PDF
ePub
Citation
Print
Introduction
Escherichia (E.) coli is an important cause of diarrhea in farm animals. According to virulence properties and the clinical symptoms of the host, pathogenic E. coli strains are designated as enterotoxigenic E. coli (ETEC), attaching and effacing E. coli, enteropathogenic E. coli, Shiga toxin-producing E. coli (STEC), and necrotoxigenic E. coli [11,19]. ETEC can cause severe diarrhea in newborn calves via the production of heat-stable enterotoxin (STa). The most common observed fimbriae on ETEC from calves with diarrhea is K99 (F5) and F41; however, strains with F17 fimbriae have also been isolated [18]. STEC strains are a well recognized cause of colibacillosis in newborn calves. Even though both healthy and diarrheic calves harbor STEC in their intestine, natural outbreaks and experimental infections have documented the association of STEC with diarrhea and dysentery in young calves [9,30]. In humans, STEC can cause severe diseases, including haemorrhagic colitis and haemolytic uremic syndrome (HUS), via production of Shiga toxins. These toxins are subdivided into two groups, Shiga toxin 1 (Stx1) and Shiga toxin 2 (Stx2). Stx1 is a homologous group in which three variants (Stx1, Stx1c and Stx1d) have been described [8,39]. Stx2 is more heterogeneous and composed of several subtypes (Stx2, Stx2c, Stx2d, Stx2e, Stx2f, Stx2g and activatable Stx2d) [15,19,28,32]. In addition to toxin production, STEC strains may possess other virulence factors such as intimin (encoded by the eae gene) [19], the plasmid-encoded enterohemolysin (encoded by the ehxA gene) [31] and the STEC autoagglutinating adhesion (Saa) [26].
In Vietnam, 40% of farm animals have diarrhea, which results in major economic losses [20]. However, no studies regarding the prevalence of pathogenic E. coli as a cause of diarrhea in calves have been published to date in Vietnam. This study was designed to investigate the prevalence and characteristics of pathogenic E. coli strains from diarrheic calves in Vietnam.
Materials and Methods
Sampling and isolation of E. coli strains
Between 2006 and 2008, samples from 322 diarrheic calves (<3 months of age) were cultured for E. coli. The calves resided on 247 farms, 234 of which had one to four animals and 13 of which had >50 animals. The farms were located in six different provinces in Central Vietnam. Of the 322 calves, 46 were <7 days old and 276 were between 8- and 90-days old. None of the calves had been vaccinated and all showed symptoms of watery diarrhea at the time of sampling. Fecal samples were collected using sterile rectal swabs that were placed into tubes containing Stuart medium and immediately transported to the Laboratory in ice-cooled containers. The samples were then plated onto MacConkey's agar and incubated overnight at 37℃. From each sample, 4 to 5 lactose positive colonies were selected and confirmed to be E. coli by standard biochemical tests (Indole, methyl red, Voges-Proskauer and citrate utilization tests). One or two isolates per calf were selected for screening of virulence genes. E. coli isolates were stored in tryptic soy broth containing 20% glycerol at -70℃ for further characterization.
Detection and genotypic characterization of virulent factors of E. coli strains
The primers used for PCR and multiplex PCR are shown in Table 1. Bacterial DNA was obtained by boiling the cells at 100℃ for 15 min and then pelleting the cells by centrifugation. The supernatant was then used in the PCR reaction. All isolates were examined for the presence of the stx1, stx2, sta, f5, f41 and eae genes by multiplex PCR as described by Franck et al. [12]. The genes for F17 fimbriae and enterotoxin (LT) were screened as described by Vu-Khac et al. [35]. The STEC strains were further tested for the presence of ehxA [31] and saa [25] genes by PCR. To distinguish the Stx2, Stx2c, and Stx2d genes, the restriction fragment length polymorphism-PCR method described by Piérard et al. [28] was used. The detection of stx2g was conducted as described by Leung et al. [15]. The DNA of isolates positive for stx1 was amplified using the Stx1cF and Stx1cR primers, which are specific for the stx1c subtype [39].
Reference strains, O antiserum, and O serogroup determination
The E. coli strains used as positive controls were E329 (f5), E322 (f17), E320 (f41), E281 (lt), E256 (sta, stb), P80 (eae), E389 (stx1), E391 (stx2), BKH-4.1 (saa/ehxA/stx1c/stx2c), CT55 (stx2g), and BHK (stx2d). O antiserum and antisera for the H7 flagella were purchased from Denka-Seiken (Japan). O serogroup determination of the E. coli isolates was conducted using standard slide agglutination techniques according to the manufacturer's instructions. The flagella antigen H7 was determined in strains belonging to serogroup O157.
Results
Prevalence of the genes of fimbriae, toxins and intimin in E. coli isolated from calves with diarrhea
A total of 345 E. coli isolates were analyzed by PCR and the results are summarized in Table 2. Overall, 108 of the 345 isolates (31.3%) carried genes of one of the fimbriae tested (F5, F17, and F41). Of the isolates that had fimbriae genes, 50 also carried genes for STx, one had both Stx and enterotoxin genes and 57 did not have any toxin genes. More than half of the isolates (177/345) carried genes for Stx and 23 also possessed the eae gene. Amongst isolates that carried the stx genes, 46 were stx1, 73 were stx2 and 58 were stx1/stx2.
Characterization of virulence genes
The STEC strains were further tested for two additional virulence factors, enterohemolysin (EHEC-hly) and autoagglutinating adhesion (Saa). The ehxA and saa genes were detected in 51% and 49% of the STEC isolates, respectively. The majority of the strains (73/87) carrying the saa gene were also positive for ehxA. Further characterization of the 104 isolates carrying the stx1 gene showed that 58 (55.7%) strains were positive for the Stx1c variant and 46 strains (44.2%) were classified as Stx1. Of the 131 strains carrying the genes for Stx2 variants, 48 (36.6%) were stx2, 48 (36.6%) were stx2c, 11 (8%) were stx2d, 17 (12.9%) were stx2g, and seven (5.3%) contained both stx2c and stx2g.
O serogroups
The isolates were tested against 33 different O serogroups commonly associated with pathogenic bovine E. coli and 89 strains (25.8%) belonged to one of these groups. The majority of strains (60) belonged to only four O serotypes, which included O15 (21 strains), O20 (9 strains), O103 (20 strains) and O157 (10 strains) (Table 3). Two of the ten O157 strains belonged to serotype O157:H7 and one carried genes for the Shiga toxin and intimin.
Discussion
This study is the first report of the prevalence of virulence factors in E. coli isolated from calves with diarrhea in Vietnam. It is well known that among the pathotypes of E. coli that cause diarrhea in calves, the ETEC strains are the most important agents, and strains expressing F5 and/or F41 and producing STa toxin are the most common [18]. F5 and/or F41 play a role in the colonization of bovine small intestine epithelial cells by ETEC and are primarily isolated from 1- to 7-day-old diarrheic calves [18]. In the present study, 53 of 345 isolates carried the f5 (36 isolates) and f41 (17 isolates) genes, and all of the F5-positive strains isolated from diarrheic calves were less than 7-day old. The susceptibility of the calves to ETEC in this age period is in agreement with previous reports [21,38]. Interestingly, in this study, we found that none of the F5-positive strains encoded enterotoxin genes (STa) when analyzed by PCR and they were also negative when tested with baby mice (data not show). However, more than half of the strains contained genes for Shiga toxin. While the F5-positive strains do not have enterotoxin genes, they do contain genes for Shiga toxin, which would indicate the emergence of a new phenotype causing diarrhea in calves. The observation that a high number of E. coli strains carried genes for fimbriae but did not have toxin genes has been reported before [17]. The F17 fimbriae, which are formally known as FY or Att25, are prevalent in E. coli strains isolated from calves with diarrhea or septicemia. In the present study, we found that 55 (16%) of 345 E. coli isolates from diseased calves carried genes for F17 fimbriae. Our results are in agreement with those of previous studies [24,33,34], which showed that F17 fimbriae are commonly found among E. coli strains isolated from 12% to 19% of diarrheic calves. Surprisingly, 50 (46%) of the 108 strains carrying fimbrial genes also possessed genes for Shiga toxins. The high prevalence of the Shiga toxin genes in these strains is particularly striking and has not been previously reported.
STEC has been implicated as an etiological factor of calf diarrhea [9,23,30,37], and these animals form a principle reservoir of STEC that is pathogenic for humans [1,6]. In the present study, we found that a total of 177 isolates (51%) were positive for the stx genes. Similarly, high percentages (40% or more) of stx gene positive E. coli strains have been reported in Brazil [29] and India [2]. However, other authors [23,24] have reported a lower rate (less than 10%) of STEC in diarrheic calves. STEC strains belonging to serogroups O15, O20, O103, and O157 have previously been found to be associated with diarrhea and enteritis in calves in Belgium [16], Spain [22], and the United States [10]. The majority of the STEC strains isolated in this study did not belong to the serogroups (O8, O9, O26, O111, O113, O126, O145) that have previously been found to be associated with diarrhea and enteritis in calves in other countries [16]. Interestingly, we found that four of the eight strains belonged to serotype O157:H7 and carried genes for both Shiga toxin and fimbriae. All of these O157:H7 strains were isolated from eight different diarrheic calves and fermented in sorbitol sugar (data not shown). The isolation of two strains belonging to serotype O157:H7 is of concern to human health [4,5]. Several studies [16,23,37] have demonstrated that most STEC from diarrheic calves only produce Stx1, whereas Stx2-positive strains are the dominant types in healthy calves [16,23,37]. This is in contrast to the results of the present study, wherein the stx2 gene (131 isolates) was observed as frequently as stx1 (104 isolates). Among the 23 STEC strains containing the eae genes, 11 isolates had only the stx1 gene, seven had the stx2 gene, and five had both genes.
It has previously been reported that the stx2c toxin genes were predominant among the stx2-positive E. coli strains isolated from diarrheic calves [24]. This is consistent with the results of the present study in which 55/131 stx2-positive strains were found to possess genes encoding Stx2c. Several studies have demonstrated that toxin types and variants could be important in the pathogenicity of STEC strains [27,28]. In humans, infection with isolates producing Stx2 toxins resulted in HUS more frequently than was observed for Stx1-producing strains [1,6,19]. Conversely, Stx2d-producing strains showed a low cytotoxicity toward Vero cells and were less frequently associated with diarrhea and HUS [27]. In the present study, stx2d was detected in 11 of 131 stx2-positive strains. The novel variant of bovine Stx2, which was designated Stx2g [15], was detected in 24/131 isolates in which seven isolates also possessed the genetic determinants for the Stx2c toxin variant. Previous studies have shown the toxin type 1c (Stx1c) to be closely associated with sheep [7,14], but not with cattle [7,40]. However, in this study, we found the stx1c genes in 44.2% of 104 stx1-positive strains, which suggests that Stx1c-producing strains are not restricted to sheep.
Recently, Paton et al. [26] described a novel megaplasmid-encoded adhesin, denoted Saa, in a LEE-negative O113: H21 STEC strain responsible for an outbreak of HUS. These authors also showed that the saa gene was associated with the ehxA gene [25,26]. In the present study, we found the saa gene in 87 STEC strains, and the majority of these strains also carried the ehxA gene. Enterohemolysin is widely distributed among STEC strains isolated from calves [3,13,24], although the role of enterohemolysin in causing diarrhea in calves has not been demonstrated. It has been suggested that this virulence factor may compliment the effects of Shiga toxin [3,19], and that it can be used as a diagnostic marker because its presence is strongly correlated with Shiga toxin [3,36]. The percentage of enterohemolysin among bovine STEC strains has been reported to range from 46.7 to 70% [3,13,24,36]. Our results showed that 51% of STEC isolates possessed the ehxA gene.
In conclusion, the results of the present study indicate that a high prevalence of strains carrying fimbrial genes also contained Shiga toxin genes. This suggests that stx genes, which are encoded on bacteriophages, are continuing to expand in the E. coli population. The high number of STEC strains isolated from diarrheic calves implies that these animals are an important reservoir of STEC strains that are potentially pathogenic toward farm animals and humans.
XML Download