Differentiation of Enterotoxigenic and Enteropathogenic Escherichia coli Isolated from Humans and Calves by rep-PCR

To determine the virulence markers of enterotoxigenic E. coli [ETEC] and enteropathogenic E. coli [EPEC], 224 stool specimens from diarrheic humans and 60 diarrheic calves were examined by PCR. This study was conducted to achieve a better understanding of the genetic relationships and evolutionary patterns within E. coli pathotypes isolated from humans and calves. Stool specimens from 224 diarrheic humans and 60 diarrheic calves were cultured for E. coli. The samples were obtained from North West of Iran. The rep-PCR fingerprint technique, which performed with primer BOX A1R, was used for discriminating between human and calf sources. A total of 25 ETEC and EPEC strains isolated from humans and calves. The discriminant analysis showed an Average Rate of Correct Classification [ARCC] of 97% for ETEC and 94.5% for EPEC isolates. This result reveals that the rep-PCR fingerprint technique with primer BOX A1R may be an effective technique for discriminating and grouping E. coli isolates, and can be empolyed as a source following tool for differentiation and identification of host sources.


INTRODUCTION
Diarrhoeal disease is a significant worldwide problem, especially in the developing countries 1 .Diarrheagenic Escherichia coli are one of the most important etiologic agents of diarrhea, and a major agent of diarrhea leading to high morbidity and mortality, mostly among children in the developing countries 2,3 .E. coli pathotypes are important pathogens in bovine neonates and the main cause of economic loss on farms 4,5 .Cattle and sheep, are main reservoirs for the transmission of pathogenic E. coli to humans through the food chain 6,7 .
The identification of pathogenic E. coli requires finding agents that determine the virulence of these organisms; it cannot be based only on cultural and biochemical criteria.Polymerase Chain Reaction [PCR] is a usually used method based on the virulence-associated factors or genes.It is capable of identifying E. coli pathotypes as well as other non-E.coli pathogens.With the finding of PCR, it has become feasible to detect virulence genes in bacterial isolates, allowing the fast and reliable recognition of pathogenic E. coli 10 .
In order to find effective control plans, it is necessary to define the source of fecal pollution and clonal relatedness of diarrheagenic E. coli 11,12 .Nowadays, there are a number of DNA analysis techniques for discriminating between human and animal origins of fecal contamination.These methods include pulsed-field electrophoresis [PFGE], ribotyping, ribosomal DNA heterogeneity, and repetitive extragenic palindromic-PCR [rep-PCR].There have been several works of analytical techniques for finding human and animal origins of fecal contamination 13 .
This study was conducted to achieve a good perception of the genetic relationships and evolutionary patterns within E. coli pathotypes isolated from humans and calves.The rep-PCR DNA fingerprint, which uses repetitive exteragenic DNA sequences, was examined as a method for grouping the EPEC and ETEC isolates.This method is simple, fast and less expensive than PFGE, and has proven to work well to distinguish between species of E. coli pathotypes.

Bacterial sources and isolation
Stool specimens from 224 diarrheic humans and 60 diarrheic calves were cultured for E. coli.The isolates were obtained from North West of Iran.The samples were collected using sterile rectal swabs, inoculated in Cary Blair medium tubes, and carried to the laboratory.They were cultivated on MacConkey agar (Merck, Germany) and incubated at 37°C for 24 h, two to three lactose positive per plate were selected and submitted to the biochemical tests.The isolates were stocked in trypticase soy broth supplemented with 20% glycerol at -80°C for further procedures.

DNA extraction
Bacterial DNA extraction was taken by using the Promega DNA extraction kit (A11125, USA).The extracted DNA was used as a template for PCR according to the described protocols.

Detection of virulence genes by PCR
The DNA templates were subjected to polymerase chain reaction with appropriate primers (Table 1) for detection of the following virulence factors: eae and bfp for EPEC, lt and st for ETEC labile and stable toxins, respectively.The PCR assays were accomplished in a 25 µl reaction mixture, consisting of 2X PCR Master Mix (2X concentrated solutions of Taq DNA polymerase, reaction buffer, MgCl 2 and dNTPs) with a BioRad T100 TM thermal cycler.2X PCR Master Mix (CinnaGen Inc.) contain all components for PCR, except DNA template and primers.Primers were provided by GeNetBio Inc. (Korea).
A duplex-PCR was optimized for detection of ETEC.The PCR conditions for lt and st amplification were: predenaturation at 95°C for 5 min for one cycle followed by 35 cycles of 95°C for 45 sec, 49°C for 45 sec, 72°C for 45 sec and final extension at 72°C for 7 min.The second duplex-PCR was standardized for detection of EPEC.The PCR conditions for eae and bfp determinants of EPEC were: predenaturation at 95°C for 3 min for one cycle followed by 38 cycles of 95°C for 1 min, 53°C for 1 min, 72°C for 1min and final extension at 72°C for 10 min.Amplified PCR products were observed after electrophoresis on 1% agarose and staining with safe dye.The PCR products were visualized under UV transilluminator and photographed.

The rep-PCR condition
The rep-PCR assay was carried out with 50 ng template DNA and 2 µM BOX A1R primer (52 -CTA CGG CAA GGC GAC GCT GAC G-32 ) 14 .The rep-PCR amplifications were performed with predenaturation at 95°C for 2 min 35 cycles including of denaturation at 94°C for 30 sec and 55°C for 1 min, and annealing at 65°C for 1 min, followed by a single step extension at 65°C for 8 min.The separation of amplified DNA fragments was achieved by electrophoresis on 1% agarose gel with 1kb and 100bp DNA ladder (GeNetBio Inc.Korea).The gels were stained with safe dye, and imaged under UV illumination.

RESULTS
Diarrheagenic E. coli was isolated from 26.8% of the humans with diarrhea and 83.3% of the calves with diarrhea.Distribution of ETEC and EPEC is shown in Table 2.We found the incidence of ETEC, typical EPEC and atypical EPEC pathotypes in humans as 16.6%, 1.6%, 3.3%, and in calves were 4%, 8%, 12%, respectively.Fig. 1 shows the typical fingerprint for E. coli isolates by using rep-PCR carried out with BOX A1R primer.All strains were studied in terms of the electrophoretic profiles of their DNA fragments.The fingerprinting generated separate bands ranging in size from 500 to 2000 bp.The band patterns of E. coli pathotypes from humans and calves were very similar, and it was deduced that the strains were closely related.
The rep-PCR fingerprints were handily determined and statistical analysed with NTSYSpc program (Ver.2.2).The dendrogram in Fig. 2 shows the clonal structure and the genetic relationships of the 25 strains isolated from humans and calves.It was generated using Jaccard similarity coefficients and UPGMA algorithm, and produced two major clusters (A, B).Cluster A is genetically homogenous including the majority of ETECs obtained from humans.In contrast to cluster The strains were manually assigned to the correct group, and a Jackknife analysis was used to determine how precisely the similarity coefficients predicted the source group.According to findings of the present study, humans ETEC, calves ETEC, humans EPEC and calves EPEC were highly classified with rate of correct classification [RCC] of 96%, 98%, 94%, and 95%, respectively.Also, humans and calves isolates were highly classified with an average rate of correct classification [ARCC] of 97% for ETEC and 94.5% for EPEC isolates.It was deduced that both cluster and Jackknife analyses are suitable statistical methods for bacterial source tracking.
Matrix correlation, a clustering method based on correlation and dependence, can be used to investigate any statistical relationship between multiple random variables at the same time and it is very useful to find the most correlated variables in a data table having the correlation coefficient between each variables.The correlation coefficient, denoted by r, values between -1.0 to +1.0.The closer r is to +1 or -1, the more strong relation the two variables have.If r is close to 0, it means there is no linear correlation between the variables 15 .In present study, tests for association were generated by Matrix correlation plot using an option that accounted for comparison of similarity matrices (Fig. 3).According to the results, correlation coefficient is 0.8.It means that variables are closely related.

DISSCUSSION
ETEC and EPEC are two of the six pathogenic E. coli identified in our study.Among the recognized E. coli pathotypes, ETEC is very important and usual, particularly in the developing countries 16 .ETEC can be dangerous due to serious fluid loss and critical dehydration.Beyond its burden in endemic areas, ETEC is the main reason of diarrhea in travelers to developing countries 6,9 .ETEC also causes diarrhea in new born calves and it is one of the main causes of economic loss in farms 5 .EPEC strains have eae gene but do not produce Shiga toxin, which can cause diarrhea in humans and different animal species 17 .The results further showed the significance of ETEC as the reason of human diarrhea in the analyzed area the North West of Iran.Similar results have been reported by Tornieporth et al., 18 and Vilchez et al., 1 .Our findings are the documented results by Bueris et al., in Salvador, Bahia 10 .However, in a study carried out by Osman et al., 7 the results obtained from the calve diarrhea were in contrast to our findings.
RCC values of the present study are within the range of RCC values showed by Carson et al., 19 and Dombek et al., 12 .But, they were higher than the values found by Seurink et al., 20 and Mohapatra et al., 11 .The ARCC obtained in present work is different in comparison with the results described by Mohapatra et al., and Carson et al., with E. coli isolates from human and animal sources 11,19 .Also the results of our study were comparable to the other studies, worked with fecal E. coli isolates from different host groups 12,20,21 .
Up to now, there is no definite standard of precision available to bacterial source tracking methods.Any Classification Technique with a rate of correct classification from 60%-70% is considered to be useful for the fecal pollution control authorities to prevent more infections 11 .
Based on the discriminatory power and higher RCC, BOX A1R-PCR was found to be appropriate technique for discrimination of fecal E. coli isolates from different sources.BOX A1R-PCR genomic fingerprinting, which is fast, low cost and easy, may be used as a supplementary molecular tool for detection of fecal E. coli origins.Latter surveys should be carried out to determine efficacy of BOX A1R-PCR with fecal E. coli isolates from other possible sources of fecal contamination as well as with other usual techniques, such as ribotyping and antibiotic resistance analysis before its application in field studies 7 .Our study is the first and most comprehensive report on genetic relationship in ETEC and EPEC pathotypes isolated from diarrheic calves and humans in Iran.It was deduced that, calves could be a source of infection to humans as the major pathogens previously explained and associated with severe disease in humans.
In conclusion, our results indicate that rep-PCR is able to discriminate between ETEC and EPEC strains.We believe that rep-PCR can be an appropriate method for differentiating and clustering the E. coli isolates.This method could further be helpful for determining the sources of closely associated E. coli strains obtained from environmental samples.

Fig. 2 :Fig. 3 :
Fig. 2: Dendrogram illustrates the clonal structure and the genetic relationships of the 25 strains isolated from humans and calves