BACKGROUND: Although the assessment of the density of Helicobacter pylori (H. pylori) colonization may be important, precise quantitative evaluation of H. pylori in the gastric mucosa has been difficult. We have developed a novel real-time quantitative PCR technique based on 5' -3' exonuclease assay for detection of H. pylori. Using this system, we evaluated number of H. pylori organisms present in gastric biopsy specimens. METHODS: A primer pair for PCR amplification was made according to the reported sequence of ureA gene of H. pylori, and an oligonucleotide probe labeled with two fluorescent dyes (a reporter dye at the 5' end and a quencher dye at the 3' end) was also designed. While the probe is intact, the quencher dye greatly reduces the fluorescent signal by the reporter dye (Fluorescence resonance energy transfer phenomenon). During the PCR extension phase, the annealed probe is cleaved by 5' -3' exonuclease activity of Taq DNA polymerase, increasing the reporter dye signal. The reporter dye signal during a PCR assay was monitored with a ABI PRISM 7700 Sequence Detector (Perkin-Elmer). Purified PCR products of ureA gene were used as standard samples for making standard curves plotting log starting copy number and threshold cycle (Ct: PCR cycle number at which the reporter fluorescence signal reached above baseline). Biopsy specimens were collected from antral mucosa of consecutive 32 patients underwent gastroscopy (GU 9, DU 9, GC 1, NUD 13). Three biopsy specimens were obtained from each patient to avoid the effect of patchy distribution of H. pylori. DNA was extracted by conventional phenol-chloroform method. Since the size of each biopsy specimen was variable, quantitative PCR assay described above was performed using 0.5 ~tg of DNA extracted from each sample. !~_!LL_.T~: In each amplification reaction, a liner standard curve could be generated in a wide range of starting copy numbers (6xl07 to 6xl0 copies of the target sequences, R 2 > 0.99). Nine of 32 samples (DU 2, NUD 7) showed no increase in the fluorescence signal during PCR process, indicating the absence of H. pylori, which was also confirmed by histological examinations. In H. pylori positive patients, the detected ureA copy numbers (i.e. number of H. pylori organisms) were variable among patients (range 1.4x102 1.2x105 copies/0.5Mg DNA, mean 2.6x104 copies/0.51ag DNA). The average ureA copy numbers of samples from DU and GU patients were not significantly different from that of NUD patients [GU (2.1 _+ l.l)xl04, DU (1.8 _+ 0.6)xl04, NUD (2.6 + 1.1)xl04 copies/0.51ag DNA, respectively]. The sample of one GC patient showed high density of H. pylori colonization (1.2x105 copies/0.5Mg DNA). CONCLUSIONS: Accurate quantitation of H. pylori in biopsy specimens was possible with this PCR system, and it will be useful for 1-1. pylori research.