Introduction

Verotoxigenic Escherichia coli (VTEC or its synonym STEC) is a zoonotic emergent pathogen that represents an important cause of diseases in humans like hemorrhagic colitis (HC) and hemolytic uremic syndrome (HUS) [19]. The VTEC family is very diverse, being E. coli O157:H7 the most common serotype associated with sporadic cases and large outbreaks of diseases in many countries [1, 44]. However, there is growing concern about the risk to human health associated with non-O157 VTEC serotypes. About 250 non-O157 VTEC serotypes have been reported and more than 100 of them have been associated with human illness [http://www.microbionet.com.au/frames/feature/vtec/brief01.html].

The Verotoxin (Vtx) and different variants of this bacteriophage-encoded toxin are considered the cardinal virulence trait of VTEC [17]. However, additional virulence determinants also contribute to the pathogenesis of VTEC infections [3, 18]. The genes responsible for an intimate adherence of the bacteria and the attaching and effacing lesions are located on a chromosomal pathogenicity island termed the locus of enterocyte effacement (LEE) [23]. The majority of VTEC serogroups related to severe disease, including the most prevalent O157:H7, possess this locus. However, some VTEC without LEE are associated with sporadic and outbreak cases of HUS [28, 39].

In addition to Vtxs, LEE, and other pathogenicity islands as well as non-LEE encoded Type III efectors, which are chromosomally encoded, all clinical isolates of E. coli O157:H7 possess a putative virulence plasmid called pO157. Similar plasmids are commonly present in non-O157 VTEC strains [15, 27] and a considerable variability in them has also been observed among them. Furthermore, some O26 and O145 strains that belonged to identical pulsed-field gel electrophoresis types could be further discriminated by the detection of plasmid-encoded genes [43, 47].

Several plasmid genes have been previously characterized including those that code for a haemolysin (ehxA) [42], a catalase-peroxidase (katP) [5], an extracellular serine protease (espP) [6], a zinc metalloprotease (stcE, also called tagA) [20, 34], a subtilase cytotoxin (subAB) [36], among others. EhxA was the first described virulence factor of pO157 [42] and a sequence of the gene has been used as a diagnostic probe for E. coli O157:H7 and often EHEC. KatP may help E. coli O157:H7 colonize host intestines by reducing oxidative stress and using the by-product oxygen in deprived oxygen conditions of the host intestine [5, 21]. EspP is known to cleave pepsinA and human coagulation factor V which could contribute to mucosal hemorrhage observed in HC [6] and influences the intestinal colonization and adherence in bovines [8]. StcE contributes to intimate adherence of this bacterium to host cells [14]. SubAB has been described in certain highly virulent VTEC strains which are negative for the LEE and shown to be cytotoxic to Vero cells and lethal for mice [26, 38].

A positive PCR reaction with primers specific for vtx1 or vtx2 can evidence the presence of VTEC in a sample, but the use of primers capable of detecting accessory virulence genes provides additional information that may also have great epidemiological value [35]. The aim of this study was to perform a multiplex PCR assay for the simultaneous detection of five putative virulence plasmid genes. This complementary technique would allow a further virulence characterization of VTEC useful to detect interstrain differences for epidemiological studies and to provide information that could be related to the risk of human infection.

Materials and Methods

Bacterial Isolates

A total of 39 VTEC strains from the collection of the Laboratorio de Inmunoquímica y Biotecnología (UNCPBA, Tandil, Argentina) were investigated. The strains belong to 20 different serotypes and were selected at random from a variety of serotypes in order to evaluate the multiplex assay. The serotypes analysed were: O2:H5, O8:H16, O8:H19, O20:H19, O22:H8, O39:H49, O79:H19, O91:H21, O112:H2, O113:H21, O113:H-, O117:H7, O141:H7, O145:H-, O157:H7, O171:H2, O171:NT (O171:H-non-typeable), O174:H21, O178:H19, ONT:H21 (O-non-typeable:H21). The strains had been isolated from cattle, patients with diarrhoea and contaminated food and previously analysed by PCR to detect the presence of genes coding for verotoxin 1 and 2 (vtx1 and vtx2), intimin (eae), enterohaemolysin (ehxA), STEC autoagglutinating adhesin (saa) and subtilase cytotoxin (subAB) [4, 13, 22, 31, 32, 41]. Reference strain EDL933 (serotype O157:H7) was kindly provided by Dr. Jorge Blanco (E. coli Reference Laboratory, University of Santiago de Compostela, Lugo, Spain).

Multiplex PCR Design

The plasmid genes detected in this study were katP (periplasmic catalase-peroxidase) [5], subAB (subtilase cytotoxin) [36], espP (extracellular serine protease) [6], stcE (metalloprotease) [20], ehxA (enterohaemolysin A) [42]. PCR primers specific for four of these genes, were designed with reference to published sequence data for the pO157 (AF074613) [7] using Primer-BLAST software [40], from the BLAST page on the NCBI site (http://blast.ncbi.nlm.nih.gov/Blast.cgi). In order to get a convenient electrophoretic separation, each primer set was designed to obtain a certain amplicon size. Previously published PCR primers were used for the detection of subA [36]. Details of the primers nucleotide sequence, the specific gene region amplified, and the size of the PCR products are listed in Table 1.

Table 1 Characteristics of the PCR primers used in the study
Full size table

PCR Reactions

Crude DNA extracts were obtained from a loop of frozen bacteria suspended in 500 μl of sterile water and boiled for 15 min. Each PCR reaction was carried out in a 25 μl volume containing, 1× PCR Buffer (50 mM KCl, 10 mM Tris pH9, 0.1% Triton X-100), 2 mM MgCl2, 200 μM of each dNTP, 25 pmol of each primer for subA, espP, katP, hlyA genes, 5 pmol of each primer for stcE gene, 1U Taq DNA polymerase (InBio, Highway) and 2.5 μl of crude DNA extract. After an initial denaturation step of 5 min at 94°C the samples were subjected to 30 cycles, each consisting of 30 s at 94°C (denaturing), 60 s at 58°C (annealing) and 150 s at 72°C (extension). The reaction was completed with a final extension step of 5 min at 72°C. PCR products were visualized in 2% agarose gels stained with SYBR Safe (Invitrogen, Carlsbad, CA, USA).

Sensitivity of the Multiplex PCR

To assess the sensitivity of the multiplex PCR, serial tenfold dilutions of a colony of O157:H7 VTEC strain (FC O157) resuspended in 1 ml of saline solution were performed. Crude extracts of these samples were then subjected to the pentavalent PCR assay.

Results

The reaction conditions for the multiplex PCR assay were optimized to ensure that all target gene sequences could be satisfactorily amplified, rendering amplicons easily visible on a gel. First, to standardize the conditions of the assay each one of the five genes (katP, espP, subA, stcE, ehxA) was amplified alone, testing different annealing temperatures and primer concentrations.

The five primer pairs used in the assay did not interfere with each other and generated amplification products of 914, 774, 556, 399, and 262 bp. The genes were detected with good efficiency in the multiplex PCR and the reaction conditions resulted in strong reproducible PCR amplicons of the predicted size in all the studied isolates. The five PCR products could still be seen in the sample that contained a 103-fold-diluted VTEC colony (equivalent to ~10 VTEC CFU per PCR reaction), meanwhile two of them (stcE and ehxA) could be detected even at higher dilutions.

The PCR products had sizes different enough to be easily identified by agarose gel electrophoresis. In order to resolve all the bands it was essential to run 2% agarose gels at 100 V for at least 60 min (Fig. 1). Clear PCR products were obtained for all five genes from pooled DNA of one O157:H7 strain (katP + espP + stcE + ehxA+) and one O20:H19 strain (espP + subA + ehxA+), indicating a lack of interference between any of the primer pairs or amplicons (figure not shown).

Fig. 1
figure 1

Multiplex PCR of selected analysed strains. Bands corresponding to katP, espP, subA, stcE, and ehxA are indicated. Lanes correspond to PCR products obtained from different strains

Full size image

The most commonly detected gene was espP, which was present in 37/39 (95%) of the VTEC strains tested. The ehxA gene was present in 23/39 (59%), meanwhile subA in 12/39 (31%), katP in 4/39 (10%), and stcE in 2/39 (Table 2). Those strains that were positive for subA harbored neither stcE nor katP.

Table 2 Putative plasmid virulence genes detected in 39 VTEC strains by the multiplex PCR analysis
Full size table

The multiplex PCR assay allowed the detection of six plasmid profiles (Table 3). Among the strains studied, the most frequent profile was espP + (16/39 strains), followed by the profile espP + subA + ehxA + (12/39 strains).

Table 3 Different distribution of plasmid encoded genes observed by serotype and frecuencies
Full size table

All the strains tested had been previously screened for the presence of two of the virulence genes detected by this PCR assay (ehxA and subA) [4, 13, 22, 31, 32]. There was a total correlation between the previous results for those genes and those obtained by the multiplex PCR described in this study.

Discussion

Several multiplex PCRs have been described for the simultaneous gene amplification of virulence factors, traits and serotype identification in VTEC isolated from different sources [9–12, 24, 25, 29, 30, 33, 37, 45], however, none of them amplified plasmid genes exclusively. In this study, the authors developed a pentavalent PCR assay for the detection of genes encoded in VTEC plasmids, katP, espP, subA, stcE, and ehxA, designing specific primers for katP, espP, stcE, and ehxA genes and combining these primers with those proposed by Paton and Paton [36] for subA gene.

The results of this study indicate that it is possible to perform simultaneous amplification and search for recognized plasmid-encoded virulence markers from different E. coli serotypes and apply this technique to the genetic characterization of E. coli strains isolated from reservoirs, foods or patients. This pentavalent PCR showed high sensitivity and specificity.

The aim of the study was to develop and evaluate a multiplex PCR assay with strains selected at random and not to analyse the relationship between the plasmid genes detected by this assay and the strain serotypes. However, in this set of strains the authors identified that espP was the most prevalent gene, meanwhile katP was present only in serotypes O145:H- and O157:H7, and stcE only in O157:H7 strains.

Pathogenic VTEC may arise when a given set of virulence genes come together in an E. coli host [28], but the bacterial determinants necessary and sufficient for VTEC to cause severe disease remain unknown. The identification of such virulence determinants is expected to have important public-health implications [46]. In view of the increasing prevalence of non-O157 serotypes in human VTEC diseases [16], there is now a need for comprehensive data on the virulence properties of these VTEC strains, including those encoded by plasmids. This knowledge could allow the correlation of some patterns of plasmid- encoded virulence genes and disease severity caused by particular serotypes other than O157:H7.

This rapid PCR assay can be used in combination with, for example, the one described by Paton and Paton [35] for direct detection of VTEC. The multiplex PCR described here offers a complementary virulence characterization of VTEC in order to provide data to assess the potential risk for human infection of isolates from different sources.