Considerations in Evaluating the Applicability of Universal Detection of Oral Pathogens

Akihiro Yoshida et al. (4). recently published an interesting paper describing the development of a 5′ fluorogenic nuclease-based real-time PCR assay for quantitative detection of the oral pathogens Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis. The authors provide the description of a set of primers and probe for amplification and detection of a broad range of bacteria based on highly conserved regions in the 16S rRNA gene. Alignment of the forward and reverse primers and probe sequences, however, showed mismatches at the 5′ as well as the 3′ ends of the 16S rRNA sequences with different oral pathogens (Fig. 1). Most likely these mismatches interfere with the amplification of P. gingivalis. In addition, according to Yoshida et al., the primers and probe sequences were based on earlier published data (2). We noticed that except for the probe sequence, the primer sequences presented in the publication of Greisen et al. (2) are located 172 and 254 bases, respectively, upstream and 180 bases downstream from the site indicated by Yoshida et al.

FIG. 1.

Alignment of the 16S rRNA sequence region described by Yoshida et al., including six oral bacteria compared to the E. coli sequence. The locations of the universal primers (Uni152-F and Uni220-R) and the probe (Uni177-T) are identical to the E. coli sequence as indicated. Strains: 1, DH5α; 2, W83; 3, ATCC 29523; 4, ATCC 43037; 5, ATCC 33270, 6, ATCC 25611; 7, ATCC 10953. Dots, nucleotides identical to those in the E. coli sequence; dashes, deletions.

We have tested this universal probe and primer set with one nonoral bacterium and six oral bacteria: Escherichia coli, P. gingivalis, A. actinomycetemcomitans, Tannerella forsythensis (formerly Bacteroides forsythus), Micromonas micros (formerly Peptostreptococcus), Prevotella intermedia, and Fusobacterium nucleatum. Growth of the bacteria, preparation of serial dilutions of pure cultures, DNA isolation, and real-time PCR amplification were performed as described previously (1).

The E. coli amplification signals consistently showed an increasing C_T range (i.e., decreasing amplification signal) between 15.9 and 26.3, which corresponded to 4.95 × 10⁷ to 4.95 × 10⁴ CFU equivalents/PCR. All the oral bacteria, however, showed a C_T range between 29.5 and 33.5, which is at the level of the negative control signal (C_T = 29.5). For A. actinomycetemcomitans, an amplification signal was obtained only with a high DNA concentration, corresponding to 10,000 CFU/ml, and the signals obtained with lower DNA concentrations were at the level of the negative control. To validate the DNA isolation, a real-time PCR with a specific primer-probe combination for E. coli (3) and P. gingivalis (1) was used. The amplification signals obtained with the specific PCR on the same DNA confirmed that the amount of DNA present in the dilution corresponded to the amount isolated.

In summary, the results with the eubacterium real-time PCR confirm the mismatches in the alignment but are in contradiction with the results on the same strains presented by Yoshida et al. In our laboratory, the primer and probe set described by Yoshida is not universal, does not amplify P. gingivalis, T. forsythensis, M. micros, P. intermedia, or F. nucleatum, and amplifies A. actinomycetemcomitans only at a high DNA concentration. We conclude that the universal real-time probe and primer set published by Yoshida et al. to detect A. actinomycetemcomitans and P. gingivalis does not seem applicable for sensitive detection of oral bacterium species belonging to the major bacterial periodontal pathogens.