Abstract

Escherichia coli is a commensal bacterium of the bird's intestinal tract, but it can invade different tissues resulting in systemic symptoms (colibacillosis). This disease occurs only when the E. coli infecting strain presents virulence factors (encoded by specific genes) that enable the adhesion and proliferation in the host organism. Thus, it is important to differentiate pathogenic (APEC, avian pathogenic E. coli) and non-pathogenic or fecal (AFEC, avian fecal E. coli) isolates. Previous studies analyzed the occurrence of virulence factors in E. coli strains isolated from birds with colibacillosis, demonstrating a high frequency of the bacterial genes cvaC, iroN, iss, iutA, sitA, tsh, fyuA, irp-2, ompT and hlyF in pathogenic strains. The aim of the present study was to evaluate the occurrence and frequency of these virulence genes in E. coli isolated from poultry flocks in Brazil. A total of 138 isolates of E. coli was obtained from samples of different tissues and/or organs (spleen, liver, kidney, trachea, lungs, skin, ovary, oviduct, intestine, cloaca) and environmental swabs collected from chicken and turkey flocks suspected to have colibacillosis in farms from the main Brazilian producing regions. Total DNA was extracted and the 10 virulence genes were detected by traditional and/or real-time PCR. At least 11 samples of each gene were sequenced and compared to reference strains. All 10 virulence factors were detected in Brazilian E. coli isolates, with frequencies ranging from 39.9% (irp-2) to 68.8% (hlyF and sitA). Moreover, a high nucleotide similarity (over 99%) was observed between gene sequences of Brazilian isolates and reference strains. Seventy-nine isolates were defined as pathogenic (APEC) and 59 as fecal (AFEC) based on previously described criteria. In conclusion, the main virulence genes of the reference E. coli strains are also present in isolates associated with colibacillosis in Brazil. The analysis of this set of virulence factors can be used to differentiate between APEC and AFEC isolates in Brazil.

INTRODUCTION

Modern poultry production practices are intensive and provide conditions for the emergence and spread of infectious diseases. Colibacillosis is a localized or systemic infection caused by the bacteria Escherichia coli and frequently reported in the main poultry-producing regions of the world. It occurs typically in young birds after infection by other pathogens or in stress situations caused by poor diet and/or environmental conditions (Ewers et al., 2003; Kabir, 2010). The incubation period is usually short (approximately 1 to 3 d) and clinical manifestations include enteritis, arthritis, omphalitis, coligranuloma, salpingitis, septicemia, airsacculitis, chronic respiratory disease (CRD), swollen head syndrome, peritonitis, osteomyelitis, synovitis, cellulitis, and others (Barnes et al., 2008; Matter et al., 2011; Barbieri et al., 2013).

Colibacillosis occurs only in birds infected by pathogenic strains of E. coli known as APEC (avian pathogenic E. coli). Most APEC isolated from poultry are specific clonal types adapted to life as pathogens that infect only birds and represent a low risk of disease for people or other animals. However, these microorganisms share common characteristics with other strains of E. coli associated with diseases outside the intestinal tracts of any species (generally known as extra-intestinal pathogenic E. coli, ExPEC). On contrast, non-pathogenic strains (known as AFEC, avian fecal E. coli) are always present in the enteric tract, including birds without colibacillosis (Johnson et al., 2008). Increasing evidence has shown that pathogenic strains differ from the fecal ones by having a peculiar set of genes clustered into chromosomal or, more often, plasmid-located pathogenicity islands (Johnson et al., 2008; Tivendale et al., 2009; Mellata et al., 2010). These pathogenicity genes are called virulence factors and are usual in the E. coli strains associated with colibacillosis (Knöbl et al., 2008).

A previous study has already identified some bacterial genes that code to virulence factors in APEC strains. First, it was demonstrated that 9 (cvaC, iroN, iss, iutA, sitA, tsh, fyuA, irp-2 and ompT) of a total of 38 studied genes occur more frequently in APEC than in AFEC isolates (Rodriguez-Siek et al., 2005). The authors also proposed genetic criteria for the classification of E. coli strains: isolates with a minimum of 5 virulence factors should be classified as pathogenic (APEC), while those with fewer than 5 as fecal (AFEC). In a later study, 46 genes were investigated in a more complete panel of samples, and the results showed that only 5 genes (iutA, iss, ompT, iroN and hlyF) were more associated with pathogenesis and could be used to differentiate pathogenic and fecal E. coli strains (Johnson et al., 2008).

Avian colibacillosis diagnosis is based on the isolation of E. coli in birds with clinical suspicion of this disease in Brazil. The main limitation of the isolation method is the presence of fecal strains in all birds, resulting in the isolation and characterization of non-pathogenic E. coli in several situations. The objective of this study was to analyze the occurrence and genetic sequences of the 10 most-studied APEC virulence factors in E. coli isolates from Brazilian poultry farms. Additionally, the genetic criteria proposed in previous studies were used to differentiate AFEC and APEC isolates.

MATERIALS AND METHODS

Samples

A total of 138 isolates of E. coli, 74 from chickens (40 breeders and 34 broilers) and 64 from turkeys, were obtained from poultry flocks of different Brazilian producing regions: Northeast (Bahia state), Midwest (Goiás and Mato Grosso do Sul states), Southeast (São Paulo and Minas Gerais states) and South (Paraná and Santa Catarina states) (Figure 1). All flocks were suspected to have colibacillosis, most of them with respiratory signs, secondary infections, and high mortality. Isolates were obtained after culture from different tissue/organ samples (spleen, liver, kidney, trachea, lungs, skin, ovary, oviduct, intestine, cloaca) and environmental swabs collected between April 2010 and April 2011.

Figure 1.

Geographical origin and host species (chicken or turkey) of the 138 Escherichia coli isolates.

Figure 1.

Geographical origin and host species (chicken or turkey) of the 138 Escherichia coli isolates.

Bacteriological Procedures

All samples were seeded in blood agar and McConckey (MC) media and incubated for 24 hours at 36°C. The isolates were considered presumptive positive for E. coli with the occurrence of colonies in the middle of the blood agar and digestion of lactose in the MC. This result was confirmed by Gram stain and 3 biochemical tests: triple sugar iron (TSI), sulfide indole motility (SIM), and methyl red (MR). The isolates were then maintained on nutrient agar medium.

DNA Extraction

DNA extraction was performed using the NewGene commercial kit (Simbios Biotecnologia, Cachoeirinha, RS, Brazil). Briefly, 100 μL of each isolate was mixed with 400 μL of lysis solution (5 M guanidine thiocyanate, 0.1 M Tris-HCl [pH 6.4]). Each tube was incubated at 60°C for 10 min. After centrifugation (8,609 × g, 1 min), the supernatant was transferred to a tube containing 20 μL of a silica suspension. After agitation and centrifugation (8,609 × g, 1 min), the pellet was washed twice with 150 μL of solution A (5 M guanidine thiocyanate, 0.1 M Tris-HCl [pH 6.4]) twice with solution B (80% ethanol) and once with solution C (96% ethanol). The tubes were heated at 60°C for 10 min for drying; afterward, 50 μL of elution solution (10 mM Tris-HCl [pH 8.0], 1 mM EDTA) was mixed with the pellet.

PCR Amplification and Analysis

Ten virulence genes (cvaC, iroN, iss, iutA, sitA, tsh, fyuA, irp-2, ompT and hlyF) were detected by PCR and real-time PCR with procedures previously described (Rodriguez-Siek et al., 2005; Johnson et al., 2008; Ikuta et al., 2014). PCR assays were carried out in the thermocycler Veriti (Applied Biosystems by Life Technologies Co., Foster City, CA). PCR results were directly evaluated in polyacrylamide gel (10%) stained with silver nitrate in comparison to controls and commercial molecular weight markers. TaqMan real-time PCR assays were performed on StepOne Plus® (Applied Biosystems). Cycle threshold (Ct) for each sample was calculated in comparison to positive and negative controls.

Sequencing

Sequencing reactions were performed with purified amplification products of the 10 virulence genes. Between 5 and 20 ng of DNA were mixed with BigDye Terminator reagent (Applied Biosystems) and submitted to 2 independent PCR (with forward and reverse primers) with the following conditions: one cycle of 3 min at 95°C followed by 40 cycles of 10 s at 95°C, and 4 min at 60°C. The results were analyzed using the ABI 3130xl Genetic Analyzer (Applied Biosystems). The 2 sequences of nucleotides in each sample were pooled in a single strand of DNA (consensus) using the SeqMan program. Consensus sequences of each gene were aligned with reference sequences for comparative genetic similarity analysis using MegAling Program (Lasergene, DNAStar, Madison, WI). The main reference sequences were genes from the plasmids of the strain APEC-O2-CoLV (Skyberg et al., 2006) and from the chromosome of the strain APEC-O1 (Johnson et al., 2007). The nucleotide sequence data reported in this paper were submitted to the GenBank nucleotide sequence database and have been assigned with the accession numbers KP657523 to KP657558.

Statistical Analysis

Virulence factors frequencies were compared between isolates classified in APEC and AFEC using the chi-square (X2) in Winpepi version 11.39 (Abramson, 2011). Differences were considered significant when P < 0.05. Z test was also performed to measure the difference between the proportions for each target (Snedecor and Cochran, 1980).

RESULTS

Detection and Sequences of Virulence Genes

The 10 virulence genes were detected in the Brazilian E. coli isolates. Genes hlyF, sitA (both in 95 isolates, 68.8%) and ompT (in 94, 68.1%) presented the highest frequency, while fyuA (60, 43.5%) cvaC (58, 42%) and irp-2 (55, 39.9%) were the least frequent ones. All the virulence genes were also detected in the 2 species of birds (chickens and turkeys) and sitA (48, 64.9%) was the most frequent in chickens, while ompT and hlyF (64, 81.3%) were the most prevalent in turkeys (Table 1). In the analysis of the geographical distribution, it was observed that all 10 virulence genes were detected in isolates from the 4 geographic regions analyzed. Gene hlyF was the most common in the South (30, 78.9%), Midwest (13, 92, 9%) and Northeast (25, 73.5%), while sitA was the most frequent in the Southeast (37, 71.2%). Uncommon genes were irp-2 in Northeast (14, 41.2%) and Southeast (11, 21.2%), sitA and tsh (10, 71.4%) in Midwest and cvaC (8, 21.1%) in the South (Table 2).

Table 1

Virulence gene frequency according to the avian host species (chicken and turkeys) and identity among nucleotide (NT) and amino acid (AA) sequences of the Brazilian isolates and reference strains.

Genes Total (n = 138) Chickens (n = 74) Turkeys (n = 64) Identity NT Identity AA 
 n(%) n(%) n(%) n(%) n(%) 
hlyF 95(68.8) 43(58.1) 52(81.3) 99.5(99.0–100) 99.4(98.5–100) 
sitA 95(68.8) 48(64.9) 47(73.4) 99.2(98.0–100) 99.4(98.8–100) 
ompT 94(68.1) 42(56.8) 52(81.3) 99.9(99.8–100) 100 
iroN 83(60.1) 39(52.7) 44(68.8) 99.8(99.6–100) 99.4(98.8–100) 
iss 79(57.2) 38(51.4) 41(64.1) 99.9(99.7–100) 100 
tsh 73(52.9) 23(31.1) 50(78.1) 99.6(99.2–100) 99.3(98.5–100) 
iutA 72(52.2) 31(41.9) 41(64.1) 99.3(98.6–100) 100 
fyuA 60(43.5) 31(41.9) 29(45.3) 99.9(99.7–100) 99.9(99.6–100) 
cvaC 58(42.0) 36(48.6) 22(34.4) 100 100 
irp-2 55(39.9) 27(36.5) 28(43.8) 99.8(99.6–100) 100 
Genes Total (n = 138) Chickens (n = 74) Turkeys (n = 64) Identity NT Identity AA 
 n(%) n(%) n(%) n(%) n(%) 
hlyF 95(68.8) 43(58.1) 52(81.3) 99.5(99.0–100) 99.4(98.5–100) 
sitA 95(68.8) 48(64.9) 47(73.4) 99.2(98.0–100) 99.4(98.8–100) 
ompT 94(68.1) 42(56.8) 52(81.3) 99.9(99.8–100) 100 
iroN 83(60.1) 39(52.7) 44(68.8) 99.8(99.6–100) 99.4(98.8–100) 
iss 79(57.2) 38(51.4) 41(64.1) 99.9(99.7–100) 100 
tsh 73(52.9) 23(31.1) 50(78.1) 99.6(99.2–100) 99.3(98.5–100) 
iutA 72(52.2) 31(41.9) 41(64.1) 99.3(98.6–100) 100 
fyuA 60(43.5) 31(41.9) 29(45.3) 99.9(99.7–100) 99.9(99.6–100) 
cvaC 58(42.0) 36(48.6) 22(34.4) 100 100 
irp-2 55(39.9) 27(36.5) 28(43.8) 99.8(99.6–100) 100 
Table 2.

Virulence gene frequency according to the Brazilian geographic regions (Northeast, Midwest, South and Southeast) of the Escherichia coli isolates.

Genes Northeast Midwest Southeast South 
 n = 34 n = 14 n = 52 n = 38 
 n(%) n(%) n(%) n(%) 
hly25(73.5) 13(92.9) 27(51.9) 30(78.9) 
iro21(61.8) 12(85.7) 25(48.1) 25(65.8) 
omp25(73.5) 12(85.7) 27(51.9) 30(78.9) 
Iss 20(58.8) 11(78.6) 24(46.2) 24(63.2) 
iut21(61.8) 11(78.6) 17(32.7) 23(60,5) 
cva19(55.9) 12(85.7) 18(34.6) 8(21.1) 
sit24(70.6) 10(71.4) 37(71.2) 24(63.2) 
Tsh 16(47.1) 10(71.4) 12(23.1) 35(92.1) 
fyu17(50.0) 12(85.7) 13(25.0) 18(47.4) 
irp-14(41.2) 12(85.7) 11(21.2) 18(47.4) 
Genes Northeast Midwest Southeast South 
 n = 34 n = 14 n = 52 n = 38 
 n(%) n(%) n(%) n(%) 
hly25(73.5) 13(92.9) 27(51.9) 30(78.9) 
iro21(61.8) 12(85.7) 25(48.1) 25(65.8) 
omp25(73.5) 12(85.7) 27(51.9) 30(78.9) 
Iss 20(58.8) 11(78.6) 24(46.2) 24(63.2) 
iut21(61.8) 11(78.6) 17(32.7) 23(60,5) 
cva19(55.9) 12(85.7) 18(34.6) 8(21.1) 
sit24(70.6) 10(71.4) 37(71.2) 24(63.2) 
Tsh 16(47.1) 10(71.4) 12(23.1) 35(92.1) 
fyu17(50.0) 12(85.7) 13(25.0) 18(47.4) 
irp-14(41.2) 12(85.7) 11(21.2) 18(47.4) 

Eleven to 21 amplified products of each gene, obtained from different isolates, were sequenced and compared with the reference plasmid APEC-O2-CoLV and chromosome APEC-O1. The results showed a high degree of identity (over 99%) among all samples of each gene and the respective coding amino acid sequences (Table 1). Most nucleotide mutations were synonymous and did not affect the amino acid sequence of the respective proteins. Non-synonymous mutations were observed in 5 genes of a total of 18 isolates, with only 1 amino acid modification in 11 sequences of 3 genes (hlyF, sitA and fyuA) and 2 amino acid changes in the remaining 7 sequences of 4 genes (tsh, hlyF, iroN and sitA).

Detection of APEC and AFEC

All Brazilian E. coli isolates were classified in APEC or AFEC according to the analysis of 9 genes (Rodriguez-Siek et al., 2005). Seventy-nine isolates were detected as APEC and 59 as AFEC. Statistical analysis showed significantly higher frequency of all 9 virulence genes (cvaC, iroN, iss, iutA, sitA, tsh, fyuA, irp-2 and ompT) in APEC than in AFEC isolates (Table 3A). In addition, all isolates were also classified according to the analysis of 5 genes (Johnson et al., 2008). A total of 81 isolates were defined as pathogenic and 57 as commensal. Again, a significantly higher frequency of the 5 virulence genes (iutA, iss, iroN, ompT and hlyF) was observed in APEC than in AFEC isolates (Table 3B). The 2 discordant results were obtained in samples of laying hens from the Midwest region that presented exactly the same result: detection of only 5 genes (hlyF, iroN, ompT, iss and sitA). As only 4 of these genes are among the 9 targeted, both samples were classified as AFEC by the first criterion (Rodriguez-Siek et al., 2005). In contrast, as the 4 of these genes are among the 5 targeted, these 2 samples were defined as APEC by the second criterion (Johnson et al., 2008).

Table 3.

Virulence gene frequency in Avian Pathogenic Escherichia coli (APEC) and Avian Fecal Escherichia coli (AFEC) isolates according classification by (A) Rodriguez-Siek et al. (2005) and (B) Johnson et al. (2008).

Genes APEC AFEC P-value1 P-value1 
 n = 79 n = 59 (X2  
 n(%) n(%)    
(A) 
iroN 78(98.7) 5(8.5) <0.0001 10.714 <0.00001 
ompT 79(100) 15(25.4) <0.0001 9.3001 <0.00001 
iss 76(96.2) 3(5.1) <0.0001 10.7039 <0.00001 
iutA 66(83.5) 6(10.2) <0.0001 8.5367 <0.00001 
cvaC 52(65.8) 6(10.2) <0.0001 6.5526 <0.00001 
sitA 70(88.6) 24(40.7) <0.0001 5.9771 <0.00001 
tsh 60(75.9) 13(22.0) <0.0001 6.2773 <0.00001 
fyuA 51(64.6) 9(15.3) <0.0001 5.78 <0.00001 
irp-2 47(59.5) 8(13.6) <0.0001 5.4525 <0.00001 
(B) 
Genes APEC AFEC P-value1 P-value1 
 n = 81 n = 57 (x2  
 (%) (%)    
hlyF 81(100) 14(24.6) <0.0001 9.4214 <0.00001 
iroN 80(98.8) 3(5.3) <0.0001 11.0464 <0.00001 
ompT 81(100) 13(22.8) <0.0001 9.5809 <0.00001 
iss 78(96.3) 1(1.8) <0.0001 11.0536 <0.00001 
iutA 66(81.5) 6(10.5) <0.0001 8.2161 <0.00001 
Genes APEC AFEC P-value1 P-value1 
 n = 79 n = 59 (X2  
 n(%) n(%)    
(A) 
iroN 78(98.7) 5(8.5) <0.0001 10.714 <0.00001 
ompT 79(100) 15(25.4) <0.0001 9.3001 <0.00001 
iss 76(96.2) 3(5.1) <0.0001 10.7039 <0.00001 
iutA 66(83.5) 6(10.2) <0.0001 8.5367 <0.00001 
cvaC 52(65.8) 6(10.2) <0.0001 6.5526 <0.00001 
sitA 70(88.6) 24(40.7) <0.0001 5.9771 <0.00001 
tsh 60(75.9) 13(22.0) <0.0001 6.2773 <0.00001 
fyuA 51(64.6) 9(15.3) <0.0001 5.78 <0.00001 
irp-2 47(59.5) 8(13.6) <0.0001 5.4525 <0.00001 
(B) 
Genes APEC AFEC P-value1 P-value1 
 n = 81 n = 57 (x2  
 (%) (%)    
hlyF 81(100) 14(24.6) <0.0001 9.4214 <0.00001 
iroN 80(98.8) 3(5.3) <0.0001 11.0464 <0.00001 
ompT 81(100) 13(22.8) <0.0001 9.5809 <0.00001 
iss 78(96.3) 1(1.8) <0.0001 11.0536 <0.00001 
iutA 66(81.5) 6(10.5) <0.0001 8.2161 <0.00001 

1The result is significant at P ≤ 0.05.

DISCUSSION

The mechanisms behind the pathogenesis and molecular epidemiology of APEC have been intensively studied in recent years. In general, pathogenic strains have an arsenal of virulence genes that code for factors that are directly associated with the ability of these bacteria to cause colibacillosis (Barnes et al., 2008). These genes have been the preferential molecular markers for the identification of the APEC strains and contributing to the reduction of losses in poultry production (Johnson et al., 2008). In our study, we evaluated the occurrence of 10 well-characterized virulence genes in 138 E. coli isolates from birds with characteristic signs of colibacillosis obtained from farms located in different Brazilian poultry-producing regions.

First, the 10 virulence genes were present in Brazilian isolates (from chickens and turkeys) with variable frequencies, but all in similar prevalence to that of the seminal studies (Rodriguez-Siek et al., 2005; Johnson et al., 2008). Although the first of these 2 studies demonstrated that the 9 targets analyzed were slightly more frequent (ranging from 86.4% for sitA to 57.2% for irp-2) when compared to our study (ranging from 68.8% for sitA to 39.9% for irp-2), the second study presented more similar gene frequencies in the 5 targets studied (68.1% for iroN to 41.5% for fyuA vs. 68.8% for hlyF to 43.5% for fyuA). Moreover, the second study analyzed a higher number of isolates (n = 994) than the first study (n = 555) (Rodriguez-Siek et al., 2005; Johnson et al., 2008).

Second, the occurrence and frequency of these genes may vary according to the geographic origin and year of the isolation. Many of these genes were also investigated in Brazilian studies across more than 10 years in the last 2 decades (Campos et al., 2008; Rocha et al., 2008; Kobayashi et al., 2011; Silva et al., 2011; Barbieri et al., 2013; Cunha et al., 2014; Maluta et al., 2014). The gene iss was the most studied (6 reports), and its frequency ranged from 51.3% in a study with a culture collection of APEC strains (Maluta et al., 2014) to 93% in a study with E. coli isolates from airsacculitis in turkeys (Cunha et al., 2014). Three genes were investigated in 4 previous studies, and the frequencies ranged from 43.4 to 95% (iroN), 23.7 to 66.7% (irp-2), and 36.8 to 66.7% (tsh). Also, 3 genes were studied in 3 more reports presenting frequencies ranging from 23 to 67% (cvaC), 25 to 46.5% (fyuA), and 40.8 to 68.1% (iutA). The most frequent gene, hlyF (68.8%), of this study occurred in high frequencies (67.1 and 70.3%) in 2 other reports (Kobayashi et al., 2011; Maluta et al., 2014). Interestingly, all the gene frequencies observed in the present study were inside these ranges demonstrated previously, indicating all of them are present in avian E. coli strains in the different poultry-producing regions and for many years in Brazil. However, this is not a definitive scenario for many other poultry farms and does not represent the virulence gene content in the whole country. More extensive studies should be performed with other industrial poultry flocks as well as with backyard chickens in Brazil.

In addition, nucleotide and amino acid sequences analyzed in this study were highly similar (over 99%) to both corresponding plasmid and chromosomal genes of the reference strains for APEC-O1 and APEC-O2 (Skyberg et al., 2006; Johnson et al., 2007). Although these genes show to be conserved and no polymorphic, they are rarely all present in the same isolate. APEC strains constitute a heterogeneous group of bacteria. Each APEC strain can have a specific combination of genes to induce colibacillosis in birds (Schouler et al., 2012). However, the main virulence genes seem to be the same and are strongly related to the occurrence of a plasmid similar to pColV in different studies conducted worldwide (Jeong et al., 2012; Schouler et al., 2012; Dissanayake et al., 2014).

The prediction of the pathogenicity of E. coli strains isolated from diseased animals is crucial. As biological assays are difficult to perform and there is a significant difference in gene prevalence between E. coli strains isolated from chickens with colibacillosis and those isolated from the feces of apparently healthy birds, new “virulence” genetic tests have been used to detect APEC strains. In the present study, we used 2 criteria previously described that consider the presence of a minimum number of virulence genes to differentiate APEC and AFEC strains (Rodriguez-Siek et al., 2005; Johnson et al., 2008). Both methods could separate pathogenic from non-pathogenic strains and were concordant in the analysis of 136 isolates (79 APEC and 57 AFEC). Only 2 isolates presented divergent results, probably because these 2 samples were positive for hlyF, the most frequent gene in APEC strains yet one that was not included in the first method described. Furthermore, all genes were more frequently detected in APEC than in AFEC isolates with highly significant differences (Table 3), similar to those observed in the original reports (Rodriguez-Siek et al., 2005; Johnson et al., 2008).

Routine diagnosis of colibacillosis in poultry farms from Brazil is based on the analysis of clinical signs with laboratory isolation of E. coli. Even serotyping using commercially available assays (which allow identification of the main avian serogroups O1, O2, and O78) is not performed in the routine of veterinary laboratories. The drawback of this procedure is the high rate of false positives obtained from the birds carrying AFEC strains. Molecular detection of virulence genes is surely easier to apply than other methods, and it would constitute a significant improvement in the control of this important disease.

The authors would like to thank the veterinarians and farmers who submitted clinical samples as well as the technicians of Simbios Biotecnologia and Laboratório de Diagnóstico Molecular (ULBRA) who performed technical support. This work was supported by Simbios Biotecnologia.

REFERENCES

Abramson
J. H.
WINPEPI updated: computer programs for epidemiologists, and their teaching potential
Epidemiol. Perspect. Innov.
 
2011
2
1
9
Barbieri
N. L.
Oliveira
A. L.
Tejkowski
T. M.
Pavanelo
D. B.
Rocha
D. A.
Matter
L. B.
Callegari-Jacques
S. M.
Brito
B. G.
Horn
F.
Genotypes and pathogenicity of cellulitis isolates reveal traits that modulate APEC virulence
PLoS One.
 
2013
8
e72322
Barnes
H. J.
Nolan
L. K.
Vaillancourt
J-P.
Saif
Y. M.
Colibacillosis
Diseases of Poultry
 
2008
Ames, IA
Blackwell Publishing
691
737
Campos
T. A.
Lago
J. C.
Nakazato
G.
Stehling
E. G.
Brocchi
M.
Castro
A. F. P.
Silveira
W. D.
Occurrence of virulence-related sequences and phylogenetic analysis of commensal and pathogenic avian Escherichia coli strains (APEC)
Pesq. Vet. Bras.
 
2008
28
533
540
Cunha
M. P.
Oliveira
M. G.
Oliveira
M. C.
Silva
K. C.
Gomes
C. R.
Moreno
A. M.
Knöbl
T.
Virulence profiles, phylogenetic background, and antibiotic resistance of Escherichia coli isolated from turkeys with airsacculitis
Scientific World J.
 
2014
2014
1
8
Dissanayake
D. R.
Octavia
S.
Lan
R.
Population structure and virulence content of avian pathogenic Escherichia coli isolated from outbreaks in Sri Lanka
Vet. Microbiol.
 
2014
168
403
412
Ewers
C.
Janssen
T.
Wieler
L. H.
Avian pathogenic Escherichia coli (APEC)
Berl. Munch. Tierarztl. Wochenschr.
 
2003
116
381
395
Ikuta
N.
Sobral
F. O. S.
Lehmann
F. K. M.
Silveira
V. P.
De Carli
S.
Casanova
Y. S.
Celmer
A. J.
Fonseca
A. S. K.
Lunge
V. R.
Taqman real-time PCR assays for rapid detection of avian pathogenic Escherichia coli isolates
Avian Dis.
 
2014
58
628
631
Jeong
Y. W.
Kim
T. E.
Kim
J. H.
Kwon
H. J.
Pathotyping avian pathogenic Escherichia coli strains in Korea
J. Vet. Sci.
 
2012
13
145
152
Johnson
T. J.
Kariyawasam
S.
Wannemuehler
Y.
Mangiamele
P.
Johnson
S. J.
Doetkott
C.
Skyberg
J. A.
Lynne
A. M.
Johnson
J. R.
Nolan
L. K.
The genome sequence of avian pathogenic Escherichia coli strain O1:K1:H7 shares strong similarities with human extraintestinal pathogenic E. coli genomes
J. Bacteriol.
 
2007
189
3228
3236
Johnson
T. J.
Wannemuehler
Y.
Doetkott
C.
Johnson
S. J.
Rosenberger
S. C.
Nolan
L. K.
Identification of minimal predictors of avian pathogenic Escherichia coli virulence for use as a rapid diagnostic tool
J. Clin. Microbiol.
 
2008
46
3987
3996
Knöbl
T.
Godoy
S. N.
Matushima
E. R.
Guimarães
M. B.
Ferreira
A. J. P.
Caracterização molecular dos fatores de virulência de estirpes de Escherichia coli isoladas de papagaios com colibacilose aviária
Braz. J. Vet. Res.
 
2008
45
54
60
Kobayashi
R. K.
Aquino
I.
Ferreira
A. L.
Vidotto
M. C.
EcoR phylogenetic analysis and virulence genotyping of avian pathogenic Escherichia coli strains and Escherichia coli isolates from commercial chicken carcasses in southern Brazil
Foodborne Pathog. Dis.
 
2011
8
631
634
Kabir
S. M. L.
Avian colibacillosis and salmonellosis: a closer look at epidemiology, pathogenesis, diagnosis, control and public health concerns
Int. J. Environ. Res. Public Health.
 
2010
7
89
114
Maluta
R. P.
Logue
C. M.
Casas
M. R.
Meng
T.
Guastalli
E. A.
Rojas
T. C.
Montelli
A. C.
Sadatsune
T.
Carvalho
M. R.
Nolan
L. K.
Silveira
W. D.
Overlapped sequence types (STs) and serogroups of avian pathogenic (APEC) and human extra-intestinal pathogenic (ExPEC) Escherichia coli isolated in Brazil
PLoS One.
 
2014
9
e105016
Matter
L. B.
Barbieri
N. L.
Nordhoff
M.
Ewers
C.
Horn
F.
Avian pathogenic Escherichia coli MT78 invades chicken fibroblasts
Vet. Microbiol.
 
2011
148
51
59
Mellata
M.
Ameiss
K.
Mo
H.
Curtiss
R.
Characterization of the contribution to virulence of three large plasmids of avian pathogenic Escherichia coli chi7122 (O78:K80:H9)
Infect. Immun.
 
2010
78
1528
1541
Rocha
A. C. G. P.
Rocha
S. L. S.
Lima-Rosa
C. A. V.
Souza
G. F.
Moraes
H. L. S.
Salle
F. O.
Moraes
L. B.
Salle
C. T. P.
Genes associated with pathogenicity of avian Escherichia coli (APEC) isolated from respiratory cases of poultry
Pesq. Vet. Bras.
 
2008
28
183
186
Rodriguez-Siek
K. E.
Giddings
C. W.
Doetkott
C.
Johnson
T. J.
Nolan
L. K.
Characterizing the APEC pathotype
Vet. Res.
 
2005
36
241
256
Schouler
C.
Schaeffer
B.
Brée
A.
Mora
A.
Dahbi
G.
Biet
F.
Oswald
E.
Mainil
J.
Blanco
J.
Moulin-Schouleur
M.
Diagnostic strategy for identifying avian pathogenic Escherichia coli based on four patterns of virulence genes
J. Clin. Microbiol.
 
2012
50
1673
1678
Silva
I. M. M.
Evêncio-Neto
J.
Silva
R. M.
Lucena-Silva
N.
Magalhães
J.
Baliza
M.
Genotypically characterization of Escherichia coli isolates from poultry
Arq. Bras. Med. Vet. Zootec.
 
2011
63
333
339
Skyberg
J. A.
Johnson
T. J.
Johnson
J. R.
Clabots
C.
Logue
C. M.
Nolan
L.K.
Acquisition of avian pathogenic Escherichia coli plasmids by a commensal E. coli isolate enhances its abilities to kill chicken embryos, grow in human urine, and colonize the murine kidney
Infect. Immun.
 
2006
74
6287
6292
Snedecor
G. W.
Cochran
W. G.
Statistical Methods, 7th ed.
1980
Ames, USA
Iowa State University Press
Tivendale
K. A.
Allen
J. L.
Browning
G. F.
Plasmid-borne virulence-associated genes have a conserved organization in virulent strains of avian pathogenic Escherichia coli
J. Clin. Microbiol.
 
2009
47
2513
2519