Rapid identification of Campylobacter spp. by melting peak analysis of biprobes in real-time PCR

Abstract

We describe rapid PCR-biprobe identification of Campylobacter spp. This is based on real-time PCR with product analysis in the same system. The assay identifies enteropathogenic campylobacters to the species level on the basis of their degree of hybridization to three 16S ribosomal DNA (rDNA) biprobes. First-round symmetric PCR is performed with genus-specific primers which selectively target and amplify a portion of the 16S rRNA gene common to all Campylobacter species. Second-round asymmetric PCR is performed in a LightCycler in the presence of one of three biprobes; the identity of an amplified DNA-biprobe duplex is established after determination of the species-specific melting peak temperature. The biprobe specificities were determined by testing 37 reference strains of Campylobacter, Helicobacter, and Arcobacter spp. and 59 Penner serotype reference strains of Campylobacter jejuni and C. coli. From the combination of melting peak profiles for each probe, an identification scheme was devised which accurately detected the five taxa pathogenic for humans (C. jejuni/C. coli, C. lari, C. upsaliensis, C. hyointestinalis, and C. fetus), as well as C. helveticus and C. lanienae. The assay was evaluated with 110 blind-tested field isolates; when the code was broken their previous phenotypic species identification was confirmed in every case. The PCR-biprobe assay also identified campylobacters directly from fecal DNA. PCR-biprobe testing of stools from 38 diarrheic subjects was 100% concordant with PCR-enzyme-linked immunosorbent assay identification (13, 20) and thus more sensitive than phenotypic identification following microaerobic culture.

FIG. 1

Biprobe fluorescence data. Shown are negative first-derivative plots, reflecting the change in fluorescence divided by the change in temperature (−dF/dT). The melting peaks seen in the range from 48 to 75°C correspond to the decrease in FRET and hence the decrease in light emitted at the Cy5 wavelength, as the biprobe melts from the single-stranded target DNA. The secondary peak seen at 86°C acts as a positive amplification control and corresponds to residual double-stranded PCR product. (A) Melting peaks obtained with biprobe A. C. jejuni/C. coli gave a peak at 64°C (dashed line), C. jejuni gave a peak at 58°C (dotted line), C. lari/C. upsaliensis gave a peak at 60°C (solid line), and C. helveticus gave a peak at 54°C (irregularly dashed line). (B) Melting peaks obtained with biprobe B. C. hyointestinalis gave a peak at 64°C (dashed line), C. jejuni/C. coli/C. lari/C. helveticus gave a peak at 62°C (solid line), C. fetus/C. lanienae gave a peak at 58°C (dotted line), and C. upsaliensis gave a peak at 51°C (irregularly dashed line), (C) Melting peaks obtained with biprobe C. C. fetus gave a peak at 67°C (dashed line), C. jejuni/C. coli/C. lanienae gave a peak at 64°C (solid line), C. jejuni/C. coli/C. lari/C. helveticus gave a peak at 60°C (dotted line), and C. upsaliensis gave a peak at 56°C (irregularly dashed line).