Cloning, Sequencing, and Phenotypic Characterization of the rpoS Gene from Pseudomonas putida KT2440

Multiple alignments of the P. putidaORF2 sequence with homologous sequences derived from the nonredundant GenBank CDS translations +PDB+SwissPot+PIR. An alignment with RpoS proteins from members of the pseudomonads and enteric bacteria is shown. The sequences of the RpoS proteins of P. fluorescens (P.f.), P. aeruginosa(P.a.), E. coli (E.c.), and S. typhimurium (S.t.) were derived from Samiguet et al. (37), Tanaka and Takahashi (43), Swiss-Prot entry 13445, and Swiss-Prot entry 37400, respectively. TheP. putida (P.p.) RpoS sequence (ORF2) was determined in this study. ∗, amino acid conserved in all sequences; ., residue that belonged to the same group in all sequences (neutral changes). Changes of amino acids in the following groups were considered neutral: (i) A, G, P, S, and T; (ii) D, N, E, and Q; (iii) R, H, and K; (iv) I, L, M, and V; (v) F, Y, and W; and (vi) C. For consensus 1, pseudomonad RpoS was used as reference; for consensus 2, both pseudomonad RpoS and enteric RpoS were used.

Replacements of the rpoS gene with theluxAB insertion mutant rpoS gene. A single homologous recombination event between the functional rpoSpresent on the chromosome of P. putida KT2440 and the inactivated rpoS present on pMIR592 was isolated by selection for resistance to streptomycin (pMIR592 [Fig. 1]). One of the Sm^r (see the footnote to Table 1 for abbreviations) transconjugants was selected and named R6C1; this merodiploid strain contained the entire plasmid pMIR592 integrated in the genome. A second crossover event at the rpoS locus was selected by cultivating R6C1 overnight in LB without streptomycin (about 10 generations) and subsequent plating on LB medium supplemented with 10% sucrose. Suc^r colonies were analyzed by replica plating. One of the Suc^r Sm^s Lux⁺ colonies was called C1R1. The genomes of three merodiploid isolates (from three independent matings) obtained as the result of the first recombination event and the genomes of six clones (two from each merodiploid) obtained as the result of the resolution of the merodiploids after the second recombination event were examined by Southern blot analysis, which revealed correct single and double recombination events, respectively, at the rpoS locus (not shown).

Growth-dependent expression of the rpoS genes of P. putida R6C1 (○, •), and P. putida C1R1 (□, ▪). Exponentially growing cells in 10 mM citrate-supplemented AB minimal medium were diluted in the same medium, and growth (open symbols) and emission of light (solid symbols) were measured. Measurement of luminescence in liquid culture was carried out as described previously (34). LU values are not normalized per cell. Notice that the same range of log units, four, is plotted in both y axes, and therefore the curve slopes of light emission and optical density are comparable. Experiments were repeated with three cultures; results of a typical experiment from a single culture are presented. Duplicate measurements of LU in a single experiment yielded an average standard deviation of 10%.

Long-term survival of P. putida strains. Carbon starvation of P. putida KT2440 (○), R6C1 (▵), and C1R1 (□) cells was accomplished through exhaustion of 1 mM citrate present in AB minimal medium as described in Materials and Methods. Exponentially growing cells in 10 mM citrate-supplemented AB minimal medium were centrifuged, the supernatant was discarded, and the cells were resuspended in 1 mM citrate-supplemented AB medium up to 3 × 10⁷ to 4 × 10⁷ cells per ml. Time zero was defined in day 0 as cultures reached stationary phase. Survival of the starved cultures was monitored by determination of viable counts on LB plates, supplemented with streptomycin in the case of R6C1. Each starvation condition was repeated at least twice with two cultures each time. Means and standard deviations of duplicate experiments with the same cultures are plotted. Some of the error bars are too small to be distinguished.

Stress challenges of carbon-starved cultures ofP. putida KT2440 (○), R6C1 (▵), and C1R1 (□). Starvation (for 48 h) was accomplished by exhaustion of 1 mM citrate from minimal medium M9. (A) Challenge with 18% (vol/vol) ethanol; (B) challenge with 200 μM H₂O₂; (C) challenge with 2.4 M NaCl. Viable counts present in each of the prechallenge samples (time zero) were normalized to 1. Survival was determined as relative viable counts. Each challenge experiment was repeated at least twice with two cultures each time. Means and standard deviations of duplicate experiments with the same cultures are plotted. Some of the error bars are too small to be distinguished.

2D-PAGE autoradiograms of carbon starvation-induced proteins of P. putida KT2440 (A) and P. putida C1R1 (B) 60 min after removal of the carbon source. Cells were cultivated in 10 mM citrate-supplemented AB minimal media. Spots positively (not present in panel B) and negatively (not present in panel A) dependent on rpoS are enclosed in boxes and stars, respectively. Molecular mass decreases from top to bottom (100 to 5 kDa), and pH decreases from left to right.

Expression from bolAp₁ andficp in P. putida KT2440 (A), P. putida R6C1 (B), and C1R1 (C) bearing either pGM112 (▵, ▴), pGM115 (□, ▪), or pGM118 (○, •). P. putidacells were electrotransformed with plasmids pGM112 (bolAp₁), pGM115 (ficp), and pGM118 (control vector, no promoter), and expression from the plasmids carrying bolAp₁ and ficp was studied in growing cultures. Exponentially growing LB cultures were used as inoculates, and cells were grown in the same LB medium supplemented with the appropriate antibiotics. OD₆₀₀ (open symbols) and specific β-galactosidase (B∼gal) activities (solid symbols) were monitored. β-Galactosidase activities are not the result of subtracting the background levels that are synthesized from the vector pGM118 alone. Duplicate measurements of β-galactosidase activities in a single experiment yielded an average standard deviation of 5%.