| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
INRA, UR454 Unité de Microbiologie, 63122 St-Genès-Champanelle, France
Correspondence
Christine Martin
cmartin{at}clermont.inra.fr
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Two supplementary figures are available with the online version of this paper.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Many non-O157 strains responsible for HUS do not possess the locus of enterocyte effacement (LEE) (Banatvala et al., 2001), a pathogenicity island encoding a type III secretion system recognized as a major virulence factor (Girardeau et al., 2005; Garmendia et al., 2005). STEC strains were classified into five seropathotypes (A to E) by Karmali et al. (2003) according to their incidence and association with HUS cases and outbreaks. Seropathotypes A and B include strains belonging to serotypes associated with HUS cases and outbreaks and containing the LEE. Seropathotype C includes LEE-negative isolates associated with sporadic cases of HUS but not with outbreaks. The virulence mechanisms of this group of strains are not well understood, and to date it is not possible to distinguish among these strains those which are a high risk to human health. Seropathotype D includes isolates rarely found in humans and associated with less severe disease (diarrhoea and HC). Seropathotype E includes serotypes that have never been found in humans.
The major characteristic of STEC linked to haemolytic symptoms is the production of Shiga toxins (Stx1, Stx2), which inhibit protein synthesis (Karmali et al., 1985; Paton & Paton, 1998). Members of the Stx family are AB holotoxins comprising one A subunit, which is the active component of the toxin, covalently bound to five identical B subunits. The B subunits form a pentameric structure required for toxin binding to its receptor, the glycolipid Gb3. Epidemiological studies, together with in vivo and in vitro experiments, have revealed that Stx2 is the most important virulence factor associated with severe human disease. Indeed, Stx2 is 1000 times more cytotoxic than Stx1 towards human renal endothelial cells, and STEC producing Stx2 are more commonly associated with serious diseases than isolates producing Stx1 or Stx1 plus Stx2 (Louise & Obrig, 1995; Boerlin et al., 1999; Paton & Paton, 1998). Several Stx2 variants have been identified on the basis of sequence homology and immunological cross-reactivity (Ito et al., 1990; Schmitt et al., 1991; Paton et al., 1995; Friedrich et al., 2002; Pierard et al., 1998). The most frequent variants identified so far in strains of human and bovine origin are Stx2, Stx2-vha, Stx2-vhb and Stx2c. Stx2 is the toxin produced by the prototype O157 : H7 strains EDL933 and Sakaï. Stx2-vha and Stx2-vhb were first described in an E. coli O91 : H21 strain isolated from a patient with HUS (Ito et al., 1990), and were also named Stx2d1 and Stx2d2, respectively (Teel et al., 2002). Stx2c was found in the E. coli O157 : H– strain E32511, a clinical isolate associated with HUS (Schmitt et al., 1991). Stx2-vha (AF479828-1), Stx2-vhb (AF479829-1) and Stx2c (M59432-1) have 99 %, 97.4 % and 100 % sequence identity in their mature A subunit and 97.1 % identity in their mature B subunit to the corresponding subunit of Stx2 (Y10775). These percentage values correspond to a maximum of three amino acid changes in each subunit. However, it has been suggested that Stx2 sequence variations may affect the capacity of a given Stx2-producing E. coli strain to cause disease (Lindgren et al., 1994; Paton et al., 1995).
The genes encoding Shiga toxins (stx) are generally carried by lambdoid bacteriophages and can be induced by DNA-damaging agents such as mitomycin C (Muhldorfer et al., 1996). As a result of the induction process, expression of stx2 genes, which is under the control of a late phage promoter (Wagner et al., 2001), is activated. Bacterial host cells lyse and release Shiga toxins and free phage particles into the environment. Epidemiological observations and animal models suggest that the severity of the disease is correlated with the amount of Stx produced during infection (Zhang et al., 2000; Dean-Nystrom et al., 2003). However, expression of stx genes and their ability to be induced depending on the stx2 variant or the relative virulence of the STEC strain is not well documented.
Most previous studies investigating the virulence traits of LEE-negative strains were based on the distribution of virulence-associated genes or of genomic islands rather than on their expression and production of the virulence factors. Here, we examined whether seropathotype A and seropathotype C STEC differ in their basal and inducible stx expression in an in vitro model. Furthermore, association of stx2 and stx2-related genes with an inducible prophage was investigated.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The mutants of E. coli O157 : H7 strain EDL933 lacking the stx1, the stx2, or the stx1 and stx2 genes were constructed in our laboratory (Gobert et al., 2007). Bacterial strains were routinely grown in Luria–Bertani (LB) medium. For Stx2 production under mitomycin C induction, overnight cultures of each isolate were subcultured in 7 ml fresh LB medium or in DMEM low glucose (Invitrogen no. 11880) with 4 mM L-glutamine and grown at 37 °C with rotary shaking at 180 r.p.m. The OD600 was measured using 1 cm cuvettes with a Jenway 6300 spectrophotometer. When the OD600 reached 0.2–0.3 (t0), the cultures were divided into two tubes, and mitomycin C (250 ng ml–1) was added to one of the tubes. Then the bacteria were grown at 37 °C with rotary shaking up to 6 h. Bacterial growth was monitored spectrophotometrically after appropriate dilution (two- to fivefold) of the samples in LB or DMEM medium. Three hours after mitomycin C treatment (t3), 6 ml of culture was harvested from each tube and centrifuged for 15 min at 3000 g. The supernatants were filtered through 0.45 µm filters (Sartorius) and used for Stx2 ELISA and detection of phage particles; the pellets were used for RNA extraction. For some strains, levels of stx mRNA were also measured at t1.5 and phage DNA at t6.
|
to discriminate between stx2 (Y10775), stx2-vhb (AF479829-1) and stx2-vha/stx2c variants. For distinction between stx2-vha (AF479828-1) and stx2c (M59432-1) variants, primers were designed in the stx2A gene (forward, 5'-GATGGCGGTCCATTATC-3'; reverse, 5'-CGGTAGAAAGTATTTGTTG-3') that amplify a 529 bp DNA fragment on the genomic DNA of strains carrying stx2-vha or stx2-c. After EcoRV digestion of this fragment, stx2c variants show two DNA fragments of 329 bp and 200 bp whereas stx2-vha DNA remains uncut. The online version of this paper contains a supplementary figure (Fig. S1) showing the nucleotide sequence alignment of the gene encoding the A subunit of stx2, stx2-vha, stx2-vhb and stx2c, with the location of the primers and the EcoRV restriction site, and a photograph of the electrophoresis gel showing the DNA fragments obtained with the PCR-RFLP assay that differentiates stx2c from stx2-vha.
The presence of genes encoding Stx2 activatable by elastase was investigated using PCR with primers SLT-II-vc and CKS2 and restriction analysis of the resulting 890 bp amplicon with PstI as described by Jelacic et al. (2003). Absence of the PstI site can be taken as an indicator of the presence of a mucus-activatable stx2 variant (Bielaszewska et al., 2006; Gobius et al., 2003; Jelacic et al., 2003).
RNA extraction and quantitative real-time PCR (q-PCR).
Following mitomycin C induction for 3 h, total RNA was extracted using the Nucleospin RNA II kit (Macherey-Nagel). RNA concentration was determined by measuring the A260 in a 96-well plate reader (Biotek µQuant) after 50-fold dilution in RNase-free water. One microgram of each RNA sample was reverse transcribed with Superscript II Enzyme (Invitrogen) and 1 µl of random primers (Invitrogen) in a final volume of 20 µl for 50 min at 42 °C. Three q-PCRs were carried out for each sample on diluted cDNA by using the LightCycler apparatus (Roche) with 0.5 µM of each primer, 4 mM MgCl2, 1 µl of LightCycler Faststart DNA Master SYBR green I (Roche) and 2 µl of cDNA (1 ng, 100 pg and 10 pg) in microcapillary tubes in a final volume of 20 µl. Primers 2SF (CACATTTACAGTGAAGGTTGA) and 2R (TTCAGCAAATCCGGAGCCTG) allowed amplification of an stx2 fragment but not of an stx2-vhb fragment. Primers 2bSF (TACATTCACAGTAAAAGTGGC) and 2R allowed amplification of an stx2-vhb fragment but not of an stx2 fragment. stx2-vha and stx2c fragments were amplified using primers 2a (TAAAAGTGGCCGGAAAAGAG) and 2R. stx2 and tufA mRNAs were quantified by noting the fluorescence crossing-point (CP) of the samples on the corresponding standard curve. Results are the mean ratios between the copy number of stx2 mRNA and the copy number of tufA mRNA. Control reactions were performed without reverse transcriptase to confirm that the target detected was RNA. Standard curves for stx quantification were obtained by PCR amplification of genomic DNA from EHEC strains containing only one of the different stx2 variants with the universal stx2 primers 2F-0 (TATATCAGTGCCCGGTGTGA) and 2R-0 (CATTATTAAACTGCACTTCAGC). The PCR products (884 bp) were purified with the Strataprep PCR purification kit (Stratagene) and DNA amounts were quantified by measuring the A260 in a 96-well plate reader (Biotek µQuant). This amount was converted to molecule number as previously described (Fronhoffs et al., 2002). Then PCR products were 10-fold serially diluted from 5x108 to 50 molecules µl–1 and three q-PCRs were carried out in a LightCycler apparatus (Roche) with primers 2SF and 2R, 2bSF and 2R, and 2a and 2R, generating standard curves for stx2, stx2-vhb, and stx2-vha/stx2c respectively. The standard curve for tufA quantification was obtained in the same way using the genomic DNA of E. coli O157 : H7 strain EDL933 and the primers TufAF (CAGGTAGGCGTTCCGTACAT) and TufAR (GTGCAAAAAGGGCATCAAAT). After quantification of the molecule number, a q-PCR was done with primers TufAqF (TGGTTGATGACGAAGAGCTG) and TufAqR (GCTCTGGTTCCGGAATGTAA) to obtain the standard curve.
Phage particle isolation and stx2 DNA quantification.
Phages were purified from non-induced or mitomycin C-induced cultures 3 h or 6 h after mitomycin C treatment as previously described (Fuchs et al., 1999) with slight modifications according to Arthur Donohue-Rolfe, Cummings School of Veterinary Medicine at Tufts University (personal communication). Briefly, 5 ml samples of the filtered culture supernatants were incubated with 10 µg ml–1 RNase A (Amersham) and 40 U ml–1 DNase I (Amersham) for 30 min at 37 °C. Phage particles were pelleted by ultracentrifugation for 16 h at 76 000 g at 4 °C. The pellets were resuspended in 100 µl PBS. Phage suspensions were boiled for 5 min at 95 °C, diluted fivefold in PBS, then stx2 and stx2-related DNA was quantified by q-PCR using the same primers and standards as for mRNA quantification. Absence of bacterial DNA in phage lysates was confirmed by the absence of tufA DNA as assayed by PCR. This method of indirect phage quantification detects phage DNA after its release from phage particles upon lysis of the host strain by use of specific primers. The method is very sensitive and circumvents the problem of extreme instability of plaque-forming capabilities of Stx2 phages, which are lost within 2 h of storage (Fuchs et al., 1999; Muniesa et al., 2004b), and the difficulty of identification of stx2-phage plaques, which are very small and not well visible on agar plates (Gamage et al., 2004; Muniesa et al., 2003).
Measurement of Stx2 concentration by ELISA.
Sandwich ELISAs were performed in 96-well plates (Maxisorp, Nunc) using a monoclonal antibody against Stx2 (STX2-BB12, Toxin Technology) diluted 1 : 500 to coat the plates and a rabbit polyclonal antiserum against Stx2 diluted 1 : 2000 to detect the toxin. A standard curve was obtained with twofold serial dilutions of purified Stx2 (Toxin Technology). Detection was performed with a 1 : 10 000 dilution of horseradish peroxidase–goat IgG anti-rabbit (Pierce) and o-phenylenediamine (OPD, Sigma)-stable peroxidase substrate (Pierce). Absorbance was measured in a 96-well plate reader (Biotek µQuant) at 492 nm.
Statistical analyses.
Data are expressed as means±SEM. To analyse the effect of the mitomycin C treatment on stx2 transcription or Stx2 production, statistical analyses were performed using SAS software (v 8.1). The origin of the strains, the stx2 variant and the serotype were considered to be independent variables. Due to the unequal number of strains in the groups, analysis of variance was performed using the Linear model procedure. Mean multiple comparison tests on log10-transformed datasets were performed. The comparison method used was the Ryan–Einot–Gabriele–Welsh multiple range test. We made pairwise adjustments using the Tukey method. A P value 
0.05 was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Ten O157 : H7 strains (nine from humans and one from cattle) belong to seropathotype A and as such possess the LEE. The remaining strains belong to seropathotype C and do not possess the LEE. Among them, 5 strains from humans and 11 strains from cattle belong to the O113 : H21 serotype, 4 from humans and 5 from cattle are O91 : H21, 1 from humans and 6 from cattle are O91 : H10. These ratios between bovine- and human-derived strains for each serotype reflect the actual occurrence quite well, since O157 : H7 strains are more frequently isolated from patients than from cattle, whereas O91 : H10 strains are rarely found in humans (Caprioli et al., 2005; Pradel et al., 2000; Nataro & Kaper, 1998; Brooks et al., 2005). Among the 42 selected strains, 22 contained only one stx2 variant, and 20 contained two stx2 variants. Only the stx2, stx2-vha, stx2-vhb and stx2c variants were found in these strains. The stx2 variant was mainly associated with human-derived strains, especially the O157 : H7 serotype, thus with strains of seropathotype A. This variant was the most frequently found in strains carrying only one stx2 gene. In contrast, 13 strains of seropathotype C among 18 possessed stx2-vhb, which was associated in 10 of them with stx2-vha, stx2-vhb or stx2c.
It has been shown that the biological activity of the Stx2-vha and Stx2-vhb toxins is activatable by elastase cleavage of the last two amino acids of the enzymically active A subunit in the human or mouse intestinal mucus (Melton-Celsa et al., 2002). Therefore all the isolates were further characterized for the possession of putative mucus-activatable Stx2 variants by PCR-RFLP analysis as described by Jelacic et al. (2003). The genes encoding stx2-vha and stx2-vhb B subunits were all associated with a putative activatable A subunit. In O157 : H7 strains, the B subunits were all associated with a non-activatable A subunit. In non-O157 : H7 strains, the association between A and B subunits was heterogeneous. Among 12 strains possessing an stx2 B gene, 6 had a putative activatable A subunit. Among 8 strains possessing an stx2c variant (indistinguishable stx2-vha/stx2c B gene, A gene with the PCR-RFLP pattern in the 5'-end characteristic of stx2c), 7 showed the PCR-RFLP pattern in the A gene 3'-end characteristic of mucus-activatable toxins (Table 1).
Are strains carrying different stx2 variants equally sensitive to mitomycin C?
A subset of 23 strains representative of each serotype and combinations of stx2 variants found in the 42-strain collection was chosen for a more detailed analysis of stx2 and stx2-related expression (Table 1). Treatment of bacteria with DNA-damaging agents such as mitomycin C results in stx-phage induction and cell lysis. It has been shown that the decrease in the culture optical density is a relevant qualitative measurement of prophage induction and stx-phage production (Muniesa et al., 2003; Tyler et al., 2004). Thus, to determine whether phages carrying different stx2 variants were equally induced by mitomycin C, we first monitored bacterial growth. Mitomycin C was added to the cultures at t0 when the bacterial concentration reached 107 to 3x107 c.f.u. ml–1. On the basis of OD600 three growth patterns were observed (Fig. 1). For the first group, which includes the majority of the strains, mitomycin C slightly slowed the growth, and lysis due to lysogenic induction became apparent around 3 h after addition of mitomycin C to the growth medium (Fig. 1a). An early lysis, beginning around 90 min after mitomycin C treatment, was representative of three strains, namely 87-307, 87-2927 and ED-76 (Fig. 1b). The growth of the third group of strains was insensitive or poorly sensitive to mitomycin C (Fig. 1c), indicating that production of phages was limited in these strains. Thus it appeared that all strains were not equally sensitive to mitomycin C. All except one of the strains harbouring only the stx2 variant belonged to the first group, whereas isolates carrying other variants, alone or in combination, were distributed in the three sensitivity groups.
|
In the absence of mitomycin C, only stx2 and stx2c mRNAs were detected (Fig. 2). Indeed, basal stx2 mRNA levels were significantly higher than variant-stx2 mRNAs (P<0.001), except for stx2c mRNA in two strains, CHvi-I and NV95. It is noteworthy that these strains are the only two O157 : H7 strains expressing stx2c. No stx2c mRNA was detected in the non-O157 : H7 strains carrying this variant. However, the stx2c variant in these strains was not classical as it showed the genotype associated with mucus-activatable activity. Among the seven strains expressing stx2 only, two did not produce significant mRNA levels. These two strains (VTH-13 and CB6775) belong to non-O157 : H7 serotypes, whereas four of the five strains expressing high mRNA levels are O157 : H7 strains. In all six strains containing both stx2 and stx2-vhb, only stx2 mRNA was detected.
|
). Globally, induced stx2 mRNA levels were higher than the others (P<0.05), except for the two O157 : H7 strains, CHVi-I and NV95, that express stx2c and the two strains that express stx2-vha and stx2-vhb. Transcription of stx2 was stimulated 25- to 100-fold in all strains expressing stx2 only, but when associated with stx2-vhb, expression of stx2 was inducible in only two of six strains. In the four remaining strains stx2 expression was as high in non-induced as in induced conditions. Three of them (CL-3, CL-15 and NV254) carry a non-classical stx2 variant showing the RFLP pattern associated with mucus-activatable variants. Expression of some stx2 variants was not induced by mitomycin C: stx2-vhb mRNA was not produced or was produced at very low levels in all strains expressing stx2-vhb, either alone or in combination with stx2. However, induction of the lytic cycle was earlier for three of these strains, 87-307, 87-2927 and ED-76 (Fig. 1b). To measure stx2-vhb mRNA levels in these strains in a large population of cells, mRNAs were isolated at the beginning of cell lysis, 90 min after mitomycin C treatment. Again, stx2-vhb mRNAs were undetectable. stx2-vha and stx2c mRNA increased 10- to 30-fold after lysogenic induction, reaching high levels in O157 : H7 strains only. An increase in stx2-related mRNA levels was also observed in strains expressing stx2-vhb in combination with stx2-vha or stx2c. However, because our attempts to design a primer pair able to differentiate between these mRNAs were unsuccessful, we failed to determine whether expression of both genes or of only one was induced. It is noteworthy that the strains carrying the non-classical stx2c variant expressed lower levels of stx2-related mRNA than the strains carrying the classical stx2c variant. Analysis of the growth curves and stx mRNA levels showed that mitomycin C-induced cell lysis was associated with high expression of at least one of the stx2 variants carried by the strain. Conversely, when treatment with mitomycin C did not result in cell lysis, the stx2 variants were not or were very poorly expressed. These observations suggest that the stx2 variants are expressed only when they are carried by an inducible prophage.
Only the stx2 variants which are expressed are carried by released phage particles
To investigate whether the stx2 variants are associated with phage DNA, we quantified stx-phage DNA in most culture supernatants by q-PCR using the same stx2 and stx2-related specific primers as used for mRNA analysis. Phage particles were harvested from strain supernatants 3 h (Fig. 2) and 6 h (data not shown) after addition of mitomycin C. stx-phage DNA was not detected or was detected at low levels in cultures in which no lysis occurred, at 3 h as well as 6 h after addition of mitomycin C. At t3 without treatment, significant amounts of stx-phage DNA were only detected in supernatants of strains expressing the corresponding stx mRNA in significant amounts. Higher amounts of stx-phage DNA were detected under mitomycin C treatment except for strains in which stx2 transcription was not inducible (CL-15, CL-3 and NV254). Under mitomycin C induction, stx2-vhb-phage DNA was detected in low amounts only in the supernatant of the three strains for which stx2-vhb mRNA was detected (NV308, DEC16A and NV254) and, surprisingly, in the supernatant of strain 87-307. High stx mRNA levels were never observed in the absence of phage particles in the supernatant. The relationships observed between growth curve patterns, stx mRNA levels and stx-phage release indicate that the stx variants carried on prophages able to produce phage particles in the medium were the only ones expressed, and that the promoter of the stx2-vhb variant, when it is not carried on such prophages (for example in CL-3, CL-15 and NV200), is inactive in the culture conditions used. In those strains in which expression of stx2 was high but not sensitive to mitomycin C (CL-3, CL-15 and NV254), stx2-phage DNA was detected at similar high levels with or without mitomycin C treatment. Thus it appears that the stx2 prophages in these strains had an unusual level of spontaneous induction. At t6, the amounts of stx-phage DNA were two- to fivefold higher than at t3, but the relative amounts of each variant remained similar (data not shown).
To investigate whether stx2 expression and stx2-phage production are affected by nutritional cues, we measured stx2 and stx2-related mRNA and phage DNA when STEC strains were grown under nutrient-limiting conditions, i.e. in DMEM medium containing a low glucose concentration (1 g l–1). The data obtained in DMEM are presented as a supplementary figure with the online version of this paper (Fig. S2). They were not significantly different from data in LB. As in LB medium, we found that without induction, only the stx2 and stx2c variants were expressed. Under mitomycin C treatment, the same levels of mRNA induction as in LB medium were observed for each variant. The stx2-vhb genes not expressed in LB remained unexpressed in DMEM. As in LB medium, stx-phage production was well correlated with stx mRNA.
Shiga toxin release
To confirm that Shiga toxin was synthesized by bacteria expressing the stx2 and stx2-related genes, amounts of Stx2 and Stx2-related variants released in culture supernatants were measured using ELISA (Fig. 3). As Stx2 variants are not immunologically distinct, the assay did not allow differentiation between Stx2, Stx2-vha, Stx2-vhb and Stx2c in the supernatant of strains producing two toxins. Thus differential production of each Stx2 variant could not be assessed by this method.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). However, expression of stx2 variants in STEC differing in their virulence traits is not well documented. In this study, we determined the distribution of stx2 variants among 42 Stx2-producing STEC obtained from patients or from healthy cattle. Ten strains belong to seropathotype A, containing O157 : H7 LEE-positive strains, which are the most frequently associated with severe outbreaks in the world (Karmali et al., 2003). The remaining strains belong to seropathotype C, which includes mainly non-O157 : H7 LEE-negative strains frequently associated with sporadic cases of HUS (Karmali et al., 2003). Then a subset of 24 strains representative of the initial group with regard to the origin, serotype, and stx2 variants was analysed for the induction by mitomycin C of stx2 variant expression, Stx2 release and stx-phage particle production.
Heterogeneity of the stx2 and stx2-related genotypes
Molecular typing of the stx2 and stx2-related variants using previously described or new PCR-RFLP methods revealed mosaic sequences. First, a gene encoding an Stx2 B-subunit can be associated with an A-subunit gene that possesses the PstI site indicative of mucus-activatable toxins. The same observation was made in a recent study on STEC strains isolated in Germany from food samples (Beutin et al., 2007). Second, A-subunit genes primarily typed as stx2c based on PCR-RFLP in their 5' end showed the PCR-RFLP type characteristic of mucus-activatable toxins in their 3' end. However this heterogeneity was not observed in O157 : H7 strains, which all showed the non-activatable stx2 and stx2c genotype, nor for the stx2-vha and stx2-vhb genes, which all showed the activatable genotype.
Heterogeneous expression of stx2 variants
Several authors have shown that Stx2-encoding prophages are highly variable (Johansen et al., 2001) and present a high degree of heterogeneity in terms of induction levels (Muniesa et al., 2004a; Wagner et al., 1999). In agreement with these reports, we found that expression of stx2 and stx2-related genes is heterogeneous in basal and in induced conditions, depending on the strain and on the stx2 variant. q-PCR analysis of stx2-related genes showed that stx2 mRNA levels were higher than stx2-vha or stx2-vhb mRNA levels. However, a more detailed analysis showed a heterogeneous expression of a given stx2 variant which correlated with a heterogeneous release of stx-phage particles depending on the strain. This could be due to different regulatory properties of the phages harbouring the same stx variant, or to bacterial host factors such as repair systems, membrane constitution, etc. For example, among nine strains harbouring stx2-vhb alone or in combination with stx2, stx2-vhb mRNA and phage particles were detected in only three, although at low levels. In Stx2-producing strains, large differences appeared in stx2-mRNA levels, in basal as well as in induced conditions. stx2 genes highly expressed without mitomycin C treatment are probably carried by a prophage with a high level of spontaneous induction, leading to high release of phage particles and Stx independently of the SOS system. Previous studies indicate that the 933W prophage, encoding Stx2 in EDL933, induces more readily than lambdoid prophages that do not encode Stx (Livny & Friedman, 2004). Here we broaden this observation to a number of other strains of different serotypes, showing that most of the prophages expressing the stx2 variant show a high level of spontaneous induction whereas the stx2-vha, stx2-vhb and stx2c variants associated with the mucus-activatable genotype are not or are poorly induced without mitomycin C treatment.
The stx2-vhb variant is probably not associated with an inducible prophage
Upon mitomycin C treatment, stx2, stx2-vha and stx2c transcription was induced, but stx2-vhb transcription was induced in only two strains out of nine. This differential expression was also seen in strains expressing both stx2 and stx2-vhb, in which high stx2 and low stx2-vhb mRNA levels were measured, in induced as well as in non-induced conditions. Thus, our results indicate that in most cases expression of stx2-vhb is not induced upon treatment by mitomycin C, and suggest that in the corresponding strains this gene is not phage-borne, or is carried by a defective prophage. This conclusion is further supported by the absence of stx2-vhb-phages in strain supernatants. Alternatively, in strains harbouring stx2-vhb together with stx2, stx2-vhb-prophages could be induced later than stx2-prophages. Thus induction of the lytic cycle of the stx2-phage could lyse the cells before high stx2-vhb mRNA levels have been produced. This could explain why low amounts of stx2-vhb-phage DNA were detected in the supernatant of strain 87-307 under mitomycin C induction. However, other studies showed that some stx2-related genes were not associated with an inducible prophage (Teel et al., 2002; Zhang et al., 2005), and that in strains harbouring two stx2-prophages only one was inducible (Teel et al., 2002; Muniesa et al., 2003). In particular, using two B2F1 isogenic mutants, producing either Stx2-vha only or Stx2-vhb only, Teel et al. (2002) have provided evidence that stx2-vha expression is bacteriophage associated but stx2-vhb is not. In a study including 168 STEC strains, Muniesa et al. (2004b) found that 5 strains possessed two stx2 copies. In each case, only one copy was found to be associated with an inducible prophage. A constitutive, poorly active promoter controlling stx2 expression in O157 strains has been described (Sung et al., 1990; Plunkett et al., 1999). The very low levels of stx2-vhb mRNA detected in induced and non-induced conditions could indicate that this gene was expressed from such a poorly active specific promoter in the conditions tested.
In summary, our results strongly suggest that stx2 and stx2-related genes are expressed only when they are carried by a prophage subject to a high spontaneous induction or/and to SOS-mediated lysogenic induction. In particular, the stx2-vhb genes are not or are very poorly expressed, in basal as well as in induced conditions, and in most cases are not found on the genome of phage particles in strain supernatants. Induction levels and amounts of stx mRNAs are significantly higher in O157 : H7 strains than in strains belonging to the other serotypes (P<0.01). Furthermore, stx mRNA levels were higher in strains belonging to seropathotype A and in the strains of seropathotype C that express the stx2 variant when compared to other strains of seropathotype C.
Stx2 release
We found variable levels of Shiga toxins in supernatants of some STEC strains in LB medium without mitomycin C. This is probably due to spontaneous phage induction, since we detected toxins in the supernatants only when phage particles were also detected. This basal amount of released toxin varied from strain to strain, and was significantly higher in strains isolated from humans than in strains isolated from cattle. There were also marked differences in the effect of mitomycin C on toxin production by different strains, but, in contrast to the findings by Ritchie et al. (2003), we found no significant difference between human- and bovine-derived strains, in terms of amounts of released toxin or of induction levels. This discrepancy could be due to the choice of strains included in the study. All but one of the human-derived strains belonged to the highly virulent seropathotype A in the study reported by Ritchie et al. (2003), whereas half of our human-derived strains belong to the low-virulent seropathotype C.
For most of the strains, the amount of toxin released in the supernatant correlated well with the measured levels of stx mRNA, in basal and in induced conditions. However, this was not the case for some particular strains. Indeed, strains CL-3, CL-15 and NV254 each produced similar high levels of stx2 mRNA and stx2-phage particles independently of the presence of mitomycin C. However, low amounts of toxin were produced without mitomycin C whereas high Stx2 amounts were produced under mitomycin C treatment; strains NV308, NV127 and NV199 produced moderate levels of stx mRNA and stx-phage DNA under mitomycin C treatment, but released high amounts of toxin. Post-transcriptional regulation leading to a more efficient translation of stx mRNA in the presence of mitomycin C could explain these observations. In contrast, strain NV200 produced high levels of stx2 mRNA, but did not release Stx2 into the medium, suggesting that stx2 mRNA was not translated. Although further investigation is required to determine the molecular mechanisms involved, these observations highlight the functional diversity of stx phages.
Conclusion
Our results indicate that the stx2 variant is mainly associated with the most pathogenic human-derived strains belonging to seropathotype A (O157 : H7 strains). Furthermore, the data show a relationship between the seropathotype and the expression level of the stx2 variant, and the amount of stx-phages and toxin released, in basal as well as in induced conditions. In seropathotype C, a subset of LEE-negative strains showed the same characteristics as the O157 : H7 strains regarding stx expression and induction and thus could be a greater risk to human health than the other strains belonging to this seropathotype.
|
ACKNOWLEDGEMENTS |
|---|
Edited by: N. J. High
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bertin, Y., Boukhors, K., Pradel, N., Livrelli, V. & Martin, C. (2001). Stx2 subtyping of Shiga toxin-producing Escherichia coli isolated from cattle in France: detection of a new Stx2 subtype and correlation with additional virulence factors. J Clin Microbiol 39, 3060–3065.
Beutin, L., Miko, A., Krause, G., Pries, K., Haby, S., Steege, K. & Albrecht, N. (2007). Identification of human-pathogenic strains of Shiga toxin-producing Escherichia coli from food by a combination of serotyping and molecular typing of Shiga toxin genes. Appl Environ Microbiol 73, 4769–4775.
Bielaszewska, M., Friedrich, A. W., Aldick, T., Schurk-Bulgrin, R. & Karch, H. (2006). Shiga toxin activatable by intestinal mucus in Escherichia coli isolated from humans: predictor for a severe clinical outcome. Clin Infect Dis 43, 1160–1167.[CrossRef][Medline]
Boerlin, P., McEwen, S. A., Boerlin-Petzold, F., Wilson, J. B., Johnson, R. P. & Gyles, C. L. (1999). Associations between virulence factors of Shiga toxin-producing Escherichia coli and disease in humans. J Clin Microbiol 37, 497–503.
Brooks, J. T., Sowers, E. G., Wells, J. G., Greene, K. D., Griffin, P. M., Hoekstra, R. M. & Strockbine, N. A. (2005). Non-O157 Shiga toxin-producing Escherichia coli infections in the United States, 1983–2002. J Infect Dis 192, 1422–1429.[CrossRef][Medline]
Caprioli, A., Morabito, S., Brugere, H. & Oswald, E. (2005). Enterohaemorrhagic Escherichia coli: emerging issues on virulence and modes of transmission. Vet Res 36, 289–311.[CrossRef][Medline]
Dean-Nystrom, E. A., Melton-Celsa, A. R., Pohlenz, J. F., Moon, H. W. & O'Brien, A. D. (2003). Comparative pathogenicity of Escherichia coli O157 and intimin-negative non-O157 Shiga toxin-producing E. coli strains in neonatal pigs. Infect Immun 71, 6526–6533.
Friedrich, A. W., Bielaszewska, M., Zhang, W. L., Pulz, M., Kuczius, T., Ammon, A. & Karch, H. (2002). Escherichia coli harboring Shiga toxin 2 gene variants: frequency and association with clinical symptoms. J Infect Dis 185, 74–84.[CrossRef][Medline]
Fronhoffs, S., Totzke, G., Stier, S., Wernert, N., Rothe, M., Bruning, T., Koch, B., Sachinidis, A., Vetter, H. & other authors (2002). A method for the rapid construction of cRNA standard curves in quantitative real-time reverse transcription polymerase chain reaction. Mol Cell Probes 16, 99–110.[CrossRef][Medline]
Fuchs, S., Muhldorfer, I., Donohue-Rolfe, A., Kerenyi, M., Emody, L., Alexiev, R., Nenkov, P. & Hacker, J. (1999). Influence of RecA on in vivo virulence and Shiga toxin 2 production in Escherichia coli pathogens. Microb Pathog 27, 13–23.[CrossRef][Medline]
Gamage, S. D., Patton, A. K., Hanson, J. F. & Weiss, A. A. (2004). Diversity and host range of Shiga toxin-encoding phage. Infect Immun 72, 7131–7139.
Garmendia, J., Frankel, G. & Crepin, V. F. (2005). Enteropathogenic and enterohemorrhagic Escherichia coli infections: translocation, translocation, translocation. Infect Immun 73, 2573–2585.
Girardeau, J. P., Dalmasso, A., Bertin, Y., Ducrot, C., Bord, S., Livrelli, V., Vernozy-Rozand, C. & Martin, C. (2005). Association of virulence genotype with phylogenetic background in comparison to different seropathotypes of Shiga toxin-producing Escherichia coli isolates. J Clin Microbiol 43, 6098–6107.
Gobert, A. P., Vareille, M., Glasser, A.-L., Hindré, T., de Sablet, T. & Martin, C. (2007). Shiga toxin produced by enterohemorrhagic Escherichia coli inhibits PI3K/NF-
B signaling pathway in globotriaosylceramide-3-negative human intestinal epithelial cells. J Immunol 178, 8168–8174.
Gobius, K. S., Higgs, G. M. & Desmarchelier, P. M. (2003). Presence of activatable Shiga toxin genotype (stx(2d)) in Shiga toxigenic Escherichia coli from livestock sources. J Clin Microbiol 41, 3777–3783.
Ito, H., Terai, A., Kurazono, H., Takeda, Y. & Nishibuchi, M. (1990). Cloning and nucleotide sequencing of Vero toxin 2 variant genes from Escherichia coli O91 : H21 isolated from a patient with the hemolytic uremic syndrome. Microb Pathog 8, 47–60.[CrossRef][Medline]
Jelacic, J. K., Damrow, T., Chen, G. S., Jelacic, S., Bielaszewska, M., Ciol, M., Carvalho, H. M., Melton-Celsa, A. R., O'Brien, A. D. & other authors (2003). Shiga toxin-producing Escherichia coli in Montana: bacterial genotypes and clinical profiles. J Infect Dis 188, 719–729.[CrossRef][Medline]
Johansen, B. K., Wasteson, Y., Granum, P. E. & Brynestad, S. (2001). Mosaic structure of Shiga-toxin-2-encoding phages isolated from Escherichia coli O157 : H7 indicates frequent gene exchange between lambdoid phage genomes. Microbiology 147, 1929–1936.
Karmali, M. A., Petric, M., Lim, C., Fleming, P. C., Arbus, G. S. & Lior, H. (1985). The association between idiopathic hemolytic uremic syndrome and infection by verotoxin-producing Escherichia coli. J Infect Dis 151, 775–782.[Medline]
Karmali, M. A., Mascarenhas, M., Shen, S., Ziebell, K., Johnson, S., Reid-Smith, R., Isaac-Renton, J., Clark, C., Rahn, K. & other authors (2003). Association of genomic O island 122 of Escherichia coli EDL 933 with verocytotoxin-producing Escherichia coli seropathotypes that are linked to epidemic and/or serious disease. J Clin Microbiol 41, 4930–4940.
Lindgren, S. W., Samuel, J. E., Schmitt, C. K. & O'Brien, A. D. (1994). The specific activities of Shiga-like toxin type II (SLT-II) and SLT-II-related toxins of enterohemorrhagic Escherichia coli differ when measured by Vero cell cytotoxicity but not by mouse lethality. Infect Immun 62, 623–631.
Livny, J. & Friedman, D. I. (2004). Characterizing spontaneous induction of Stx encoding phages using a selectable reporter system. Mol Microbiol 51, 1691–1704.[CrossRef][Medline]
Louise, C. B. & Obrig, T. G. (1995). Specific interaction of Escherichia coli O157 : H7-derived Shiga-like toxin II with human renal endothelial cells. J Infect Dis 172, 1397–1401.[Medline]
Melton-Celsa, A. R., Kokai-Kun, J. F. & O'Brien, A. D. (2002). Activation of Shiga toxin type 2d (Stx2d) by elastase involves cleavage of the C-terminal two amino acids of the A(2) peptide in the context of the appropriate B pentamer. Mol Microbiol 43, 207–215.[CrossRef][Medline]
Muhldorfer, I., Hacker, J., Keusch, G. T., Acheson, D. W., Tschape, H., Kane, A. V., Ritter, A., Olschlager, T. & Donohue-Rolfe, A. (1996). Regulation of the Shiga-like toxin II operon in Escherichia coli. Infect Immun 64, 495–502.[Abstract]
Muniesa, M., de Simon, M., Prats, G., Ferrer, D., Panella, H. & Jofre, J. (2003). Shiga toxin 2-converting bacteriophages associated with clonal variability in Escherichia coli O157 : H7 strains of human origin isolated from a single outbreak. Infect Immun 71, 4554–4562.
Muniesa, M., Blanco, J. E., De Simon, M., Serra-Moreno, R., Blanch, A. R. & Jofre, J. (2004a). Diversity of stx2 converting bacteriophages induced from Shiga-toxin-producing Escherichia coli strains isolated from cattle. Microbiology 150, 2959–2971.
Muniesa, M., Serra-Moreno, R. & Jofre, J. (2004b). Free Shiga toxin bacteriophages isolated from sewage showed diversity although the stx genes appeared conserved. Environ Microbiol 6, 716–725.[CrossRef][Medline]
Nataro, J. P. & Kaper, J. B. (1998). Diarrheagenic Escherichia coli. Clin Microbiol Rev 11, 142–201.
O'Brien, A. O., Lively, T. A., Chen, M. E., Rothman, S. W. & Formal, S. B. (1983). Escherichia coli O157 : H7 strains associated with haemorrhagic colitis in the United States produce a Shigella dysenteriae 1 (SHIGA) like cytotoxin. Lancet 1, 702[Medline]
Paton, J. C. & Paton, A. W. (1998). Pathogenesis and diagnosis of Shiga toxin-producing Escherichia coli infections. Clin Microbiol Rev 11, 450–479.
Paton, A. W., Bourne, A. J., Manning, P. A. & Paton, J. C. (1995). Comparative toxicity and virulence of Escherichia coli clones expressing variant and chimeric Shiga-like toxin type II operons. Infect Immun 63, 2450–2458.[Abstract]
Pierard, D., Muyldermans, G., Moriau, L., Stevens, D. & Lauwers, S. (1998). Identification of new verocytotoxin type 2 variant B-subunit genes in human and animal Escherichia coli isolates. J Clin Microbiol 36, 3317–3322.
Plunkett, G., III, Rose, D. J., Durfee, T. J. & Blattner, F. R. (1999). Sequence of Shiga toxin 2 phage 933W from Escherichia coli O157 : H7: Shiga toxin as a phage late-gene product. J Bacteriol 181, 1767–1778.
Pradel, N., Livrelli, V., De Champs, C., Palcoux, J. B., Reynaud, A., Scheutz, F., Sirot, J., Joly, B. & Forestier, C. (2000). Prevalence and characterization of Shiga toxin-producing Escherichia coli isolated from cattle, food, and children during a one-year prospective study in France. J Clin Microbiol 38, 1023–1031.
Ritchie, J. M., Wagner, P. L., Acheson, D. W. & Waldor, M. K. (2003). Comparison of Shiga toxin production by hemolytic-uremic syndrome-associated and bovine-associated Shiga toxin-producing Escherichia coli isolates. Appl Environ Microbiol 69, 1059–1066.
Schmitt, C. K., McKee, M. L. & O'Brien, A. D. (1991). Two copies of Shiga-like toxin II-related genes common in enterohemorrhagic Escherichia coli strains are responsible for the antigenic heterogeneity of the O157 : H– strain E32511. Infect Immun 59, 1065–1073.
Sung, L. M., Jackson, M. P., O'Brien, A. D. & Holmes, R. K. (1990). Transcription of the Shiga-like toxin type II and Shiga-like toxin type II variant operons of Escherichia coli. J Bacteriol 172, 6386–6395.
Teel, L. D., Melton-Celsa, A. R., Schmitt, C. K. & O'Brien, A. D. (2002). One of two copies of the gene for the activatable Shiga toxin type 2d in Escherichia coli O91 : H21 strain B2F1 is associated with an inducible bacteriophage. Infect Immun 70, 4282–4291.
Tyler, J. S., Mills, M. J. & Friedman, D. I. (2004). The operator and early promoter region of the Shiga toxin type 2-encoding bacteriophage 933W and control of toxin expression. J Bacteriol 186, 7670–7679.
Wagner, P. L., Acheson, D. W. & Waldor, M. K. (1999). Isogenic lysogens of diverse Shiga toxin 2-encoding bacteriophages produce markedly different amounts of Shiga toxin. Infect Immun 67, 6710–6714.
Wagner, P. L., Neely, M. N., Zhang, X., Acheson, D. W., Waldor, M. K. & Friedman, D. I. (2001). Role for a phage promoter in Shiga toxin 2 expression from a pathogenic Escherichia coli strain. J Bacteriol 183, 2081–2085.
Zhang, X., McDaniel, A. D., Wolf, L. E., Keusch, G. T., Waldor, M. K. & Acheson, D. W. (2000). Quinolone antibiotics induce Shiga toxin-encoding bacteriophages, toxin production, and death in mice. J Infect Dis 181, 664–670.[CrossRef][Medline]
Zhang, W., Bielaszewska, M., Friedrich, A. W., Kuczius, T. & Karch, H. (2005). Transcriptional analysis of genes encoding Shiga toxin 2 and its variants in Escherichia coli. Appl Environ Microbiol 71, 558–561.
Received 11 May 2007;
revised 3 September 2007;
accepted 3 October 2007.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
, Growth in the absence of mitomycin C; 
, growth in the presence of mitomycin C. (a) Growth curve of NV237 (stx2), representative of E. coli Sakaï (stx2), EDL933 (stx2), 86-24 (stx2), 85-08 (stx2), VTH-13 (stx2), NV237 (stx2), NV308 (stx2-vhb), CHVi-I (stx2c), NV95 (stx2c), CL-3, CL-15 (stx2-stx2-vhb), NV254 (stx2-stx2-vhb), NV199 (stx2c-stx2-vhb) and B2F1 (stx2-vha-stx2-vhb). (b) Growth curve of E. coli 87-307 (stx2-stx2-vhb), representative of E-D76 (stx2-stx2-vhb) and 87-2927 (stx2-vha-stx2-vhb). (c) Growth curve of E. coli NV166 (stx2c), representative of CB6775 (stx2), NV148 (stx2-vha), DEC16A (stx2-vhb), E-D226 (stx2-vhb), NV200 (stx2-stx2-vhb) and NV127 (stx2c-stx2-vhb).
) or presence (
) of mitomycin C for 3 h. The seropathotype of the strains is indicated at the top. stx2 variants indicated under the graph are the variants detected in the q-PCR experiments. The dashed line indicates values of the negative control. *Strains harbouring stx2+stx2-vhb were tested separately for stx2 and stx2-vhb.