Colony Multiplex PCR Assay for Identification and Differentiation of Campylobacter jejuni, C. coli, C. lari, C. upsaliensis, and C. fetus subsp. fetus

DNA was prepared by the whole-cell procedure. Each PCR template was prepared by using approximately half a loopful of culture transferred to 1 ml of brain heart infusion broth (Oxoid). The optical density was adjusted to give a reading of 0.3 at 600 nm. The optimized whole-cell DNA preparations from all Campylobacter species were further diluted 1:500 in distilled water and were heated at 100°C for 10 min in a 0.5-ml Eppendorf tube. Templates were used immediately for PCRs or were kept at 4°C for up to 1 month.

Oligonucleotides, ranging from 18- to 24-mers, were selected from the published DNA sequences of the various Campylobacter species (Table 1) using Oligo software (version 3.4). Synthesis of oligonucleotides was carried out at the DNA Core Facility in the National Microbiology Laboratory, Winnipeg, Canada. The six pairs of primers were designed to identify the genes hipO from C. jejuni; glyA from C. coli, C. lari, and C. upsaliensis; sapB2 from C. fetus subsp. fetus; and the internal control 23S rRNA. The primer sequences used in the multiplex PCR are outlined in Table 1. The colony PCR sensitivity was determined by both PCR and plate count methods; in brief, 10-fold serial dilutions of up to 10⁻¹⁴ were made in triplicate in brain heart infusion broth from 1- ml cultures with starting optical density readings of 0.3 at 600 nm. From these diluted samples, 100-μl aliquots from C. jejuni, C. coli, C. lari, C. fetus subsp. fetus, and C. upsaliensis isolates were plated onto Mueller-Hinton agar. Duplicate samples of each were used to perform standard colony counts and to evaluate the sensitivity of the PCR assay using agarose gel electrophoresis relative to actual plate counts within the range of 10² to 10¹³ CFU/ml. It appears that a range of 10⁸ to 10¹³ CFU/ml was most effective for all the Campylobacter colony multiplex PCR assays.

Each multiplex PCR tube contained 200 μM deoxynucleoside triphosphate; 2.5 μl of 10× reaction buffer (500 mM Tris-HCl [pH 8.3], 100 mM KCl, and 50 mM [NH₄]₂SO₄); 20 mM MgCl₂; 0.5 μM C. jejuni and C. lari primers; 1 μM C. coli and C. fetus primers, 2 μM C. upsaliensis primers; 0.2 μM 23S rRNA primer (Table 1); 1.25 U of FastStart Taq DNA polymerase (Roche Diagnostics, GmbH, Mannheim, Germany), and 2.5 μl of whole-cell template DNA. The volume was adjusted with sterile distilled water to give 25 μl. DNA amplification was carried out in a Perkin-Elmer thermocycler using an initial denaturation step at 95°C for 6 min followed by 30 cycles of amplification (denaturation at 95°C for 0.5 min, annealing at 59°C for 0.5 min, and extension at 72°C for 0.5 min), ending with a final extension at 72°C for 7 min. The six primer sets were evaluated individually for primer specificity using the reference strains C. jejuni NCTC 11168, C. coli NCTC 11353, C. lari NCTC 11352, C. upsaliensis ATCC 43954, C. fetus subsp. fetus ATCC 27374, C. fetus subsp. venerealis ATCC 19438, C. hyointestinalis ATCC 35217, C. sputorum biovar fecalis ATCC 33711, A. butzleri ATCC 49616, and A. butzleri-like CDCD2887. Only the corresponding strains showed the expected PCR amplification products. Reproducibility of the multiplex PCR assay was evaluated using 137 clinical and environmental samples.

Figure 1 illustrates the PCR-amplified products with the Campylobacter reference strains as templates following 1.5% agarose gel electrophoresis. In the assay, six bands were detected from a mixture of DNA containing each of the five Campylobacter spp. (Fig. 1, lane 8). The amplicons from the control strains were subjected to further confirmation and characterization by digestion using restriction endonucleases with cleavage sites within the amplicon. The restriction enzymes used and the predicted product sizes are shown in Table 2. Enzyme fragments with the anticipated sizes were obtained in each case (data not shown).

All 124 Campylobacter samples were identified by using biochemical assays and 16S rRNA-PCR (14). Complete agreement was obtained with the species-specific primers used in the present assay for all isolates examined. The amplicon for the Campylobacter 23S rRNA primers was present in all tested Campylobacter, Arcobacter, and Helicobacter isolates but failed to amplify E. coli and A. hydrophila isolates (Table 3). The sensitivity range of the colony multiplex PCR in number of CFU per milliliter was 10⁸ to 10¹³ for C. jejuni, 10⁶ to 10¹³ for C. coli and C. upsaliensis, 10⁷ to 10¹³ for C. lari, and 10² to 10¹³ for C. fetus subsp. fetus.

Case control studies demonstrate that human campylobacteriosis is a food-borne disease with infection most frequently resulting from handling and consuming contaminated poultry meat (21). Indeed, in one study it was reported that 73.2% of 489 meat samples were contaminated with the pathogen (12). Clinically the most important campylobacters are the members of the thermophilic group that includes C. jejuni, C. coli, C. lari, and C. upsaliensis, with C. jejuni responsible for the majority of human cases (1). C. fetus is also recognized as a human and animal pathogen and has been identified in 12.5% of ox liver samples (12). Accurate identification of these organisms is required in order to decide upon appropriate therapeutic measures, to understand the pathology of disease, and to provide clinical and epidemiological data for disease control. A number of protocols (3-8, 11, 17, 20, 22) have been described in the literature for use in the differentiation of the closely related thermophilic C. jejuni, C. coli, C. lari, and C. upsaliensis species as well as of C. fetus. Most of these methods are based on DNA probe technology or rRNA PCR-restriction fragment length polymorphism. Due to a requirement for restriction enzymes or hybridization steps with species-specific probes, both are relatively complex to perform. An assay combining PCR and DNA hybridization was developed for the rapid detection of C. fetus (5), but it is not capable of differentiating C. fetus subsp. fetus from C. fetus subsp. venerealis. In a recent study multiplex PCR was used for simultaneously differentiating C. jejuni, C. coli, and C. lari (6). This assay was based on the sequence information of the gene encoding the oxidoreductase subunit of C. jejuni (23; GenBank accession no. AL139075 ), the aspartokinase gene for C. coli (13), and 16S rRNA for C. lari (16).

Detection by PCR of the hipO gene, shown to be highly conserved in C. jejuni, provided an effective identification marker for C. jejuni (19, 20). In addition, PCR hybridization confirmed that the Campylobacter glyA gene can be used as the target to identify and differentiate C. jejuni, C. coli, C. lari, and C. upsaliensis at the species level (2). Furthermore, the sapB2 gene of C. fetus subsp. fetus was recognized as a suitable target for identifying C. fetus (4).

In this study, a colony multiplex PCR-based diagnostic protocol was developed to simultaneously detect five genes specific to each of the five pathogenic Campylobacter species, while the 23S rRNA probe was included to serve as an internal validation control to monitor PCR conditions and reagents. The 23S rRNA was found in all of the Campylobacter, Arcobacter, and H. pylori species investigated.

The present colony multiplex PCR assay proved to be accurate and simple to perform and could be completed within 3 h. It had the added advantage of detecting the hipO gene in C. jejuni strains that were hippuricase negative by phenotypic methods and therefore difficult to differentiate from C. coli (5, 20). In addition to clinical use, the method has potential as a diagnostic kit for detecting thermophilic campylobacters in complex samples, such as foods in which low pathogen numbers (<10³ CFU/ml) are frequently present. The present PCR assay offers an effective alternative to traditional biochemical typing methods for the identification and differentiation of C. jejuni, C. coli, C. lari, C. upsaliensis, and C. fetus subsp. fetus isolated from humans and poultry.

We thank Roman Benes for technical assistance.