Molecular Characterization of Enterohemorrhagic E. coli O157 Isolated from Animal Fecal and Food Samples in Eastern China

Objective. To elucidate the extent of food contamination by enterohemorrhagic Escherichia coli (EHEC) O157 in Eastern China. Methods. A total of 1100 food and animal fecal samples were screened for EHEC O157. Then, molecular characterization of each isolate was determined. Results. EHEC O157 was isolated as follows: pig feces, 4% (20/500); cattle feces, 3.3% (2/60); chicken feces, 1.43% (2/140); pork, 2.14% (3/140), milk, 1.67% (1/60); and chicken meat, 1.67% (1/60). The stx1, stx2, eae, and hlyA genes were present in 26.7% (8/30), 40% (12/30), 63.3% (19/30), and 50% (15/30) of the O157 isolates, respectively. Molecular typing showed that strains from fecal and food samples were clustered into the same molecular typing group. Furthermore, the isolates from pork and pig feces possessed the same characterization as the clinical strains ATCC35150 and ATCC43889. Biofilm formation assays showed that 53.3% of the EHEC O157 isolates could produce biofilm. However, composite analyses showed that biofilm formation of EHEC O157 was independent of genetic background. Conclusions. Animal feces, especially from pigs, serve as reservoirs for food contamination by EHEC O157. Thus, it is important to control contamination by EHEC O157 on farms and in abattoirs to reduce the incidence of foodborne infections in humans.


Introduction
Enterohemorrhagic Escherichia coli (EHEC) is an important emerging zoonotic foodborne pathogen that can cause watery and/or bloody diarrhea, hemorrhagic colitis (HC), hemolytic-uremic syndrome (HUS), and thrombotic thrombocytopenic purpura (TTP). In humans, EHEC O157 is recognized as a major etiological agent of these diseases [1][2][3]. Shiga toxin, intimin, and enterohemolysin are key virulence factors for the pathogenesis of O157 and other EHEC strains, including the 2011 O104:H4 outbreak in Germany [1,4,5]. Biofilm is composed of surface-bound microbes enclosed in an amorphous extracellular matrix [6], which is often an assembly of exopolysaccharides, proteins, and nucleic acids [7]. Residence in a biofilm community offers certain advantages to bacteria, such as enhanced resistance to the environmental stresses.
Although the prevalence of EHEC O157 in animal feces in Eastern China has been reported [11], comparative analysis of EHEC O157 in animal feces and food samples remains limited. Thus, the aim of this study was to detect and characterize EHEC O157 isolates from animal fecal and food samples in Eastern China to elucidate the extent of food contamination by animal feces.

Bacterial Strains and Growth
Conditions. The EHEC O157:H7 reference strains ATCC35150 and ATCC43889 were used as positive controls for amplification by polymerase 2 The Scientific World Journal

Isolation and Identification of EHEC O157.
In the present study, a total of 1100 samples, including animal fecal samples ( = 800) and food samples ( = 300), which were collected once per month from July 2010 to July 2012 in Eastern China (Shanghai, Jiangsu, and Zhejiang), were used for analysis. Isolation and identification of EHEC O157 were performed according to a procedure using enrichment methods and confirmed by PCR. Briefly, preliminary identification of O157 strains was obtained through phenotype characterization using sorbitol-MacConkey agar. Isolates were confirmed by PCR for the O157 lipopolysaccharide O-antigen synthesis gene rfbE (Table 1) and serotyping was performed using an agglutination assay with anti-O157 sera.

Virulence Genes
Detection. The EHEC O157 virulence genes stx1, stx2, hlyA, and eae were assessed by PCR. Descriptions of the targeted genes and primer sequences are listed in Table 1.

Quantification of Biofilm
Formation. Biofilm formation assays were performed according to the methods described in a previous report [21] with some modifications. Briefly, strains were grown to the stationary phase in LB broth at 37 ∘ C and then diluted to 1 : 100 in LB broth without salt (LB-NS). A 200 L aliquot of each dilution was dispensed into individual wells of a microtiter plate. Wells with sterile LB medium served as blank controls. After incubation for 24 h at 37 ∘ C, the cultured medium was removed and the wells were washed twice with sterile phosphate-buffered saline (PBS) to remove planktonic bacteria. The biofilms grown in the microtiter plate were stained with crystal violet (1%, w/v) at room temperature for 1 h. After removing the crystal violet solution, the wells were washed five times with PBS to remove unbound dye. Then, the plates were air dried and filled with 200 L of 95% ethanol to solubilize the dye from the cells. Finally, the optical density (OD) of the stained adherent bacteria was determined at 595 nm using a microplate reader (BioTek Instruments, Inc., Winooski, VT, USA). All tests were performed three times and the results were averaged. Strains were classified as non-biofilm producers, weak biofilm producers, moderate biofilm producers, or strong biofilm producers based on OD 595 measurements of bacterial biofilms in accordance with previously described criteria [22]. Briefly, OD values produced by bacterial biofilms were compared to a cut-off OD value (ODc, three standard deviations above the mean OD of a blank test) and the strains were classified as follows: no biofilm production (OD < ODc), weak biofilm producer (ODc < OD < 2 ODc), moderate biofilm producer (2 ODc < OD < 4 ODc), and strong biofilm producer (4 ODc < OD).

Phylogenetic
Typing. E. coli is classified to four main phylogenetic groups (A, B1, B2, and D) according to the combination of two genes (chuA and yjaA) and an anonymous DNA fragment (TSPE4.C2) as described previously [22]. Thus, EHEC O157 isolates were assigned to phylogenetic groups by triplex-PCR amplification of chuA and yjaA, and the DNA fragment TSPE4.C2.

Multilocus Sequence Typing (MLST).
All EHEC O157 strains were assigned to multilocus sequence types (STs) as described previously [23] for the amplification of seven house-keeping genes (adk, fumC, gyrB, icd, mdh, purA, and recA). Briefly, the housekeeping genes of each strain were amplified, sequenced, and aligned against allele templates of E. coli retrieved from an online database (http://www.mlst .net/).

Distribution of Virulence Genes in EHEC O157
. EHEC O157 isolates were screened for the presence of virulence genes. The distribution and combinations of virulence genes for each isolate are shown in Table 3. Ten virulence gene genotypes (G1-G10) were identified. The stx1 and stx2 genes were detected in 8 (26.7%) and 12 (40%) isolates, respectively. The eae and hlyA genes were present in 63.3% (19/30) and 50% (15/30) of the O157 isolates, respectively. Moreover, 7 (23.3%) isolates harbored the eae gene alone, which was the predominant genotype (G1). Five (16.7%) strains harbored all four virulence genes simultaneously (G3), which were all isolated from pig feces and pork samples. Generally, the prevalence of these genes in pig feces and pork was higher than in other samples.

Discussion
EHEC O157 is an important emerging zoonotic foodborne pathogen that can cause gastroenteritis often complicated by HC or HUS in humans [1][2][3]. Food contaminated with EHEC O157 is a principal source of infection in humans [8,9]. Thus, the aim of this study was to isolate and characterize EHEC O157 strains from animal feces and food samples to elucidate the extent of food contaminated by animal feces in Eastern China.
In the present study, EHEC O157 was successfully isolated from 2.73% (30/1100) of the tested animal feces and food samples. Among the 30 EHEC O157 isolates, 20 (66.7%) strains were isolated from pig feces, 3 (10%) from pork, 2 (6.67%) from cattle feces, 2 (6.67%) from chicken feces, and one each from duck feces, milk, and chicken meat, respectively. Reportedly, the proportion of EHEC O157 is often higher in pig feces and pork than in cattle feces and beef [24]. Although the number of EHEC O157 isolates in pig feces or pork was higher than in cattle feces or milk, the prevalence was similar in our investigation. Two (3.3%) of the 60 cattle fecal samples were positive for EHEC O157, but none of the beef samples were. One possible explanation for this phenomenon might be that the cattle fecal samples were collected from dairy cows used for milk production. The prevalence of EHEC O157 is dependent on various factors, including the hygienic conditions of the farms and abattoirs. Also, surrounding environments can impact the distribution of EHEC O157 in animal feces, food products, and water 4 The Scientific World Journal   Figure 1: Biofilm formation by EHEC O157 isolates. Biofilm formation by EHEC O157 isolates on polypropylene microtiter plates following incubation for 24 h at 37 ∘ C in LB broth without salt (LB-NS). All experiments were repeated at least three times. The columns represent the mean ± standard deviations of the data. Comparisons of the OD values produced by bacterial biofilms to the ODc values were used to classify the strains (non-biofilm producer, weak, moderate, or strong). The cut-off OD (ODc) value was defined as three standard deviations above the mean OD of a blank control. Strains were classified as follows: OD < ODc = no biofilm production; ODc < OD < (2 ODc) = weak biofilm producer; (2 ODc) < OD < (4 ODc) = moderate biofilm producer; and (4 ODc) < OD = strong biofilm producer.
sources worldwide [8,10,11,[25][26][27]. Similarly, the prevalence of EHEC O157 from different sources in different areas of China can also extensively vary [11,[28][29][30]. Compared to the findings of previous studies [29], the prevalence of EHEC O157 in pig and cattle feces were lower in Shanghai, which was likely due to improved hygienic conditions of farms in recent years. Moreover, our results showed that the proportion of EHEC O157 was higher in feces than in food and water samples, which was in accordance with the results of previous studies [10,25,26,29,31]. The pathogenesis of EHEC O157 is associated with several virulence factors, such as Shiga toxin 1 and 2 (encoded by stx1 and stx2, resp.), intimin (encoded by eae), and enterohemolysin (encoded by hlyA) [1,4]. Shiga toxins inhibit protein synthesis resulting in the death of host cells and are key virulence factors of other Shiga toxin-producing E. coli infections, including the recent O104:H4 outbreak in Germany in 2011 [5]. Intimin is a type III secretion system effector protein that facilitates the intimate adherence of E. coli O157 cells to the intestinal epithelium. Enterohemolysin has been implicated as a causative agent of enterocyte effacement [1,4]. Thus, the distribution of virulence genes in EHEC O157 isolates was screened by PCR analysis. Our results showed that more of the isolates harbored the stx2 gene than stx1 (40% versus 26.7%, resp.). A large proportion (63.3%) of the O157 isolates was positive for the eae gene, and 15 (50%) isolates harbored the hlyA gene. However, different distributions of these four virulence genes in EHECO157 were observed by other researchers [25,32], which might be due to the different sources and pathogenesis of E. coli collections. Composite analysis showed that these virulence genes were mostly prevalent in isolates from pig feces and pork than other samples. Similar results were also reported previously [31]. Among the genotypes, 5 (16.7%) isolates from pig feces and pork possessed a combination of all four virulence genes found in clinic strains ATCC35150 and ATCC43889.
Our findings indicated that the use of molecular typing provides valuable information that may also be useful in pinpointing sources of food contamination [14,16,33]. In the present study, three different methods, including phylogenetic typing, MLST, and RAPD fingerprinting, were used to analyze the relatedness of EHEC O157 strains isolated from fecal and food samples. Of the individual typing techniques used in this study, MLST and RAPD fingerprinting showed higher levels of discriminating power than phylogenetic typing. Moreover, the results of the MLST and RAPD fingerprinting techniques were similar. Results of phylogenetic typing showed that most (70%) of the O157 strains were classified as group D. Parallel findings, in which EHEC O157:H7 strains were almost exclusively group D, have been reported elsewhere [17,18]. In this study, seven selected housekeeping genes were chosen to assess potential sequence diversity by MLST typing according to a protocol found at http://www.mlst.net/. The results showed that O157 isolates predominantly belonged to types ST11, ST88, ST641, and ST117. However, different ST types were obtained in 6 The Scientific World Journal other studies, which were due to the selection of different housekeeping and virulence genes [3,[14][15][16].
Composite analysis of all three molecular typing methods revealed a relationship between the phylogenetic group and MLST and RAPD typing. For example, the largest cluster of MLST type (ST11) was comprised mainly of strains belonging to ECOR group D and RAPD type R1. In fact, all of the R1 type isolates were identified as ST11 and ECOR group D. Molecular characterization of EHEC O157 showed that the isolates from fecal and food samples were clustered into the same predominant group, indicating that animal feces might be a reservoir for EHEC O157. Thus, close monitoring of possible food contamination by EHEC O157 should be reinforced.
Biofilm formation aids in the survival of bacteria and enhances their ability to survival in environment. Thus, the ability of the EHEC O157 isolates to form biofilms was assessed in this study. The results indicated that biofilm formation occurred in 16 (53.3%) of the EHEC O157 strains after 24 h of incubation at 37 ∘ C. However, the composite analysis between molecular typing and biofilm formation revealed that biofilm formation of EHEC O157 was independent of the three molecular genetic typing methods used in this study, which was consistent with the conclusions of another study [17].

Conclusions
EHEC O157 was isolated from fecal and food samples and then characterized. The results showed that isolates from fecal and food samples harbored the same gene combinations. Moreover, these isolates were clustered into the same molecular typing group, indicating that animal feces are reservoirs of EHEC O157. Thus, it is important to control food contamination with EHEC O157 on farms and in abattoirs to reduce the incidence of foodborne infections in humans.

Conflict of Interests
None of the authors had any conflict of interests in the writing of this paper.