Fig 1.
Graphical representation of the primers and probes used in this study.
(A) The aroA mutant gene (1279 bp) is schematically represented by the blue line on which the mutation is indicated in red. Underneath this line, the amplicons generated by the different primer pairs (arrows) are represented. Above the schematic gene representation, the location of the probes is indicated with a star. Since this figure depicts the aroA mutant gene, the primer pair aroA_WT cannot be represented. (B) A similar representation for the xanQ gene (1401 bp) in blue with probes (above) and primers (below).
Table 1.
List of primer sequences.
Table 2.
List of probe sequences.
Fig 2.
Discrimination of the ΔaroA vaccine strain from other E. coli strains and non-E. coli strains using PCR.
The specificity of the primers targeting either the E. coli ΔaroA gene (A), the WT E. coli aroA gene (B), the previously published primers targeting both the WT and ΔaroA gene (total aroA) (D) or simultaneous detection of the ΔaroA and xanQ gene (C), was screened against a total of 132 E. coli strains and 7 non-Escherichia strains. (A) PCR specific for the ΔaroA gene. Amplification only occurred for the vaccine strain, while none of the commensal E. coli strains or non-Escherichia strains was amplified. The picture shows the results from primer pair aroA_3, which results in a band at 281bp when the aroA mutation is present. All other tested ΔaroA specific primers showed the same specificity. (B) The aroA primers targeting the WT gene (aroA_WT) result in a band at 660 bp in commensal E. coli, whereas no band occurred when the aroA mutation was present or in the non-Escherichia strains. (C) Duplex PCR with aroA_3 and xanQ_2 primers. The E. coli-specific xanQ primers result in a band at 97 bp, whereas the band at 281 bp indicates ΔaroA. Only for the ΔaroA E. coli strain, the two bands are present. (D) Primer pair aroA_LR (La Ragione et al., 2013, Table 1) is not specific for the ΔaroA mutation and results in an amplicon of 1236 bp for wild type E. coli, or 1161 bp for the ΔaroA vaccine strain. The pictures show a representative selection of the total number of strains used to assess the primer specificity. Samples: 1: blank control 2: E. coli O78:K80 ΔaroA (vaccine strain), 3: E coli O78:K80 WT, 4: E. coli Nissle 1917, 5: APEC_1 (chicken intestine), 6: APEC_2 (turkey intestine), 7: APEC019 (chicken intestine), 8: APEC023 (chicken intestine), 9: 220115–1 (chicken intestine, aromatic amino acid auxotrophic E. coli), 10: 220115–10 (chicken intestine, aromatic amino acid auxotrophic E. coli), 11: 06114–21 (chicken intestine, aromatic amino acid auxotrophic E. coli), 12: 7-35s (chicken intestine, aromatic amino acid auxotrophic E. coli), 13: 3-44s (chicken intestine, aromatic amino acid auxotrophic E. coli), 14: Salmonella Enteritidis, 15: Salmonella Typhimurium, 16: Clostridium perfringens NE18, 17: Clostridium perfringens CP20, 18: Bacillus subtilis DSM 29784. The complete list of tested strains can be found in S1 Table. Electrophoresis was performed on 1.5% agarose gel. GeneRulerTM 100 bp Plus DNA Ladder was used as a molecular weight marker.
Table 3.
Comparison of efficiency and limit of quantification in different PCR assays for all primer pairs.
For each primer pair (Table 1) the efficiency, limit of quantification (LOQ) and limit of detection (LOD) were assessed using a 10-fold dilution series of a standard fragment (108−100 copies/μl).
Table 4.
Comparison of efficiency, limit of quantification and limit of detection in the two different duplex probe-based qPCR set-ups.
For each of the two primer/probe combinations the efficiency, limit of quantification and limit of detection were assessed using a 10-fold dilution series of a standard fragment (108−100 copies/μl). The choice for these specific combinations of primer pairs for both genes was based upon limited primer/probe interactions.
Fig 3.
Amount of detected copies/ml ΔaroA vs original input.
The amount of detected ΔaroA (copies/ml) for the three different qPCRs (the SYBRgreen simplex qPCR (black triangles), probe-based simplex qPCR (dark grey squares) and probe-based duplex qPCR (light grey circles) is depicted. (A) The input of the qPCR is genomic DNA originating from a 10-fold dilution series of an overnight culture from the ΔaroA strain. (B) The input for the qPCRs is DNA isolated from fresh chicken faeces spiked with a 10-fold dilution series of the ΔaroA strain.
Fig 4.
Detection of ΔaroA vaccine-strain and total E. coli load in ileal samples of vaccinated (n = 7) or non-vaccinated birds (n = 7).
Birds from the vaccinated group received 108 CFU ΔaroA vaccine per bird at day of hatch. At 3 days of age (2 days after vaccination), ileal content was collected. (A) Simultaneous detection of the vaccine strain (targeting the mutant aroA gene) (dark grey bars) and total E. coli load (targeting the xanQ gene) (light grey bars) was obtained using the duplex qPCR. Total E. coli load in the ileum of vaccinated birds tended to be lower than in non-vaccinated birds (p = 0.053). (B) 92.4% of the total E. coli load in the vaccinated birds was attributed to the ΔaroA vaccine strain, whereas in the non-vaccinated birds, no vaccine strain was detected.