Characterization of Diarrheagenic Escherichia Coli Serotypes Isolated from Poultry and Humans

Ahmed Byomi1, Sherif Zidan1, Mohamed Diab2, Gopal Reddy3, Abiodun Adesiyun4 and Woubit Abdela3* 1Department of Animal Hygiene and Zoonoses, Faculty of Veterinary Medicine, University of Sadat City, Egypt 2Department of Animal Hygiene and Zoonoses, Faculty of Veterinary Medicine, New Valley Assiut University, Egypt 3Department of Pathobiology, College of Veterinary Medicine, Nursing and Allied Health, Tuskegee University, Tuskegee, USA 4School of Veterinary Medicine, Faculty of Medical Sciences, University of the West Indies, St. Augustine, Trinidad and Tobago SOJ Veterinary Sciences Open Access Research Article

Avian pathogenic E. coli (APEC) strains produce serious extraintestinal lesions in poultry causing high morbidity and mortality in chickens and turkeys, leading to considerable economic losses [2].In addition, E. coli is a common and an important pathogen that causes at least 5% of mortalities in poultry flocks [3],

Isolation and Identification of E. coli
One milliliter (1 ml) of homogenized BPW or 1 g of sample was transferred to Trypticase soy broth (TSB) and incubated overnight at 37oC.Loopful from incubated TSB was streaked for isolation on MacConkey agar plates and incubated for 24 h at 37oC.Lactose-fermented colonies were streaked on eosin methylene blue (EMB) agar plates and then incubated for 24 h at 37oC.The suspected purified colonies with metallic sheen were identified using standard biochemical tests [13].

Selection and Characterization of E. coli Isolates
One hundred (100) isolates comprising 50, 19, 13 and 18 E. coli isolates from broiler, ducks, backyards and humans respectively, were randomly selected for further characterization.

Vero Cell Assay (VCA) of E. coli Strains
The cytotoxicity of the E. coli isolates for Vero cells was determined according to the procedure described by Konowalchuk et al. [14].

Serotyping of E. coli Isolates
Serotyping of E. coli isolates was performed in the Serology Unit, Animal Health Research Institute, Doki, Egypt using commercially available antisera (Denka Seiken Co., LTD, Tokyo, Japan) by the slide agglutination test.Detection of stx1, stx2, eae, ST and LT genes in STEC isolates from fecal samples was performed using PCR as described by Paton and Paton [15].

Genomic DNA Extraction
Pure colonies of bacteria from MacConkey agar plates were sub-cultured into TSB and incubated for 12 h at 37oC.After incubation, bacteria were collected by centrifugation at 13000 rpm for 2 minutes and the sediment was suspended in equal volume of Tris-EDTA buffer.Thereafter, 100 µl of lysozyme solution (10 mg/L), 100 µl of proteinase K (0.3 mg/L) and 1% dodecyl sulphate were added.The DNA lysate was extracted once with chloroform/isoamyl alcohol (24:1 ratio by volume), and then extracted with phenol/chloroform/isoamyl alcohol (25: 24: 1 ratio).The aqueous phase was mixed with isopropanol alcohol and stored at -20oC for 30 minutes.The precipitated DNA was spooled out, rinsed in 70 % ethanol and dissolves in 0.5 ml of Tris EDTA buffer.

Statistical Analyses
The prevalence of E. coli strains, occurrence of virulence markers and serotypes of the isolates from poultry and human sources were compared to detect statistically significant differences using the Chi-square test.The level of significance was set at an alpha 0.05.

Prevalence of E. coli in Poultry and Human contacts
The prevalence of E. coli from three types of poultry and human beings is shown in [Table 2].Overall, of a total of 1528 samples collected from poultry and human beings, 622 (40.7%) were positive for E. coli.The prevalence of E. coli was 43.6%, 57.7%, 53.0% and 7.2% for broiler, ducks, backyard chickens (including litters) and humans respectively and the differences were statistically significant (P < 0.05; X2).Amongst poultry samples, the prevalence of E. coli in broilers (43.6%) was statistically significantly (P < 0.05; X2) lower than found in either backyard chickens (53.0%) or ducks (57.7%).

Frequency of Isolation of E. coli by type of Samples Collected
Overall, E. coli was isolated from 41.5% (78 of 188), 46.3% (193 of 417), 49.0% (148 of 302) and 49.9% (185 of 371) of unabsorbed yolk sacs, litter, cloacae and septicemic lesions respectively.The differences were however not statistically significant (P > 0.05; X2).[Figure 1] shows the isolation rates of E. coli from broilers, ducks and backyard chicken samples.For broilers, the range of isolation rates was from 39.1% (115 of 294) for litter samples to 47.8% (120 of 251) from for septicemic lesions (P < 0.05; X2).For duck samples, the lowest frequency of isolation of E. coli was 48.0% (12 of 25) which was originated from unabsorbed yolk sacs while the highest frequency was from litter samples, 65.1% (54 of 83) again the difference was not statistically significant (P > 0.05; X2).From backyard chickens, the frequency of isolation ranged from 46.7% (14 of 30) for septicemic lesions to 60.0% (24 of 40) for litter samples but the differences were not statistically significant (P > 0.05; X2).
A comparison of the prevalence of E. coli in feces of birds (without inclusion of their litters) revealed a prevalence of 45.8% (281 of 614), 54.1% (92 of 170), and 49.4% (38 of 77) for broilers, ducks and backyard chickens respectively.The differences however were not statistically significant (P > 0.05; X2).For poultry fecal samples, the prevalence of E. coli was 47.7% (411 of 861) which was ssignificant (P < 0.05; X2) and highe than the 7.2% (18 of 250) found in humans.

Frequency of Detection of Virulence Genes in E. coli Isolates
The frequency of virulence genes amongst VTEC strains isolated from poultry and human sources as detected by VCA is displayed in [Table 4].Overall, of the 30 VTEC/STEC strains from poultry and human isolates, 6 (20.0%), 16 (53.3%),8 (26.7%) and 14 (46.7%) were positive for stx1, stx2, stx1/stx2 and eae respectively.Of the isolates positive for stx genes, stx2 was most

Characterization of Diarrheagenic Escherichia Coli Serotypes Isolated from Poultry and Humans
Copyright: © 2017 Abdela, et al.
frequently detected (57.7%) in poultry isolates while stx1 was most frequently detected (50.0%) in human isolates.A similar frequency of eae was detected in poultry and human isolates of E. coli, 46.2% (5 of 7) and 50.0%(2 of 4) respectively.
The frequency of detection of the selected virulence genes in E. coli strains isolated from poultry and human sources by serogroups is shown in [Table 5].Serogroups O78 and O111 from poultry sources showed a relatively high frequency of STEC strains, 26.9% (7 of 26) and 23.1% (6 of 26) respectively.The 4 serogroups (O126, O111, O26 and O78) detected from human isolates of E. coli each had a frequency of 25.0 % (1 of 4) for STEC strains.[Figure 2] shows the presence of verocytotoxic genes stx1, stx2 and attachment gene eae, from different serotypes O119, O126 and O111 from poultry, and [Figure 3] shows serotypes O125 and O128 isolated from human carrying ST and LT genes.

Characterization of Diarrheagenic Escherichia Coli Serotypes Isolated from Poultry and Humans
Copyright: © 2017 Abdela, et al.

Discussion
The study was conducted using the 'One Health' concept which encompasses the interaction of animals, humans and the environment in the transmission of diseases [16].The primary goal was to study the similarities in the prevalence and characteristics of toxigenic E. coli strains in rural communities in Egypt where there is a high animal-human-environment interaction.This is primarily due to the fact that backyard chickens are in very close contact with humans in the households, the exposure of poultry farm workers to E.coli from broilers and ducks and finally, the consumption of poultry products from these villages.These exposures have been reported by other researchers and it is well established that chickens could serve as reservoirs of toxigenic E. coli as a result of poor hygienic practices in rural communities which may pose food safety concerns [6,11,17].
The carriage rate of E. coli in feces of poultry which ranged from 45.8 % to 54.1 % in the current study is lower than the 75.5 % reported for apparently healthy layers and their environments in India [18] and the 78% reported in ducks by Adzitey et al. [19].Considerably lower prevalence of 39.4 % was however detected in broilers feces by Abhilasha and Gupta [20].The isolation of E. coli, a normal flora of the intestinal tracks of animals and humans [21], is an indication that poultry could be potential sources of the microorganism to contaminate the environment and humans due to poultry meat consumption.Poultry have been recognized as reservoirs of pathogenic organisms such as Salmonella spp., Campylobacter spp., E. coli, amongst others [22].
In our study, although the prevalence of E. coli in the poultry species was significantly higher than found in human feces tested, frequencies of isolation of E. coli was reported to vary across countries, 60.0 % [23] and 62.5 % in Bangladesh [24].On the other hand, the frequency of isolation or detection of E. coli by various methods ranged from 0.4 % to 22 % in stool samples from human clinical laboratories in 10 European countries [25].
The prevalence of E. coli in humans in contact with poultry (9.2 %) was significantly higher than found in those without poultry contact (4.2 %), showing important zoonotic implications.A much higher frequency (73.0 %) of isolation of E. coli was detected in 30 healthy animal farm workers elsewhere [26].Several studies have reported the similarity of isolates of E. coli regarding the genes and antibiotic sensitivity, recovered from poultry and their human contacts [10,12].Riccobono et al. [11] had however suggested that cross-transmission between children and home raised chickens could not represent a major spreading mechanism for resistant E. coli in households of resource-limited settings with high human-animal interaction.It was also of zoonotic significance to have detected the four serogroups of E. coli (O126, O111, O26 and O78) isolated from humans were also recovered from poultry sources in the current study.It would however be necessary to apply more robust genetic tools, such as the pulse-field gel electrophoresis (PFGE) [27], to confirm the relatedness of the isolates.It is however pertinent to mention that the serogroups detected in the current study have also been isolated from poultry and humans elsewhere [28].
The detection of E. coli (39.1 % to 60.0 %) in the litters of poultry farms (broilers and ducks) and backyard chicken environments is considered high and the organism has been recovered in the litters of poultry houses by others [29].The significantly lower prevalence of E. coli in the litters on broiler farms compared with those from duck farms and backyard chicken environment may reflect a difference in management systems in which human contact is more prevalent on the duck farms and backyard chicken environment than the broiler farms.When changes of litters were completed on the duck farms and backyard chicken environment a higher exposure by humans to fecal matter resulted in a higher prevalence of E. coli, suggesting a higher risk factor for human infections by toxigenic E. coli [12].
The fact that E. coli was isolated at a significantly higher frequency from diarrheic humans (14.2 %) than from nondiarrheic (5.3 %) humans may be an indication that the microorganism was responsible either alone or in combination with other pathogens for the episodes.The pathogenic and toxigenic strains of E coli are an established human pathogens [2][3][4]30].
From epidemiologic, virulence and pathogenic point, out of the 100 isolates tested, 30 (30 %) were confirmed by PCA as VTEC strains and all possessed stx1, stx2 and stx1/stx2 genes making them all STEC strains.Other researchers have reported the detection of VTEC from animal sources, for example, in Côte d'Ivoire only one isolate of E.coli was determined to be a STEC strain [31] while Amézquita-Lópe et al. [32] reported isolating STEC strains from cattle, chickens and sheep.However, isolates of E. coli from chickens have been reported to be negative for STEC strains in Greece [33] and the USA [34].Although it has been established elsewhere that STEC strains are responsible for hemorrhagic gastroenteritis [35] in humans, only 4% of the isolates from humans were STEC strains.This was, higher than the 0.4% prevalence Shiga toxin strains in diarrheic humans reported by others [ [33], however only oone isolate was reported as 0157:HNM positive from human diarrheal stool specimens in Côte d'Ivoire [31].In our study, a comparatively high prevalence (26 %) of STEC was detected in poultry and their litters, which was considerably higher than the 6% reported for chickens in Burkina Faso [36].This is the first documentation of STEC strains from poultry sources in Egypt, an indication that poultry could be important reservoirs of STEC strains for human infections [36,37].
Overall in our study, VCA detected 30 % of the isolates tested to be VTEC strains, at a frequency of 31.7 % and 22.2 % in chicken and human isolates respectively.Comparatively lower frequencies of VTEC strains have been documented by others, 11% [3] and 9.1% [39] from chickens.The frequency of verocytotoxin producers amongst isolates from Egyptian villages in the current study (22%) is considerably lower than the 96.6% reported by Ananias and Yano [38], also using VCA.
The strategy employed in the current research, which used VCA followed by multiplex PCR successfully confirmed the presence of verotoxin (stx1, stx2, stx1/stx2) STEC strains as VTEC strains.Although the sensitivity of Vero cells to stx was first reported by Konowalchuk et al. [14], the cytotoxicity for this cell line remains the "gold standard" for confirmation of putative stx-producing isolates [15].In our study, there was a 100 % correlation between VCA and multiplex PCR results.However, the only inference that can be made on VTEC-positive (by VCA) and verotoxin gene-positive to classify them as STEC strains because they were not serotyped specifically for O157 strains, which is a limitation of the current study.It is known that both O157 and non-O157 STEC strains exist that are VTEC [32].
Our finding that 26.7% of the 30 STEC isolates were positive for both stx1 and stx2 genes is in agreement with published studies [43,44].The frequencies of detection have also been variable according to reports, with lower frequency of 10% for stx1 and 16.5% for stx2 genes [45] and a higher frequency than found in our study by others [46].In our study, a total of 6 (20 %) isolates were positive for only stx1 gene and 16 (53.3%) for only stx2 gene, a slight variance from the reports of others [39] who stated that stx1 was more frequently detected in STEC strains isolated from human patients with diarrhea.
The detection of 50% human isolates tested positive for the eae genes, which are usually associated with EPEC strains and known to be responsible for gastroenteritis [40], indicates possible pathogenic significance.
It is also pertinent to mention that 46.2% of the isolates from poultry from Egyptian villages were positive for EPEC strains and therefore could serve as potential sources for human infections.This by far is higher than reported elsewhere, 37% in chickens in Burkina Faso [35], 30.0% in diseased chickens in the USA [7] and 9.52% from chickens in India [43].Four serogroups (O126, O111, O78 and O1) of E. coli from poultry were eae genes positive and therefore EPEC strains.These strains have been isolated from poultry by others [41].
The prevalence of 8 % for EPEC strains detected in humans in our study compares favorably with the 7.6 % reported for Melbourne infants [42] and 5.9 % found in Berlin infants [42].However, relatively lower prevalence of 4.57 % for EPEC strains was reported in humans in while considerably higher prevalence has been reported in diarrheic (36.8%) and non-diarrheic (29.8%) humans by Garcia and others [44].
ETEC strains producing LT and ST have long been known to be responsible for diarrhea in both humans and animals [23,25].The prevalence of 28 % for ETEC strains in chicken isolates in our study is higher than the 5 % reported for chicken isolates in Burkina Faso [36] but comparative to the 22.8 % and 38.71 % detected in broilers and layers respectively in Bangladesh [23].
Regarding the frequency of detection of ETEC strains from human isolates in our study, 33 % is considerably higher than a range of 0.1% to 4.2% reported from human diarrheic patients in ten European countries [25].This is not a surprise because it has been reported that EPEC strains and other pathogenic and toxigenic strains of E. coli are more prevalent in developing countries where poor hygienic practices are more prevalent than in developed countries [25,45].Serogroups O128 and O125 detected in chicken and human isolates in this study were also reported among ETEC strains [46].The serotypes and the frequency of detection of both LT and ST genes, albeit at different frequencies, in E. coli isolates recovered from chickens and humans in rural communities in Egypt; indicate that they might play an important role as a cause of diarrhea in those areas.

Conclusion
Poultry (broilers, chickens and ducks) sampled from rural communities in Egypt displayed a significantly higher prevalence of E. coli than humans sampled in the same villages.However, the similarities detected amongst the strains of E. coli from poultry and human sources regarding the serogroups, EHEC, VTEC, ETEC and EPEC indicate that poultry in these villages have the potential to serve as important reservoirs of infection for humans.It is however imperative to use molecular methods such as the PFGE or MLST for future studies of this nature.This confirms the genetic relatedness of the E. coli strains isolated for poultry and humans in the Egyptian communities studied, which is considered a limitation of this study.It is also noted that although the sample size of 100 used in the current study which was primarily due limited resources, the study design which applied several serological and genetic methods successfully established

Characterization of Diarrheagenic Escherichia Coli Serotypes Isolated from Poultry and Humans
Copyright: © 2017 Abdela, et al.
significant similarities in occurrence and characteristics between human and poultry isolates studied.Finally, to the best of our knowledge the present study is the first report that studied the relationship of poultry-human transfer of diarrhegenic E. coli in rural communities in Egypt.
It is recommended that future studies should consider the use of PFGE and/or MLST to definitely confirm the relatedness of E. coli isolates from both poultry and human isolates in rural Egyptian communities.

Table 1 :
Primers used for PCR amplification

Table 2 :
Frequency distribution of E. coli isolation in poultry and human samples Figure 1: Prevalence of E. coli in different types of poultry sampled

Table 3 :
Serological characterization of E. coli isolates from poultry and humans

Table 4 : Frequency of virulence genes amongst VTEC strains detected by VCA
No. (%) of isolates positive for:

Table 6 :
Occurrence of selected virulence genes (ST and LT) in ETEC strains from poultry and human samples *ETEC: Enterotoxigenic E.