar, Newport, Infantis, Braenderup, Agona, Java and Stanley) from humans in England and Wales in 2001 (PHLS Salmonella Dataset 2002, unpublished) were tested. Within serotypes Enteritidis, Virchow, Hadar, Agona and Java, the strains were of different phage types but were chosen arbitrarily. S. Typhimurium SARA4 from the SARA Collection (Beltran et al., 1991) was used as a positive control in the PCR and S. Typhimurium LT2 as a negative control.

Serotyping, phage typing and antimicrobial susceptibility testing.

Strains were serotyped (Rowe & Hall, 1989), phage-typed (Threlfall & Frost, 1990) and tested for resistance to ampicillin (A), chloramphenicol (C), gentamicin (G), kanamycin (K), neomycin (Ne), streptomycin (S), spectinomycin (Sp), sulphonamides (Su), tetracyclines (T), trimethoprim (Tm), ciprofloxacin [low-level (Cp_L) and high-level (Cp_H)], nalidixic acid (Nx), furazolidone (Fu), amikacin (Ak), cephalexin (Cx), cephradine (Cr), cefuroxime (Cf), ceftriaxone (Cn) and cefotaxime (Ct) by standard methods (Frost, 1994). The final concentrations of the respective antimicrobials were (µg ml^–1): A, 8; C, 8; G, 4; K, 8; Ne, 8; S, 16; Sp, 64; Su, 64; T, 8; Tm, 2; Cp_L, 0.125; Cp_H, 1; Nx, 16; Fu, 8; Ak, 4; Cx, 16; Cr, 16; Cf, 16; Cn, 1; and Ct, 1.

Preparation of template DNA.

DNA template was prepared from 24-h nutrient broth cultures using the Wizard Genomic DNA purification kit (Promega).

Detection of sopE using a LightCycler.

A 642 bp fragment of sopE was amplified using primers SopE-F (5'-TCAGTTGGAATTGCTGT GGA-3') and SopE-R (5'-TCCAAAAACAGGAAACCACAC-3') designed using the complete sequence of the S. Typhimurium sopE gene (GenBank accession no. AF043239; Hardt et al., 1998) and manufactured by MWG Biotech UK. The PCR was carried out in a 20 µl volume containing 1x FastStart DNA Master SYBR Green I reaction mix (Roche Diagnostics), 3 mM MgCl₂, 0.25 µM (each) SopE-F and SopE-R, 12.4 µl water and 10 ng DNA template. Amplification consisted of 95 °C for 10 min followed by 45 cycles of 95 °C for 10 s, 64 °C for 5 s and 72 °C for 30 s carried out in a LightCycler (Roche Diagnostics). The accumulation of double-stranded PCR product was detected by measuring the fluorescence signals in real time at the end of each elongation step. The PCR product was melted over a temperature range from 65 to 95 °C with a temperature transition rate of 0.1 °C s^–1 with continuous detection of fluorescence to allow discrimination between primer dimers and specific product.

Statistical analysis.

Differences in the proportion of sopE-positive strains between S. Typhimurium phage types and between the ten serotypes were evaluated by exact chi-squared tests.

DNA sequencing of sopE amplicons.

sopE amplicons were purified prior to sequencing using a QIAquick PCR purification kit (Qiagen). Purified amplicons were sequenced in both directions with primers SopE-F and SopE-R using the ABI Prism BigDye terminator kit (Perkin-Elmer Applied Biosystems) by the Advanced Biotechnology Centre (Imperial College School of Medicine, London, UK). Sequence data, obtained in an electronic format, were analysed using Bioedit software version 5.0.9 (Hall, 1999).

	Results and Discussion

TOP Introduction Methods Results and Discussion ACKNOWLEDGEMENTS References

 
It remains unclear why some strains of non-typhoidal S. enterica, particularly MDR strains, are associated more frequently with outbreaks than others. It is possible that some MDR strains may have enhanced pathogenicity in their hosts in relation to their colonization ability or some other virulence attributes. The genetic background of the relatively small number of strains with the capacity both to acquire resistance genes and to colonize food animals has not been fully addressed; similarly, the distribution of virulence factors used by Salmonella to induce enteropathogenesis has not been fully explored.

PCR detection of sopE

In this study, 47/158 S. Typhimurium isolates were sopE-positive. These belonged to definitive phage types (DTs) 29, 44, 49, 204b and 204c only (Table 1), all of which either have been involved in major epidemics [DTs 29, 204b and 204c (Anderson, 1968; Rowe & Threlfall, 1984)] or were the precursors of epidemic strains (DT44 and DT49 were precursors of DT29 and the DT204 complex, respectively). A significantly higher proportion (P < 0.001) of these phage types and a significantly lower proportion of DT104 and DT193 were sopE-positive than would be expected if there was no association between phage type and presence of the sopE gene. Mirold et al. (2001) screened the SARA reference collection and clinical S. Typhimurium isolates from Germany and found all DT49, 204 and 204c isolates and some isolates of DTs 68 and 175 sopE-positive. In contrast, we identified two DT49 and a single DT204c strain that were sopE-negative, possibly representing different clones not prevalent in Germany. MDR S. Typhimurium DT104 strains were sopE-negative and yet are one of the most successful clones; therefore, other factors, such as chromosomal integration of the resistance genes within integron structures, may also be important (Threlfall et al., 1994). Lysogenization with SopE can lead to changes in susceptibility to the phages used in the Anderson typing scheme (Anderson, 1964; Anderson et al., 1977), resulting in phage-type conversion (Rabsch et al., 2002), which may explain in part why sopE is limited to certain phage types.

Presence of sopE varied significantly among the other nine serotypes (P < 0.001) (Table 2). In contrast to serotype Typhimurium, all S. Enteritidis were sopE-positive; therefore, the presence of sopE could not account for differences in prevalence between the phage types tested (Table 2). However, Prager et al. (2000) identified sopE in all isolates of some S. Enteritidis phage types (PT1, 4, 6, 8, 11 and 13) but in only a few isolates of others (PT14b and 21). PT1, 4, 6, 8 and 13 are epidemiologically important in the UK, while PT14b was relatively uncommon until recently (Communicable Disease Surveillance Centre, 2002). Carriage of sopE in these and other phage types of S. Enteritidis may contribute to their epidemiological success.

Of the S. Virchow isolates, only antimicrobial-sensitive PT30 and 31 isolates were negative (Table 2). All S. Hadar PT2, 5, 22 and 58 strains, but only antimicrobial-resistant strains of PT10, 11 and 18, were positive (Table 2). sopE was more prevalent in MDR S. Newport than in antimicrobial-sensitive strains (Table 2); however, it was not found in bovine and human MDR S. Newport isolates tested (K. L. Hopkins, unpublished observation) that are currently epidemic in the United States (Rankin et al., 2002). Only two S. Java strains and one strain of S. Infantis were sopE-positive, while all strains of S. Braenderup, S. Agona and S. Stanley were negative. In contrast to previous studies (Hardt et al., 1998; Prager et al., 2000), where only a small number of strains of each serotype were analysed, we found sopE in a large proportion of S. Virchow, S. Hadar and S. Newport and in one strain of S. Infantis. Further strains need to be tested to make any clear association between particular sero-phage types and carriage of sopE, as was noted for certain S. Typhimurium phage types.

An invA-specific PCR, previously used as a target for rapid identification of Salmonella (Chiu & Ou, 1996), was used to test the quality of sopE-negative DNA templates. All sopE-negative S. Typhimurium, S. Virchow, S. Hadar, S. Newport, S. Java, S. Infantis, S. Braenderup, S. Agona and S. Stanley produced an invA amplicon of the expected size (data not shown); therefore, a negative sopE PCR result did not reflect poor DNA template preparation.

Sequence analysis of sopE amplicons

Melting-curve analysis of the sopE amplicons is capable of revealing differences in the nucleotide sequence; here, two peaks were produced, differing in melting temperature by about 1 °C (Fig. 1). All sopE amplicons from S. Typhimurium strains had a melting point of 86.9 ± 0.3 °C, identical to the positive control SARA4 in each amplification run. Amplicons from all S. Newport and S. Java strains and 8/9 S. Virchow strains had a melting point of 87.2 °C, similar to that obtained for the S. Typhimurium strains and SARA4 (Fig. 1). However, one S. Virchow (V2) strain produced an amplification product with a higher melting point of 88.4 °C. Amplicons from S. Hadar and S. Enteritidis strains had a melting point of 88.3 ± 0.2 °C. S. Typhimurium LT2 did not produce a product peak after melting-curve analysis. RFLP patterns of sopE have been shown to correlate with serotype, with all isolates of the same serotype producing identical RFLP patterns, but patterns can also be conserved between different serotypes (Prager et al., 2000).

View larger version (18K): [in this window] [in a new window]   Fig. 1. LightCycler assay for detection of sopE in S. enterica. Melting-curve analysis of the sopE PCR product shows two peaks are produced: melting curve A (with a melting peak of 86.9 ± 0.3 °C) corresponds to PCR products amplified from S. Typhimurium, S. Newport, S. Java and S. Virchow, while melting curve B (with a melting peak of 88.3 ± 0.2 °C) corresponds to PCR products amplified from S. Enteritidis and S. Hadar.

sopE amplicons from S. Enteritidis, S. Java, S. Infantis, S. Newport, S. Hadar and S. Virchow were sequenced. The variation responsible for the difference in melting points was distributed throughout the length of the amplicon, resulting in 31 amino acid substitutions (an alignment is available as supplementary material in JMM Online). The S. Virchow V2 sopE amplicon, which had a higher melting point than the other S. Virchow strains, had a sequence identical to the sopE of S. Enteritidis. This may result from infection of an S. Virchow strain with the SopE-encoding phage from an S. Enteritidis strain. Mirold et al. (1999) were able to induce SopE from epidemic strains of S. Typhimurium and thus demonstrate that SopE is capable of infecting a range of S. Typhimurium strains, leading to the potential for horizontal transfer of sopE to other strains.

It has been suggested that acquisition of sopE by lysogeny with SopE could increase strain fitness and therefore may be an important factor in the emergence of epidemic strains (Ehrbar et al., 2002). The available data indicate that SopE may play a role in invasion of host cells. Lysogenic conversion of S. Typhimurium ATCC 14028 with SopE resulted in a 2.5-fold increase in invasive ability compared with the wild-type (Ehrbar et al., 2002), which may be sufficient to result in a small increase in transmissibility of sopE-positive strains and increase prevalence over time. Prager et al. (2003) found that all isolates of S. Paratyphi B associated with systemic infection carried sopE, compared with only a small number of S. Paratyphi B associated with enteric infection. These strains are distinguished phenotypically only by their ability to utilize D-tartrate, but show markedly different levels of virulence in man. However, studies analysing the role of SopE in Salmonella pathogenesis showed that inactivation of sopE had no effect on the LD₅₀ or mean time of death of orally inoculated BALB/c mice and did not affect the macrophage cytotoxicity of S. Typhimurium (Hardt et al., 1998). However, Bossi (2003) suggested that, in addition to the improved interaction between Salmonella and the host cell due to carriage of the sopE-encoding phage, lysogeny by this phage may kill sensitive bacteria and result in a population of sopE-positive strains. This may be an additional advantage over sopE-negative strains and contribute to the epidemic success of sopE-positive strains.

The data presented here suggest that, with the exception of S. Enteritidis, there is no clear association of sopE with a particular serotype. However, sopE appears to be associated with particular phage types within a given serotype. Sequence polymorphisms between sopE genes belonging to different serotypes may enable ‘fine-tuning’ of host cell signalling and perhaps result in differences in invasion potential. Detection of sopE may provide a method for the identification of Salmonella strains with the potential for epidemic spread and lead to introduction of targeted intervention strategies to combat the appearance and spread of MDR strains through the food chain.

	ACKNOWLEDGEMENTS

TOP Introduction Methods Results and Discussion ACKNOWLEDGEMENTS References

 
This study was funded by a food safety research grant from McDonalds Restaurants (UK) Ltd. We thank the Health Protection Agency (HPA) Salmonella Reference Unit for serotyping, phage typing and information relating to the strains, Dr R. H. Davies (Veterinary Laboratories Agency – Weybridge) for provision of some of the S. Typhimurium animal isolates, Dr K. Grant (Specialist and Reference Microbiology Division, HPA) for help with optimization of the LightCycler assay and Neville Verlander (Statistics Unit, HPA Communicable Disease Surveillance Centre) for help with statistical analysis.

	References

TOP Introduction Methods Results and Discussion ACKNOWLEDGEMENTS References