Abstract
Escherichia coli strains designated as avian pathogenic E. coli (APEC) are responsible for avian colibacillosis, an acute and largely systemic disease that promotes significant economic losses in poultry industry worldwide because of mortality increase, medication costs, and condemnation of carcasses. APEC is a subgroup of extraintestinal pathogenic E. coli pathotype, which includes uropathogenic E. coli, neonatal meningitis E. coli, and septicemic E. coli. We isolated E. coli from commercial chicken carcasses in a Brazilian community and compared by polymerase chain reaction-defined phylogenetic group (A, B1, B2, or D) with APEC strains isolated from sick chickens from different poultry farms. A substantial number of strains assigned to phylogenetic E. coli reference collection group B2, which is known to harbor potent extraintestinal human and animal E. coli pathogens, were identified as APEC (26.0%) in both commercial chicken carcasses and retail poultry meat (retail poultry E. coli [RPEC]) (21.25%). The majority of RPEC were classified as group A (35%), whereas the majority of APEC were groups B1 (30.8) and A (27.6%). APEC and RPEC presented the genes pentaplex, iutA, hly, iron, ompT, and iss, but with different virulence profiles. The similarity between APEC and RPEC indicates RPEC as potentially pathogenic strains and supports a possible zoonotic risk for humans.
Introduction
Extraintestinal pathogenic E. coli (ExPEC) strains are characterized by the possession of virulence factors that enable their extraintestinal lifestyle and make them distinct from commensal and diarrheagenic E. coli strains (Russo and Johnson, 2000). Along with uropathogenic E. coli (UPEC), neonatal meningitis E. coli, or septicemic E. coli strains, APEC strains also fall under the category of ExPEC (Russo and Johnson, 2000).
APEC strains are very diverse, mostly because of their several virulence factors and serotypes, including adhesins, toxins, iron acquisition systems, colicin V plasmid, serum resistance proteins, F hemolysin, temperature-sensitive hemagglutinin, and capsule as well as lipopolysaccharide complexes (Dho-Moulin and Fairbrother, 1999). Many of these characteristics of virulence are found in human E. coli strains as UPEC, and genomic similarities between UPEC and APEC were evidenced, indicating that they are not host-specific strains (Mokady et al., 2005; Kariyawasam et al., 2007).
Several researchers have reported the possibility of transference of ExPEC from birds to humans, indicating that these contaminants may be epidemiologically linked to human disease (Johnson et al., 2003, 2005; Manges et al., 2007; Zhao et al., 2009).
Phylogenetic analysis of the genes chuA, a gene required for heme transport in enterohemorrhagic O157:H7 E. coli, and also yjaA and TSPE4 C2, genes with no known function, have shown that E. coli is composed of four main phylogenetic groups (A, B1, B2, and D) (Clermont et al., 2000). It has also shown that ExPEC from human sources belongs mainly to E. coli reference collection (EcoR) group B2 and occasionally to group D, whereas groups B1 and A are rarely found (Bingen-Bidois et al., 2002).
Recently, APEC isolates were clearly distinguished from avian fecal E. coli isolates by their possession of five genes carried by plasmids iutA, hlyF, iss, iroN, and ompT, suggesting that this pentaplex panel has diagnostic applications and underscores the close association between avian E. coli virulence and the possession of ColV plasmids (Johnson et al., 2008a).
To define the genetic relationship between a collection of E. coli strains isolated from colibacillosis (APEC) and commercial chicken carcasses in southern Brazil (subsequently referred as retail poultry E. coli [RPEC]), we compared RPEC and APEC using genotyping and phylogenetic typing.
Materials and Methods
Bacterial strains
Fifty-seven commercial refrigerated chicken carcasses from PR supermarkets (six different supermarkets and six different poultry wholesale companies) in Londrina were analyzed. Twenty-five grams of each carcass was taken and homogenized with 225 mL of saline added with 0.1% peptone. After homogenization, 1 mL was smeared into crystal violet red neutro bile agar by pour plate and incubated at 37°C for 24/48 h. The suspected colonies were confirmed as E. coli by biochemical tests such as EPM, MILi (Toledo et al., 1982a, 1982b), and Simons citrate agar (Difco).
APEC strains were isolated from colibacillosis lesions of chickens with clinical signs of respiratory tract infection and coming from different poultry farms (Delicato et al., 2003). RS218 and V27 (Johnson and Stell, 2000) were used as reference strains and E. coli K12 HB101 and 711F as negative control for all tests. All isolates were stored at −20°C in brain heart infusion broth to which 25% glycerol was added after growth.
EcoR grouping
E. coli strains were classified according to EcoR using the rapid phylogenetic grouping polymerase chain reaction (PCR) technique described by Clermont et al. (2000). The isolates were assigned to one of four groups (A, B1, B2, or D) based on their possession of two genes (chuA and yjaA) and a DNA fragment (TSPE4.C2), as determined by PCR. Boiled lysates from overnight cultures were used as the source of template DNA for this study.
Virulence genotyping
E. coli strains isolated from colibacillosis (APEC) and from commercial chicken carcasses were examined for the presence of the best five genes with respect to pathogenicity: iutA (aerobactin siderophore receptor gene), hlyF (putative avian hemolysin), iss (episomal increased serum survival gene), iroN (salmochelin siderophore receptor gene), and ompT (episomal outer membrane protease gene). The analysis was carried out using the PCR primer sets previously described (Johnson et al., 2008a).
Statistical analysis
Difference between two independent proportions and Friedman nonparametric tests (Sokal and Rohlf, 1994) for univariate analysis of significance were used to investigate virulence factors and associations between APEC and RPEC. Differences were considered statistically significant if p ≤ 0.05 for all tests.
Results
One to three colonies were selected out of the isolates obtained from 40 of the 57 carcasses analyzed. A total of 80 E. coli strains were tested for phylogenetic groups and submitted to PCR pentaplex test. One hundred eighty-five APEC strains (Delicato et al., 2003) were tested for phylogenetic groups and PCR pentaplex test.
The distribution of APEC and isolates from carcasses or retail poultry meat (RPEC) among the four phylogenetic groups is shown in Table 1. The analyzed isolates were assigned to all four phylogenetic groups. Of 185 APEC strains, 51 (27.6%) were assigned to group A, 57 (30.8%) to group B1, 48 (26.0%) to group B2, and 29 (15.6%) to group D. Of 80 RPEC, 28 (35.0%) were assigned to group A, 12 (15.0%) to group B1, 17 (21.25%) to group B2, and 23 (28.75%) to group D. The majority of the RPEC were assigned to group A (35.0%), whereas the majority of APEC strains belonged to group B1 (30.8%). Significantly fewer isolates of RPEC (p < 0.05; 15.0%) were assigned to phylogenetic group B1. The percentage of strains assigned to B2 was similar between APEC and RPEC (26.0% and 21.25%, respectively). Significantly fewer isolates of APEC (p < 0.05; 15.6%) were assigned to phylogenetic group D (Table 1).
Difference between two independent proportions: p-value of ≤0.05 for all tests.
NS, differences not significant.
E. coli contaminating poultry carcasses were assessed for their genetic similarities to APEC. Table 2 summarizes the presence of the genes iutA, hlyF, episomal iss, iroN, and episomal ompT for the two groups of isolates, using a subset of genes of pentaplex PCR. Among the APEC, the five aforementioned genes showed high prevalence, but among RPEC only the gene iutA had a rate of appearance above 50%.
Values show the number of isolates; values in parentheses show the proportion of isolates containing each gene as a proportion of the total number of isolates.
Difference between two independent proportions: p-value of ≤0.05 for all tests.
APEC, avian pathogenic E. coli; RPEC, retail poultry E. coli.
The test for differences between two proportions demonstrated that the frequency of the gene iut A was not significantly different (p = 0.9179) for the two types of strains, although the presence of hlyF, iss, iroN, and ompT was significantly higher (p < 0.0001) in APEC than in RPEC (Table 2).
Moreover, APEC and RPEC phylogenetic groups (A, B1, B2, or D) were compared with the virulence genotype obtained by pentaplex (Table 3). An analysis of the distribution of the five more important virulence genes among APEC's four phylogenetic groups revealed the occurrence of these genes at a higher rate in groups B2 and D and at a lower rate in group A. Among RPEC, only the gene iutA occurred at a higher rate in groups B2 and D.
Group B2 differs from the others by the presence of 4–5 virulence factors and group A presents more cases of absence or presence of only one gene than the other phylogenetic groups. According to Table 3, Friedman test demonstrated a significant difference (p < 0.0001) for the presence of virulence factors in APEC's B2 group in comparison with all other phylogenetic groups assigned to RPEC (α = 5%). Among APEC strains, there was a statically significant difference between the groups A, B2, and D (p = 0.0037), considering that group B2 had a higher frequency of virulence factors than groups A and D (p < 0.05). No significant differences (p = 0.5280) were observed among virulence factors and the phylogenetic groups in RPEC strains.
Discussion
APEC isolates remain an important problem for poultry producers and a potential concern for public health professionals, because growing evidences suggest a possible role of APEC in human disease.
We classified outbreak strains isolated from sick chickens (APEC = 185) and nonoutbreak strains isolated from commercial chicken carcasses (RPEC = 80) and determined that all strains were distributed in the four main EcoR phylogenetic groups. Our data agree with other authors who found that APEC was associated with all four EcoR groups and that group A presented higher percentage among APEC strains, 38.0%, 36.0%, and 46.1% (Rodriguez-Siek et al., 2005; Ewers et al., 2007; Kariyawasam et al., 2007).
We found a higher percentage of groups A (35.0%) and D (28.75%) in RPEC. Our data agree with Johnson et al. (2003), who isolated E coli assigned to groups A (32%) and D (25%) in retail chicken products. Also, fecal and environmental strains were significantly associated with groups D (38.9%) and A (65.7%) (Ewers et al., 2009). Group A usually contains commensal strains, but group D contains pathogenic intestinal strains and ExPEC. We also found a substantial number of APEC (26.0%) and RPEC (21.25%) assigned to group B2; similarly high percentages of APEC assigned to group B2 were found in others works (Ewers et al., 2009; Zhao et al., 2009). Also, group B2 was found at a substantial percentage (23.2%) among fecal E. coli strains from clinically healthy chickens, which is known to include potent human and animal ExPEC strains (Ewers et al., 2009). B2 strains were the most virulent for mice, which were in agreement with the presence of a large number of the virulence-associated genes (Picard et al., 1999).
The pentaplex panel described by Johnson et al. (2008a) showed that the genes iutA, hlyF, episomal iss, iroN, and episomal ompT were useful to detect APEC-like strains occurring in poultry production, along the food chain, and in human disease. This panel may be helpful toward clarifying the potential roles of APEC in human disease, ascertaining the source of APEC in animal outbreaks, and identifying effective targets for avian colibacillosis control.
In this work, phylogenetic groups (A, B1, B2, or D) were compared with virulence genotype obtained by pentaplex of APEC and RPEC strains. Similar to the data obtained by Johnson et al. (2009), both types of strains contained five virulence-associated genes; although higher rates of these genes occurred in APEC. The majority of APEC strains had four or five virulence genes (65%), independent of phylogenetic group, confirming the importance of pentaplex for the diagnostic of APEC. APEC and RPEC strains of phylogenetic group A showed no significant differences for the presence of virulence genes. These strains presented lower number of these genes, in accordance with data from other authors who observed the presence of few virulence factors and suggested low virulence and the environmental character of these strains.
Our data also suggest the possibility of zoonotic risk because eight (10%) of RPEC strains presented the five virulence-associated genes of APEC and ExPEC, even though the majority of RPEC strains assigned to group B2 did not present the same virulence of APEC strains (Table 3). We demonstrate that phylogenetic group B2 of APEC had more incidence of possession of all five virulence-associated genes, being capable of surviving in multiple environments and hosting or initiating an infection, as it was suggested by other authors (Johnson et al., 2008b).
In conclusion, APEC and RPEC harbor ExPEC traits that are known to contribute to the onset of diseases in human beings. The presence of RPEC with characteristics of APEC increases the group of strains with ExPEC traits and holds important zoonotic risks.
Footnotes
Disclosure Statement
No competing financial interests exist.