Real-time TaqMan PCR for Yersinia enterocolitica detection based on the ail and foxA genes.

Yersinia enterocolitica, a cause of emerging enteric infections, is a foodborne pathogen associated with various enteric and systemic syndromes, e.g., diarrhea, enteritis, and enterocolitis. Therefore, the detection of this pathogen has important significance. Previous real-time PCR for detection of Y. enterocolitica was primarily based on the ail gene; biotype 1A nonpathogenic strains were not included (1–3). However, recent studies (4, 5) showed that biotype 1A ystB-positive strains are potentially pathogenic and related outbreaks were reported. We therefore designed a TaqMan real-time PCR method for detection of both pathogenic and nonpathogenic Y. enterocolitica strains. Using data from sequence analysis of the ail and foxA genes from many Y. enterocolitica strains (6), we designed TaqMan probes and primers (Table 1). An experiment using the entire reaction system was performed using a 20l volume containing 10 l premix (TaKaRa; China), 7.2 l ultrapure distilled water, 0.2 l ROXII, and 0.2 l (100 nmol/liter) of each primer and probe. A two-step method was adopted. The cycling conditions for the use of a Rotor-Gene Q system consisted of 1 cycle of initial denaturation at 95°C for 10 s followed by 40 cycles of melting at 95°C for 5 s and elongation at 60°C for 30 s. And for the use of a ABI 7500 Fast system, cycling was performed using one initial denaturation at 95°C for 20 s followed by 40 cycles of melting at 95°C for 3 s and elongation at 60°C for 30 s. A total of 168 pathogenic Y. enterocolitica strains (3 ail sequence patterns and 8 foxA sequence patterns) and a total of 41 nonpathogenic Y. enterocolitica strains (13 foxA sequence patterns) were used to assess the sensitivity and specificity of the method. Most of these strains were isolated from animals, mainly swine and mice, in China. Furthermore, 258 non-Y. enterocolitica strains were used to test the specificity of the two detection systems. Most of the strains were Gram-negative bacteria of various genera. All the strains were isolated from patients and identified by using a Vitek Compact 2 biochemical identification instrument (bioMérieux). The results showed that both the ail and foxA gene detection systems have 100% specificity. Standard curves and sensitivity data were obtained by amplifying standard plasmid that had been serially diluted 10-fold. In parallel, we detected the sensitivity of conventional PCR. The results suggested that the slope was 3.09 and the R was 0.99 for the ail system and that the slope was 3.16 and the R was 0.99 for the foxA system. The detection limit was 10 copies/ l for both detection systems. This represented sensitivity 10 times greater than that of the conventional PCR detection. To exclude false-negative results caused by potential inhibitors, we used the internal amplification control (IAC) developed by Fricker et al. (7). A total of 15 pathogenic Y. enterocolitica strains were used to test the IAC with the ail and foxA detection systems. When the ail and foxA detection systems were mixed with the IAC, both of them amplified well with IAC and also had no nonspecific amplification, thus showing that the IAC that we used was suitable for our detection systems. In our laboratory, combining conventional PCR and culture isolation achieves good results; currently, we first employ PCR screening to find positive samples with the Y. enterocolitica conserved foxA gene or pathogenic ail gene and then inoculate positive samples onto cefsulodin-irgasan-novobiocin isolation media (CIN agar; Difco). To compare real-time PCR detection, conventional PCR detection, and culture isolation methods, we tested 228 separate an-