AIM:H pylori genomes are highly diversified.This projectwas designed to genotype H pyloriisolates by the polymerasechain reaction (PCR)-based randomly amplified polymorphicDNA (RAPD) fingerprinting technique and to verify its stabilityby Southern blotting and DNA sequencing.METHODS:Clinical isolates of Hpyloriwere cultured fromgastric antra and cardia of 73 individuals,and genomic DNAwas prepared for each isolate.RAPD was carried out underoptimized conditions.23S rDNA was regarded as an internalcontrol,and a 361 bp rDNA fragment (RDF) was used as aprobe to screen the RAPD products by Southern blotting.Ten RDFs from different clinical isolates and the flankingregions (both upstream and downstream) of four RDFs wereamplified and sequenced.RESULTS:H pylori isolates from different individuals haddifferent RAPD profiles,but the profiles for isolates culturedfrom different gastric sites of a given individual were identicalin all but one case.Isolates from 27 individuals were RDFpositive by Southern blotting.Sequences of the RDFs andtheir flanking regions were almost the same between theRDF positive and negative isolates as determined by Southernblotting.There was no binding site for random PCR primerinside the sequences.CONCLUSION:RAPD is very useful in genotyping H pylorigrossly on a large scale.However,it seems unstable inamplification of low yield fragments,especially those thatdo not appear as visible bands on the agarose gel stainedwith EB,since the primer is partially matched to the template. AIM: H pylori genomes are highly diversified.This projectwas designed to genotype H pyloriisolates by the polymerasechain reaction (PCR) -based randomly amplified polymorphicDNA (RAPD) fingerprinting technique and to verify its stabilityby Southern blotting and DNA sequencing.METHODS: Clinical isolates of Hpyloriwere cultured fromgastric antra and cardia of 73 individuals, and genomic DNAwas prepared for each isolate. RAPD was carried out under optimized conditions.23S rDNA was viewed as an internal control, and a 361 bp rDNA fragment (RDF) was used as aprobe to screen the RAPD products by Southern blotting. Ren RDFs from different clinical isolates and the flanking regs (both upstream and downstream) of four RDFs were amplified and sequenced .RESULTS: H pylori isolates from different individuals haddifferent RAPD profiles, but the profiles for isolates culturedfrom different gastric sites of a given individual were identicalin all but one case. Isolates from 27 individuals were RDFpositive by S outhern blotting. Sequences of the RDFs and the flanking regions were almost the same between the RDF positive and negative isolates as determined by Southern blotting. There was no binding site for random PCR primerinside the sequences. CONCLUSION: RAPD is very useful in genotyping H pylorigrossly on a large scale.However, it seems unstable inamplification of low yield fragments, especially those that do not appear as visible bands on the agarose gel stainedwith EB, since the primer is partially matched to the template.