Differential gene content among phlD+ genotypes.

In order to determine whether specific subsets of DAPG producing Pseudomonas spp. are involved in soil suppressiveness, over 300 phlD+ isolates including reference strains from a worldwide collection, isolates from soils with a history of at least 5 years monoculture of wheat or flax, and isolates from the pea wilt suppressive soil were analyzed by restriction analysis of 16S ribosomal DNA and genomic fingerprinting. The genomic fingerprinting by RAPD, rep-PCR, and phlD-RFLP analyses have resolved at least 17 different genotypes phlD+ genotypes (A-Q) within American and European collections (1). The groupings defined by different methods have been largely consistent, indicating clonal growth in soils from different locations.

Although P.fluorescens Q8r1-96 and other BOX-PCR group D strains have unique colonization properties, they are virtually indistinguishable from members of most other phlD+ BOX-PCR groups in culture, suggesting that relatively few differences at the genetic level might be sufficient to confer the colonization phenotype typical of premier PGPR. As a first approach to identifying novel genes associated with this trait, we are using a PCR-based suppression subtractive hybridization (SSH) technique (2) to isolate unique DNA fragments present in strain P.fluorescens Q8r1-96 (the D genotype "tester") but not P.fluorescens Q2-87 (the B genotype "driver"). Genomic subtraction, which involves the removal of sequences common to two genomes and allows the direct cloning of strain-specific genes, is among the best methods currently available for exploring differences between closely related bacterial genomes. As a result of genomic subtraction with Q8r1-96 vs Q2-87 we identified 32 unique subtracted (SSH) clones.

Homology searches revealed that 28 of the 32 selected sequences contain open reading frames of unknown function. Twelve sequences (SSH3, 18, 25, 45, 53, 58, 64, 81, 101, 103, and 164) resemble conserved hypothetical ORFs of unknown function reported in other bacterial species, whereas 16 others (SSH5, 8, 12, 13, 26, 32, 36, 41, 62, 66, 74, 80, 83, 93, 132, and 133) have no significant matches with known sequences. Three other clones, SSH61, 85, and 127, resemble putative regulatory genes present in other Pseudomonas genomes, and one, SSH6, has significant protein sequence similarity with E. coli colicin M (see table below).

Sequence analysis of some Q8r1-96-specific loci

Of the clones with similarity to known bacterial proteins, SSH61 encodes a predicted protein resembling proteins related to the global response regulator Lrp from E. coli. Among proteins with the highest similarity to the product of SSH61, BkdR from P. putida (accession No. P14279) has been characterized and shown to activate the expression of genes for the branched-chain keto acid dehydrogenase complex (3). The SSH61 gene product could regulate the synthesis or utilization of certain amino or keto acids and thus affect the ability of pseudomonads to survive in the rhizosphere.
Clones SSH85 and SSH127 encode a putative two-component sensor kinase similar to the predicted sensor kinase PA1396 from P. aeruginosa PAO1, and a putative regulator with similarity to proteins of the LysR family of transcriptional regulators, respectively.
Clone SSH6 exhibited similarity to colicin M, a pore-forming protein from E. coli that inhibits the biosynthesis of peptidoglycan and O-antigen, resulting in autolysis of sensitive cells. Bacteriocin production is common among Gram-negative and Gram-positive bacteria, but the predicted product of SSH6 is distinct from other bacteriocins produced by fluorescent pseudomonads and may be the first example of a Pseudomonas bacteriocin similar to colicin M. Based on the ecological role of colicins (4), we speculate that SSH6 may play a role in intraspecific interactions or contribute to the competitiveness of Q8r1-96 in the environment.

References

 | Top |