Phylogenetic relationships of Pseudomonas putida strains deduced from the nucleotide sequences of gyrB, rpoD and 16S rRNA genes

Phylogenetic analysis of 20 Pseudomonas strains (Pseudomonas putida, Pseudomonas fluorescens and Pseudomonas chlororaphis) was conducted by using the nucleotide sequences of the genes for 16S RNA, DNA gyrase B subunit (gyrB) and RNA polymerase delta 70 factor (rpoD), which have been determined by the direct sequencing of PCR-amplified fragments. On the basis of gyrB and rpoD sequences, these strains were split into two major clusters: one including the type strain of P. putida and all biovar A strains and the other including all P. putida biovar B strains, P. fluorescens stains and the P. chlororaphis strain. In the phylogenetic tree reconstructed from the 16S rRNA sequences included variable regions, P. Putida biovar A and B strains were not separated into two independent clusters, whereas in the phylogenetic tree reconstructed from the 16S rRNA sequences excluding the variable region sequences, these strains were separated into P. putida biovar A and biovar B clusters. The pairwise distances estimated from the variable regions of 16S rRNA correlated poorly with the synonymous distances estimated from the gyrB and rpoD genes. On the other hand, a highly significant correlation was observed between the pairwise distances estimated from the non-variable regions of 16S rRNA and the synonymous distances from gyrB and rpoD genes. Consequently, only the 16S rRNA sequences in the non-variable regions should be used for the phylogenetic analysis. The gyrB and rpoD analyses showed the necessity for the reclassification of P. putida biovar B strains.