This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schierack, P.
Right arrow Articles by Vahjen, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schierack, P.
Right arrow Articles by Vahjen, W.
Agricola
Right arrow Articles by Schierack, P.
Right arrow Articles by Vahjen, W.

 Previous Article  |  Next Article 

Applied and Environmental Microbiology, October 2006, p. 6680-6686, Vol. 72, No. 10
0099-2240/06/$08.00+0     doi:10.1128/AEM.02952-05
Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Virulence Factor Gene Profiles of Escherichia coli Isolates from Clinically Healthy Pigs

Peter Schierack,1* Hartmut Steinrück,2 Sylvia Kleta,1 and Wilfried Vahjen3

Institut für Mikrobiologie und Tierseuchen, Fachbereich Veterinärmedizin, Freie Universität Berlin, Berlin, Germany,1 Nationales Referenzlabor für E. coli, Bundesinstitut für Risikobewertung (BfR), Berlin, Germany,2 Institut für Tierernährung, Fachbereich für Veterinärmedizin, Freie Universität Berlin, Berlin, Germany3

Received 14 December 2005/ Accepted 17 June 2006


arrow
ABSTRACT
 
Nonpathogenic, intestinal Escherichia coli (commensal E. coli) supports the physiological intestinal balance of the host, whereas pathogenic E. coli with typical virulence factor gene profiles can cause severe outbreaks of diarrhea. In many reports, E. coli isolates from diarrheic animals were classified as putative pathogens. Here we describe a broad variety of virulence gene-positive E. coli isolates from swine with no clinical signs of intestinal disease. The isolation of E. coli from 34 pigs from the same population and the testing of 331 isolates for genes encoding heat-stable enterotoxins I and II, heat-labile enterotoxin I, Shiga toxin 2e, and F4, F5, F6, F18, and F41 fimbriae revealed that 68.6% of the isolates were positive for at least one virulence gene, with a total of 24 different virulence factor gene profiles, implying high rates of horizontal gene transfer in this E. coli population. Additionally, we traced the occurrence of hemolytic E. coli over a period of 1 year in this same pig population. Hemolytic isolates were differentiated into seven clones; only three were found to harbor virulence genes. Hemolytic E. coli isolates without virulence genes or with only the fedA gene were found to be nontypeable by slide agglutination tests with OK antisera intended for screening live cultures against common pathogenic E. coli serogroups. The results appear to indicate that virulence gene-carrying E. coli strains are a normal part of intestinal bacterial populations and that high numbers of E. coli cells harboring virulence genes and/or with hemolytic activity do not necessarily correlate with disease.


arrow
INTRODUCTION
 
Escherichia coli bacteria are a dominant aerobic, intestinal bacterial population in swine, colonizing the intestine distal to the stomach. It is estimated that 1 to 4% of all cultivable bacteria of the colon are E. coli bacteria, and up to 1010 E. coli bacteria can be detected in 1 gram of feces (24, 26). Nonpathogenic E. coli (commensal E. coli) strains are thought to maintain the physiological milieu of the gut and support digestion as well as defend against enteric pathogens. Other E. coli strains carry and express virulence genes, can cause severe outbreaks of diarrhea (pathogenic E. coli), and are responsible for major economic losses in pig rearing (28). Typical swine-pathogenic E. coli strains include edema disease-causing E. coli (EDEC), which expresses hemolysin, F18 fimbriae (fedA), and the exotoxin Shiga toxin 2e (stx2e), and enterotoxigenic E. coli (ETEC), which expresses F4 (K88; the faeG gene), F5 (K99; the fanA gene), F6 (P987; the fasA gene), or F41 (fimF41a) fimbriae and the heat-stable (ST; the est genes) or heat-labile (LT; the elt genes) toxins (8).

Virulence genes are often located on transmissible genetic elements and can thus be transmitted to receptive E. coli recipient strains (3). In addition to virulence factor typing, serotyping of pathogenic E. coli strains is routinely used to classify isolates and to determine associations between serotypes, virulence, and epidemiology. Porcine pathogenic E. coli strains belong to a limited number of serogroups, with O8, O108, O138, O139, O141, O147, O149, and O157 being the most commonly reported (8). Screening of large numbers of different E. coli isolates from diarrheic animals of different animal populations has been used to survey associated virulence genes and serotype distributions.

In our study, we focused on clinically healthy pigs of a single animal population. Two approaches were chosen. (i) We screened randomly isolated intestinal E. coli strains for the stx2e, faeG, fanA, fasA, fedA, fimF41a, est-Ib, est-II, and elt-Ia virulence genes (10 isolates per animal). We also tested isolates from a diarrheic piglet of this population. (ii) We monitored the herd for the presence of hemolytic E. coli over a period of 1 year. We concentrated on hemolytic E. coli strains because these are readily identified on blood agar plates and because hemolysis appears to be associated with pathogenicity (8). We followed the occurrence of hemolytic E. coli and the transmission from sows to the corresponding litters over the same period and classified isolates by chromosomal restriction enzyme analysis and determination of virulence genes and serotypes.

The data from this study show numerous virulence factor gene profiles for E. coli strains from clinically healthy pigs and a poor association between hemolysis and virulence genes. In addition, slide agglutination tests with OK antisera intended for screening live cultures against the O8:K87, O108:K–, O138:K81, O139:K91, O141:K85ab, O141:K85ac, O147:K89, O149:K91, and O157:K– groups were found to be reliable only for the identification of "typical" pathogenic E. coli isolates and insufficient for the classification of other isolates.


arrow
MATERIALS AND METHODS
 
Animals and housing.
Pigs were housed as previously described (15). Briefly, Landrace x Duroc sows were housed in an environmentally regulated building in groups of three to four sows without straw bedding on a concrete floor. After weaning, piglets from the sows were reared together with littermates as pairs or triplets in pens in flat-deck batteries (2 m2/pen), with pens segregated from each other by continuous partitions 40-cm high topped with an additional 40 cm of metal grid. Since monitoring of the microbial composition in feces and digesta was one of the main aims of the study, the application of prophylactic and therapeutic antibiotics to gestating and lactating sows and to piglets was prohibited during and for at least 3 months prior to the trial. Animals (in the case of nursing piglets, the whole litter) requiring medication which might have the potential to alter the intestinal microbiota were excluded from the trial. Pigs were fed according to the recommendations of the Verband Deutscher Landwirtschaftlicher Untersuchungs- und Forschungsanstalten (Association of German Agricultural Tests and Research Institutions) (19).

Isolation of E. coli from clinically healthy piglets and sows.
Feces from grower piglets (42 days old) were sampled from the first defecation after the daily cleaning of the pens. From each piglet pen (n = 21; corresponds to 21 animals), one fecal sample was collected. From each sow (n = 13), one rectal fecal sample was taken. Fecal samples from the animals were collected and processed within 24 h. One gram of feces was resuspended in 10 ml 0.85% NaCl, and serial dilutions were plated onto McConkey-lactose agar plates. The plates were incubated overnight at 37°C. Lactose-positive (pink to red) colonies were counted as E. coli. Ten E. coli isolates per piglet (n = 210, where n is the total number of isolates) and 10 E. coli isolates per sow (n = 121; 9 isolates were lost during subsequent cultivation) were chosen for virulence gene determination. According to the calculations of Schlager et al. (25), 10 isolates are sufficient to determine the most common clones in fecal samples, with a 90% chance of detection of a clone which has a frequency of at least 20% in the sample (25).

Isolation of E. coli from one diarrheic piglet.
A 35-day-old piglet with watery diarrhea was sacrificed, and samples of the mucosa were taken from the stomach (pars esophagus), duodenum, jejunum (proximal, medial, and distal), ileum, cecum, and colon ascendens. Mucosal samples were washed once with phosphate-buffered saline and incubated for 30 min with minimal medium (15 g · liter–1 Na2HPO4 · 12H2O, 3 g · liter–1 KH2PO4, 1 g · liter–1 NaCl, 1 g · liter–1 NH4Cl, 80 g · liter–1 L-methionine, 0.5 g · liter–1 L-cysteine, 0.3 g · liter–1 Na formaldehyde sulfoxylate, 10 mg · liter–1 hemin, and 4 mg · liter–1 vitamin K1). Mucosal tissues were obtained by scraping with a scalpel, homogenized, and after serial dilution, plated as described above. More isolates from the single clinically ill piglet were investigated (96 E. coli isolates) than from the healthy piglets to determine whether there might be a correlation between numbers of E. coli clones and specific virulence factor gene profiles and illness.

Isolation of hemolytic E. coli.
The same pig population was screened over a period of 1 year for the occurrence of hemolytic E. coli, beginning shortly after the determination of the virulence factor gene profiles of E. coli from the clinically healthy sows and piglets. E. coli was isolated as previously described from fecal samples taken over a period of 1 year (24). Fecal samples from sows (81 samples from 27 sows) were collected rectally at intervals of 1 or 2 months for each sow. Intestinal fecal samples from piglets (46 samples) were collected from individual animals of rearing groups (four animals per rearing group) which had been sacrificed for isolation of intestinal sections. Intestinal sections were clamped and sealed off with surgical thread before removal to prevent loss of contents and to maintain the intestinal conditions during anaerobic transport. The intestinal contents of the sacrificed piglets were taken from ileal and/or colon sections at day 14, 28, 35, and 56 postpartum. Samples of intestinal contents or feces were used directly for growing E. coli isolates on blood agar plates, followed by single-colony purification on Gassner and CHROMagar orientation agars. The hemolysin genotype was determined by PCR using primers hlyAfw (5'GTCCATTGCCGATAAGTTT3') and hlyArev (5'AAGTAATTTTTGCCGTGTTTT3'). Briefly, colonies of E. coli isolates were diluted in deionized, distilled water (ddH2O) and heated for 5 min at 99°C. After 5 min of incubation on ice, samples were centrifuged for 5 min at 6,000 x g. One-microliter volumes of the supernatants (lysates) were used for PCR. PCR conditions consisted of denaturation for 15 min at 95°C, followed by 10 cycles of 94°C for 30 s, 50°C for 30 s, and 72°C for 60 s; 20 cycles of 94°C for 30 s, 54°C for 30 s, and 72°C for 60 s; and 10 min at 72°C. The hemolysin phenotype was determined by description of the hemolytic zone (kinetics, morphology, and intensity), with determinations made 3 and 16 h after culture from both surface and stab inoculations onto washed sheep blood agar plates.

Determination of E. coli OK groups (slide test).
E. coli isolates from blood agar plates subjected to single-colony purification on Gassner and CHROMagar orientation plates (Mast Diagnostica, Reinfeld, Germany) were resuspended in physiological saline and subjected to slide agglutination tests with OK antisera intended for screening live cultures against the O8:K87, O108:K–, O138:K81, O139:K91, O141:K85ab, O141:K85ac, O147:K89, O149:K91, and O157:K– groups (BfR Dessau, Berlin, Germany).

Determination of E. coli O:H groups (standard methods).
Determinations of O:H serotypes were performed as previously described (4, 20) using diagnostic antisera produced in the German E. coli National Reference Laboratory (Nationales Referenzlaboratorium für E. coli, Berlin). The corresponding E. coli O1 through O181 (n = 173) and H1 through H56 (n = 53) antigenic test strains were kindly provided by Ida and Frits Ørskov and Flemming Scheutz (Statens Seruminstitut, Copenhagen, Denmark).

O typing was performed with 96-well U-form microtiter plates. For O typing, 20 ml brain heart infusion broth was inoculated with single, smooth colonies grown on blood agar plates and incubated at 37°C for about 6 h. Cell suspensions were boiled for 1 h or autoclaved for 2 h to destroy A antigens of O nontypeable strains. One-hundred-microliter volumes of the suspensions were tested first against 21 O antiserum pools, A to W (100 µl antiserum). The composition of the different O serum pools according to the antigenic relationships of the O groups has been described by the International Escherichia and Klebsiella Centre (http://www.ssi.dk). Bacterial suspensions were then tested with the individual sera of the positive pools. Single serum samples with positive reactions were finally titrated against the test antigens. The plates were read after 6 h (stages 1 and 2) or after overnight incubation at 50°C.

Only well-flagellated, motile E. coli cells are suitable for H typing. Therefore, strains were propagated by two or more passages through semisolid agar in U tubes. Bacteria from the zones with the highest motility were inoculated into brain heart infusion broth. H antigens were prepared by incubation for 5 to 6 h in roller tubes at 37°C, followed by the addition of 0.5% formaldehyde. H agglutinations were performed in small tubes with 200 µl cell suspension and 200 µl antiserum. The tubes were incubated in a water bath at 50°C for 1 to 2 h. An agglutinoscope or hand lens was used for reading the agglutinants. H typing also consisted of three stages: examination with 10 H antiserum pools, A to K, followed by determination of a positively reacting single component(s), and confirmation or exclusion of the putative H type by quantitative agglutination (titration).

Virulence gene determination using PCR and dot blot hybridization.
The presence of the stx2e, faeG, fanA, fasA, fedA, fimF41a, est-Ib, est-II, and elt-Ia virulence genes in the E. coli isolates was assayed using multiplex PCR followed by dot blot hybridization. The primers and multiplex PCR conditions used were previously described (17). Briefly, colonies of E. coli isolates were diluted in ddH2O and heated for 5 min at 99°C. After 5 min of incubation on ice, the samples were centrifuged for 5 min at 6,000 x g. One-microliter volumes of the supernatants (lysates) were used for PCR. PCR conditions consisted of denaturation for 15 min at 95°C, followed by 10 cycles of 94°C for 30 s, 50°C for 30 s, and 72°C for 60 s; 20 cycles of 94°C for 30 s, 54°C for 30 s, and 72°C for 60 s; and 10 min at 72°C. For additional confirmation, we tested the PCR products by dot blot hybridization. The oligonucleotide probes for the hybridizations were created using the program Primer3 (22) and are indicated in Table 1. Twenty-five-microliter volumes of the PCR products were diluted into 150 µl ddH2O, and 50-µl volumes of the dilutions were spotted onto nylon membranes using the dot blot 96 system (Biometra, Göttingen, Germany). The membranes were baked at 102°C for 30 min and incubated overnight at 40°C with digoxigenin-labeled oligonucleotide probes (10 pmol/100-mm2 nylon membrane). The hybridized membranes were washed twice with 2x SSC (1x SSC is 0.15 M NaCl plus 0.015 mM sodium citrate), 0.1% sodium dodecyl sulfate for 5 min at 20°C and twice with 0.1x SSC, 0.1% sodium dodecyl sulfate for 15 min at hybridization temperatures. Development of the membranes for visualization was performed according to the manufacturer's recommendations (Roche, Mannheim, Germany).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Oligonucleotide probes for dot blot hybridization used for detection of virulence genes of porcine E. coli isolates

Differentiation of E. coli clones by pulsed-field gel electrophoresis.
To differentiate the E. coli clones, we used pulsed-field gel electrophoresis as previously described (11). Bacterial clones were grown in LB medium overnight and adjusted to an optical density at 600 nm of 1.4. Samples (1.5 ml) were centrifuged and washed by resuspension twice with phosphate-buffered saline. The bacterial solutions were embedded in 1.2% pulsed-field certified agarose (Bio-Rad, München, Germany) in TE buffer (10 mM Tris-HCl, 10 mM EDTA, pH 7.5). The solidified agar blocks were incubated for 24 h with proteinase K in ESP buffer (500 mM EDTA, 1% sarcosyl, pH 9.5) and washed three times for 1.5 h each with TE buffer. Bacterial DNA was digested with 20 U XbaI at 37°C overnight. The digested blocks were embedded in a 1.2% pulsed-field agarose gel, and DNA fragments were separated for 22 h at 6 V and 50 Hz and examined by ethidium bromide staining. Isolates were designated as single clones if they did not differ by more than one band.


arrow
RESULTS
 
Virulence factor gene profiles of E. coli isolates from clinically healthy pigs.
We isolated E. coli from feces from pigs of a clinically healthy pig population and screened 10 isolates per individual for virulence gene-carrying E. coli. From 13 sows, we identified E. coli isolates with 12 different virulence factor gene profiles. From 21 piglets from the same population, we detected E. coli isolates showing 21 different virulence factor gene profiles. In total, we identified 24 different virulence factor gene profiles, with up to four different virulence genes identified in a single E. coli isolate (Table 2). Of all E. coli isolates from the piglets, 73.8% carried at least one virulence gene, compared to 59.5% of all E. coli isolates from the sows. In total, 68.6% of the isolates from the piglets and the sows carried at least one virulence gene. Up to seven E. coli isolates with different virulence factor gene profiles were found in one fecal sample from the piglets. Of the 10 E. coli isolates from each piglet, as many as 10 carried virulence genes. For comparison, we included in Table 2 information about virulence factor gene profiles of classical porcine pathogenic E. coli, EDEC, and ETEC. To illustrate possible different virulence factor gene profiles of E. coli isolates from different piglets, we chose the results from six exemplary piglets for representation in Table 3. The most obvious difference in the isolates was the frequency of fedA-positive E. coli in the piglets compared to that in the sows. The fedA gene was present in 34 isolates from 10 piglets, whereas no isolate from the sows showed this factor. There were no other obvious patterns specific for either piglets or sows.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Occurrence of virulence factor gene profiles in E. coli isolates from a clinically healthy pig population


View this table:
[in this window]
[in a new window]
  		TABLE 3. Virulence factor gene profiles of 10 E. coli isolates from six exemplary healthy piglets

Virulence factor gene profiles of E. coli from one diarrheic piglet.
During the course of the study after the sampling of the healthy pigs, one piglet from the population showed clinical symptoms of diarrhea (watery diarrhea). From 96 E. coli isolates from this single animal, we identified 18 different virulence factor gene profiles with up to four different virulence genes in 1 isolate (Table 2). Of all E. coli isolates from this piglet, 66.6% carried at least one virulence gene. The intestinal distribution of the virulence gene-carrying E. coli isolates from the sick animal is shown in Table 4. Quantitative differences in occurrence of virulence genes between this animal and the clinically healthy piglets have to be considered carefully since the data were obtained from only one clinically ill piglet and therefore are unlikely to be representative. However, there were more stx2e-positive E. coli isolates obtained from the diarrheic piglet than from the clinically healthy piglets..


View this table:
[in this window]
[in a new window]
  		TABLE 4. Intestinal distribution of virulence factor gene profiles in E. coli from one diarrheic piglet

Occurrence of hemolytic E. coli over a period of 1 year.
We also monitored the virulence factor gene profiles of E. coli isolates in the pig population over a period of 1 year, concentrating on hemolytic E. coli since it is readily identified using blood agar plates and since hemolytic activity appears to be associated with pathogenicity (8). Over the course of 1 year, fecal samples from 27 sows (three samples per sow) and piglets from the litters of these sows (4 piglets per litter, litters from 11 sows) were tested for the presence of hemolytic E. coli. Twenty-one fecal samples from the sows were found to harbor hemolytic E. coli. However, in only four sows were we able to detect hemolytic E. coli more than once. A total of 15 piglets from 10 different litters carried hemolytic E. coli. However, only one piglet in each of 10 litters carried hemolytic E. coli, and in 5 litters two piglets carried hemolytic E. coli. In general, where we isolated hemolytic E. coli in piglets, we also found hemolytic E. coli in the corresponding sows.

To determine whether the different hemolytic E. coli isolates were clones, we used genomic DNA macrorestriction analyses. As shown in Table 5, the 40 independent hemolytic isolates belonged to seven different clones. There were no similarities in the band patterns of the individual clones. Only three clones (clones 3, 4, and 7) carried virulence genes. All isolates of one clone showed the same virulence factor gene profile. The most common clone (clone 1) was found in 11 sows and 5 piglets from different litters. Two other clones were found only once (clones 6 and 7). Hemolytic E. coli isolated repeatedly in any one sow was found, with a single exception, to represent the same clone. In contrast, hemolytic E. coli isolates found more than once in a litter of piglets were of different clonal origins. Of the 15 hemolytic isolates from piglets, only 1 was found in the corresponding sow. The hemolysis type was determined to be alpha-hemolysin for all hemolytic clones by PCR (genotype) and by description of the hemolytic zone (phenotype).


View this table:
[in this window]
[in a new window]
  		TABLE 5. Macrorestriction analysis types (clones) of hemolytic E. coli isolates from clinically healthy pigs over a 1-year period

Twenty-five randomly chosen hemolytic isolates from sows and piglets were serotyped in a blind manner with antisera against groups O8:K87, O108:K–, O138:K81, O139:K91, O141:K85ab, O141:K85ac, O147:K89, O149:K91, and O157:K– by slide agglutination tests. The seven fedA- and stx2e-positive isolates (clones 4 and 7) were all identified as O141:K85ab and O141:K85ac. All other hemolytic E. coli clones (clones 1, 2, and 3) were nonserotypeable. One clone was classified into several serogroups (Table 6). All clones were additionally serotyped by standard methods (Table 6).


View this table:
[in this window]
[in a new window]
  		TABLE 6. Serotyping of hemolytic E. coli with slide agglutination tests with OK antisera and determination of E. coli O:H groups (standard methods)


arrow
DISCUSSION
 
The genome of E. coli can be considered to be composed of a universal core of genes providing the backbone of genetic information and a flexible gene pool which is clone specific (3). This flexible gene pool includes virulence genes, which are often carried by mobile genetic elements. The gene for Shiga toxin Stx2e is located on the chromosome on a defective prophage (18), whereas the genes for the F4, F5, F6, and F18 fimbriae and for the heat-labile and heat-stable enterotoxins are located on plasmids. Some genes seem to occur in association more frequently than others, e.g., est-Ib with elt, est-Ia with fanA, est-Ia with fasA, and est-Ia with est-Ib (5, 16). Mobile genetic elements can be laterally transferred, and it was originally estimated that between 10% and 16% of the whole E. coli chromosome arose through lateral gene transfer, enabling clones with high lateral gene transfer rates to adapt to special environmental conditions (12). Frequently transferred virulence plasmids carrying genetic information required for pathogenic features generally confer a selective advantage to their host (9). Conjugation is probably the major mode of exchange of genetic information in the natural environment (23, 29) and in the intestinal tracts of humans and animals (10).

Previous attempts to predict the kinetics of plasmid transfer in the animal gut have been based on the assumption that the intestine can be compared to a chemostat (6, 7). In a mixed liquid culture, conjugation rates (transconjugants per donor per minute) of 0.2 have been observed, with recent transconjugants functioning as donors only after a maturation time of 40 to 80 min (1). Other studies suggest that the intestinal environment displays transfer kinetics different from those expected of a mixed liquid culture but similar to those observed in biofilms. In biofilms, plasmid transfer occurred very rapidly only in the initial phase of colonization (13). Escherichia coli grows rapidly in the intestinal mucus layer, whereas only very slow growth takes place in the gut contents (14, 21, 27). Plasmid transfer might therefore occur mainly within the fraction of intestinal E. coli bacteria present in the mucus (21). Clones of E. coli originally isolated from animals were capable of exchanging a 60-kb plasmid under culture conditions simulating the anaerobic gut environment. Regarding our results, the high recovery rates of virulence gene-harboring E. coli combined with the different virulence gene profiles implies high rates of horizontal gene transfer and a dynamic gene pool in our probed E. coli population. The presence of high numbers of virulence gene-carrying E. coli cells in a limited population might accelerate the diversity of virulence factor gene profiles and numbers of virulence gene-carrying E. coli cells in individual animals.

Multiple virulence factors are required to create a pathogenic E. coli isolate. Frydendahl (8) reported that most E. coli isolates (of 219 isolates) from pigs with postweaning diarrhea or edema disease carried the faeG (44.7% of tested isolates), fedA (39.3%), fasA (0.9%), est-Ib (77.6%), elt (61.6%), est-Ia (26.5%), and stx2e (16.4%) genes and that six pathotypes accounted for 65.7% of all isolates investigated. Both fimbrial adhesins and toxins are correlated with the pathogenicity of E. coli in pigs. In this study, we screened a clinically healthy pig population and found that 73.8% of the piglet E. coli isolates and 59.5% of the sow E. coli isolates carried at least one virulence gene. As many as all of the isolates from one piglet were found to harbor virulence genes, and as many as all 10 isolates tested carried genes encoding enterotoxins. As the pigs examined were clinically healthy, this suggests that the virulence genes of these E. coli isolates were either not or only very poorly expressed in the intestine, that the ability of these E. coli isolates to cause disease might be suppressed, or that these E. coli isolates represented only a very small fraction of the total intestinal microflora. Since the E. coli isolates were randomly chosen for screening, the last suggestion appears unlikely. Rather, the results indicate that virulence gene-carrying E. coli isolates are a normal part of the intestinal bacterial population. Additionally, these virulence gene-carrying E. coli isolates must be considered to be commensal E. coli isolates unless they are shown to cause disease.

The virulence factor gene profiles of E. coli from the diarrheic piglet were similar to the virulence factor gene profiles of E. coli from healthy piglets, making it difficult to identify a particular clone as the causative agent of the diarrhea. In addition, quantitative differences in the occurrence of virulence genes between this animal and the clinically healthy piglets have to be considered carefully, as the distribution of E. coli strains might be dependent on intestinal localization, and we tested E. coli from different intestinal sections only from the clinically ill piglet, not from the clinically healthy pigs. E. coli populations from feces (clinically healthy pigs) do not always reflect the populations in the lower bowel, as feces remain for longer periods in the rectum and the populations could change compared to those in the colon. However, as numbers of stx2e-positive isolates increased in this diarrheic piglet, stx2e-positive strains might have been involved in the diarrhea.

In this single diarrheic animal, we detected 18 different virulence factor gene profiles. The virulence gene-carrying E. coli cells were evenly distributed in all sections of the gut. Since we did not examine the intestinal distributions of the various E. coli isolates from the healthy pig population, we cannot draw any conclusions about the colonization of the small intestine versus that of the lower bowel by virulence gene-carrying E. coli in the clinically healthy pigs.

In one fecal sample each from both a healthy piglet and the diarrheic piglet, we found different E. coli isolates harboring single fimbrial as well as single toxin genes. This suggests that the identification of possible intestinal pathogenic E. coli strains as the causative agents of disease by PCR from fecal DNA preparations will lead to false interpretations. Intestinal distribution and enumeration of potential pathogenic E. coli clones can be determined only by single (viable)-isolate testing.

The expression of hemolysin and hemolytic activity is thought to correlate with the pathogenicity of E. coli (8). In addition to classification by virulence factor gene profiles and serotyping, pathogenic E. coli isolates have also been classified according to their hemolytic activities (8). During our 1-year study, we followed the occurrence of hemolytic E. coli in a single, healthy pig population. Of 27 sows, we detected hemolytic E. coli at least once in 17 sows and in 10 of 11 piglet litters. It seems clear that these hemolytic E. coli bacteria were established in this pig population and were transmitted from the sows to their piglets. We were able to differentiate the 40 different hemolytic isolates into seven different clones. Only three of the hemolytic E. coli clones carried at least one virulence gene. The virulence factor gene profile of one isolate was assigned to a specific clone. One hemolytic E. coli clone (clone 1) was found in 11 sows and 5 piglets, whereas the other hemolytic E. coli clones were found just once. While measures were taken to keep the sows and their litters physically separated from the other groups, within a given sow-litter group, it might be expected that all E. coli bacteria had the possibility of spreading uniformly in the population. The results show, however, that the establishment of an organism in a single animal is individual and time dependent. We were able to detect transmission of only one particular hemolytic clone from a sow to its litter in a single case; none of the other hemolytic E. coli clones isolated from piglets were found in the corresponding sows. These observations suggest that the intestinal milieus of sows and piglets differ to such an extent that only distinct E. coli clones are able to become established or that the colostrum of the sows inhibited the colonization of distinct clones. Regardless of the reasons for the differing establishment capabilities, the hemolytic activity of E. coli clones itself is not sufficient evidence to establish the pathogenicity of E. coli isolates, nor does it serve to trace the transmission of E. coli from sows to piglets.

Serotyping is one tool for determining the pathogenicity and epidemiology of E. coli strains. For slide agglutination, sera produced with a live or formalin-treated E. coli culture (so-called OK sera) contain antibodies against most of the known and unknown structures found on the surface. The agglutination observed is the combined consequence of several independent antigen-antibody systems, and it is not simple to demonstrate whether one or more possible surface antigens of the unknown cultures are involved. Nevertheless, slide agglutination is an extremely simple, sensitive, and for many purposes useful method and is used primarily to screen many colonies in a limited number of sera. But, usually, the slide agglutination test needs confirmation by an agglutination titration under more-standardized conditions in tubes or trays (20). In practice, the slide agglutination test with OK sera is commonly and solely used to screen porcine E. coli from diarrheic animals (2). Antisera against the O8:K87, O108:K–, O138:K81, O139:K91, O141:K85ab, O141:K85ac, O147:K89, O149:K91, and O157:K– groups should capture the most prevalent porcine pathogenic E. coli strains. In our study, we show that conclusions about pathogenicity based on hemolysis and slide agglutination tests of E. coli might lead to false results. We were unable to definitively serotype hemolytic E. coli isolates that were without virulence genes or those with a single gene for the fimbrial adhesin (fedA) with the slide agglutination test. Therefore, in the absence of accompanying toxin genes, E. coli strains, including hemolytic E. coli strains, with only fimbrial genes should be considered nonpathogenic. In addition, these hemolytic E. coli strains do not fit well in the serotyping scheme of pathogenic E. coli strains compared to commensal E. coli strains. In contrast, serotypes of classical pathogenic, hemolytic porcine E. coli clones (carrying the fedA and stx2e genes) were unequivocally determined by slide agglutination tests. Slide agglutination tests are therefore valuable only for determination of pathogenic E. coli when their results are combined with data for the presence of virulence genes, since nonpathogenic, commensal E. coli strains show cross-reactivity to the antisera against the O-specific antigens of pathogens. Additionally, since one animal can harbor different hemolytic clones (our unpublished data), the isolation and serotyping of one hemolytic isolate from one diarrheic animal can also lead to false interpretations.

The distinction between pathogenic E. coli strains and commensal E. coli strains which harbor presumed virulence-associated genes is blurred. We have shown that clinically healthy pigs can harbor high numbers of E. coli cells that carry virulence genes, and hemolytic activity alone is not clearly associated with pathogenicity. The occurrence of virulence factors therefore does not serve as proof of the pathogenicity of a given clone. The verification of virulent E. coli clones as agents of disease and the distinction between virulent and commensal E. coli clones will require a number of different approaches, including determination of the relative proportion of a given isolate to the total E. coli population of the animal.


arrow
ACKNOWLEDGMENTS
 
This work was supported by grant FOR 438/1-1 from the Deutsche Forschungsgemeinschaft. S. Kleta was supported by NaFöG Berlin.

We thank Annette Kurz, Stefanie Bernhard, Susanne Göbel, and Gudrun Hultsch for excellent technical assistance; Antina Lübke-Becker and coworkers for serotyping the E. coli isolates; Peter Schwerk and Karsten Tedin for isolation of the hemolytic E. coli; and Karsten Tedin and Lothar Wieler for careful reading of the manuscript.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Institut für Mikrobiologie und Tierseuchen, Postfach 040225, D-10116 Berlin, Germany. Phone: 49-30-2093 6704. Fax: 49-30-2093 6067. E-mail: schierack.peter{at}vetmed.fu-berlin.de. Back


arrow
REFERENCES
 
    1
  1. Andrup, L., and K. Andersen. 1999. A comparison of the kinetics of plasmid transfer in the conjugation systems encoded by the F plasmid from Escherichia coli and plasmid pCF10 from Enterococcus faecalis. Microbiology 145:2001-2009.[Abstract/Free Full Text]
    2. 2
  2. Do, T., C. Stephens, K. Townsend, X. Wu, T. Chapman, J. Chin, B. McCormick, M. Bara, and D. J. Trott. 2005. Rapid identification of virulence genes in enterotoxigenic Escherichia coli isolates associated with diarrhoea in Queensland piggeries. Aust. Vet. J. 83:293-299.[Medline]
    3. 3
  3. Dobrindt, U., U. Hentschel, J. B. Kaper, and J. Hacker. 2002. Genome plasticity in pathogenic and nonpathogenic enterobacteria. Curr. Top. Microbiol. Immunol. 264:157-175.[Medline]
    4. 4
  4. Ewing, W. H. 1986. The genus Escherichia, p. 93-122. In P. R. Edwards and W. H. Ewing (ed.), Edwards and Ewing's identification of Enterobacteriaceae, 4th ed. Elsevier Science Publishing Co. Inc., New York, N.Y.
    5. 5
  5. Fekete, P. Z., J. Gerardin, E. Jacquemin, J. G. Mainil, and B. Nagy. 2002. Replicon typing of F18 fimbriae encoding plasmids of enterotoxigenic and verotoxigenic Escherichia coli strains from porcine postweaning diarrhoea and oedema disease. Vet. Microbiol. 85:275-284.[CrossRef][Medline]
    6. 6
  6. Freter, R., H. Brickner, M. Botney, D. Cleven, and A. Aranki. 1983. Mechanisms that control bacterial populations in continuous-flow culture models of mouse large intestinal flora. Infect. Immun. 39:676-685.[Abstract/Free Full Text]
    7. 7
  7. Freter, R., E. Stauffer, D. Cleven, L. V. Holdeman, and W. E. C. Moore. 1983. Continuous-flow cultures as in vitro models of the ecology of large intestinal flora. Infect. Immun. 39:666-675.[Abstract/Free Full Text]
    8. 8
  8. Frydendahl, K. 2002. Prevalence of serogroups and virulence genes in Escherichia coli associated with postweaning diarrhoea and edema disease in pigs and a comparison of diagnostic approaches. Vet. Microbiol. 85:169-182.[CrossRef][Medline]
    9. 9
  9. Gordon, D. M. 1992. Rate of plasmid transfer among Escherichia coli strains isolated from natural populations. J. Gen. Microbiol. 138:17-21.[Abstract/Free Full Text]
    10. 10
  10. Jarrett, P., and M. Stephenson. 1990. Plasmid transfer between strains of Bacillus thuringiensis infecting Galleria mellonella and Spodoptera littoralis. Appl. Environ. Microbiol. 56:1608-1614.[Abstract/Free Full Text]
    11. 11
  11. Laturnus, C., J. Jores, I. Moser, P. Schwerk, and L. H. Wieler. 2005. Long-term clonal lineages within Campylobacter jejuni O:2 strains from different geographical regions and hosts. Int. J. Med. Microbiol. 294:521-524.[CrossRef][Medline]
    12. 12
  12. Lawrence, J. G., and H. Ochman. 1997. Amelioration of bacterial genomes: rates of change and exchange. J. Mol. Evol. 44:383-397.[CrossRef][Medline]
    13. 13
  13. Licht, T. R., B. B. Christensen, K. A. Krogfelt, and S. Molin. 1999. Plasmid transfer in the animal intestine and other dynamic bacterial populations: the role of community structure and environment. Microbiology 145:2615-2622.[Abstract/Free Full Text]
    14. 14
  14. Licht, T. R., T. Tolker-Nielsen, K. Holmstrom, K. A. Krogfelt, and S. Molin. 1999. Inhibition of Escherichia coli precursor-16S rRNA processing by mouse intestinal contents. Environ. Microbiol. 1:23-32.[Medline]
    15. 15
  15. Macha, M., D. Taras, W. Vahjen, A. Arini, and O. Simon. 2004. Specific enumeration of the probiotic strain Enterococcus faecium NCIMB 10415 in the intestinal tract and in faeces of piglets and sows. Arch. Anim. Nutr. 58:443-452.[CrossRef][Medline]
    16. 16
  16. Mainil, J. G., G. Daube, E. Jacquemin, P. Pohl, and A. Kaeckenbeeck. 1998. Virulence plasmids of enterotoxigenic Escherichia coli isolates from piglets. Vet. Microbiol. 62:291-301.[CrossRef][Medline]
    17. 17
  17. Moon, H. W., L. J. Hoffman, N. A. Cornick, S. L. Booher, and B. T. Bosworth. 1999. Prevalences of some virulence genes among Escherichia coli isolates from swine presented to a diagnostic laboratory in Iowa. J. Vet. Diagn. Investig. 11:557-560.[Free Full Text]
    18. 18
  18. Muniesa, M., J. Recktenwald, M. Bielaszewska, H. Karch, and H. Schmidt. 2000. Characterization of a Shiga toxin 2e-converting bacteriophage from an Escherichia coli strain of human origin. Infect. Immun. 68:4850-4855.[Abstract/Free Full Text]
    19. 19
  19. Naumann, C., and R. Bassler. 1993. Die chemische Untersuchung von Futtermitteln. VDLUFA-Verlag, Darmstadt, Germany.
    20. 20
  20. Ørskov, F., and I. Ørskov. 1984. Serotyping of Escherichia coli. Methods Microbiol. 14:43-112.
    21. 21
  21. Poulsen, L. K., T. R. Licht, C. Rang, K. A. Krogfelt, and S. Molin. 1995. Physiological state of Escherichia coli BJ4 growing in the large intestines of streptomycin-treated mice. J. Bacteriol. 177:5840-5845.[Abstract/Free Full Text]
    22. 22
  22. Rozen, S., and H. J. Skaletsky. 2000. Primer3 on the WWW for general users and for biologist programmers, p. 365-386. In S. Krawetz and S. Misener (ed.) Bioinformatics methods and protocols: methods in molecular biology. Humana Press, Totowa, N.J.
    23. 23
  23. Salyers, A. A. 1993. Gene transfer in the mammalian intestinal tract. Curr. Opin. Biotechnol. 4:294-298.[CrossRef][Medline]
    24. 24
  24. Scharek, L., J. Guth, K. Reiter, K. D. Weyrauch, D. Taras, P. Schwerk, P. Schierack, M. F. Schmidt, L. H. Wieler, and K. Tedin. 2005. Influence of a probiotic Enterococcus faecium strain on development of the immune system of sows and piglets. Vet. Immunol. Immunopathol. 105:151-161.[CrossRef][Medline]
    25. 25
  25. Schlager, T. A., J. O. Hendley, A. L. Bell, and T. S. Whittam. 2002. Clonal diversity of Escherichia coli colonizing stools and urinary tracts of young girls. Infect. Immun. 70:1225-1229.[Abstract/Free Full Text]
    26. 26
  26. Selander, R. K., J. M. Musser, D. A. Caugant, M. N. Gilmour, and T. S. Whittam. 1987. Population genetics of pathogenic bacteria. Microb. Pathog. 3:1-7.[CrossRef][Medline]
    27. 27
  27. Wadolkowski, E. A., D. C. Laux, and P. S. Cohen. 1988. Colonization of the streptomycin-treated mouse large intestine by a human fecal Escherichia coli strain: role of adhesion to mucosal receptors. Infect. Immun. 56:1036-1043.[Abstract/Free Full Text]
    28. 28
  28. Wieler, L. H., A. Ilieff, W. Herbst, C. Bauer, E. Vieler, R. Bauerfeind, K. Failing, H. Klos, D. Wengert, G. Baljer, and H. Zahner. 2001. Prevalence of enteropathogens in suckling and weaned piglets with diarrhoea in southern Germany. J. Vet. Med. B 48:151-159.[CrossRef]
    29. 29
  29. Yin, X., and G. Stotzky. 1997. Gene transfer among bacteria in natural environments. Adv. Appl. Microbiol. 45:153-212.[Medline]

Applied and Environmental Microbiology, October 2006, p. 6680-6686, Vol. 72, No. 10
0099-2240/06/$08.00+0     doi:10.1128/AEM.02952-05
Copyright © 2006, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Schierack, P., Romer, A., Jores, J., Kaspar, H., Guenther, S., Filter, M., Eichberg, J., Wieler, L. H. (2009). Isolation and Characterization of Intestinal Escherichia coli Clones from Wild Boars in Germany. Appl. Environ. Microbiol. 75: 695-702 [Abstract] [Full Text]  
  • Schierack, P., Walk, N., Reiter, K., Weyrauch, K. D., Wieler, L. H. (2007). Composition of intestinal Enterobacteriaceae populations of healthy domestic pigs. Microbiology 153: 3830-3837 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schierack, P.
Right arrow Articles by Vahjen, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schierack, P.
Right arrow Articles by Vahjen, W.
Agricola
Right arrow Articles by Schierack, P.
Right arrow Articles by Vahjen, W.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»