In the current study, we compared the efficiency of PCR detection to culture-dependent isolation of E. coli, as a model, from the liver and intestinal contents. Further, the incidence differences of E. coli isolation from the liver and intestine. Samples were collected from birds suffering from respiratory manifestation and/or diarrhea. Ninety E. coli isolates were recovered from 60 birds (52 intestinal and 38 liver samples) by bacteriological culture on selective broth and selective agar. PCR was performed using pho-A gene as a general marker for E. coli on DNA directly purified from the samples and assigned PCR (a). Negative PCR (a) samples were cultured on broth and another PCR was done, assigned PCR (b). Bacteriological isolation was more sensitive than PCR (a) indicating that inhibitors in the samples could have reduced or totally blocked the amplification capacity of PCR (a), which limited its diagnostic usefulness. PCR (b) was more sensitive than PCR (a) and more practical than bacteriological isolation. Detection of three virulence genes; iut-A, iss and tsh, showed genotypic variations of avian pathogenic E. coli (APEC) isolated from the liver and the intestine. In conclusion, an enrichment step results in more sensitive PCR than culture performed one. Further, general genes rather than virulence genes should be used as an indicator for E. coli detection in infected materials.
Key words: E. coli, PCR, pho-A, colibacillosis