Validity, sensitivity and resolution limit of the PCR-RFLP analysis of the rrs (16S rRNA gene) as a tool to identify soil-borne and plant-associated bacterial populations

The value of the restriction fragment length polymorphism (RFLP) analysis of polymerase chain reaction (PCR)-amplified rrs (16S ribosomal ribonucleic acid [rRNA] gene) for the rapid identification of bacteria isolated from soil or plants, and the limits of this technique, were evaluated using bacterial genera characteristic of the above-mentioned ecosystems. Results showed that up to two restriction site differences may occur between rrs operons within the same bacterial genome as well as between bacteria belonging to the same genospecies. In spite of these limited differences, members of the same genospecies yield very similar rrs RFLP patterns. The identification limit varies according to the analyzed taxa. Species can be differentiated amongst members of both the family Rhizobiaceae and the genus Stenotrophomonas, while the technique only allows grouping of closely related species amongst Xanthomonas spp. and amongst species related to Psendomonas syringae. On the basis of their rrs RFLP profiles, all presently known species of Agrobacterium can be routinely identified using only the enzymes HpaII (or MspI), AluI and HaeIII. Moreover, the method was also found to be valuable in assessing the biodiversity of a bacterial community isolated from the rhizosphere. From the comparison of empiric rrs RFLPs, published sequences and deoxyribonucleic acid (DNA) pairing studies, we suggest that the occurrence of five different restriction sites within two rrs genes is the minimum difference required to clearly establish that two relevant bacteria belong to different genospecies. (C) Inra/Elsevier, Paris