DksA is required for growth phase-dependent regulation, growth rate-dependent control, and stringent control of fis expression in Escherichia coli

Abstract

DksA is a critical transcription factor in Escherichia coli that binds to RNA polymerase and potentiates control of rRNA promoters and certain amino acid promoters. Given the kinetic similarities between rRNA promoters and the fis promoter (Pfis), we investigated the possibility that DksA might also control transcription from Pfis. We show that the absence of dksA extends transcription from Pfis well into the late logarithmic and stationary growth phases, demonstrating the importance of DksA for growth phase-dependent regulation of fis. We also show that transcription from Pfis increases with steady-state growth rate and that dksA is absolutely required for this regulation. In addition, both DksA and ppGpp are required for inhibition of Pfis promoter activity following amino acid starvation, and these factors act directly and synergistically to negatively control Pfis transcription in vitro. DksA decreases the half-life of the intrinsically short-lived fis promoter-RNA polymerase complex and increases its sensitivity to the concentration of CTP, the predominant initiating nucleotide triphosphate for this promoter. This work extends our understanding of the multiple factors controlling fis expression and demonstrates the generality of the DksA requirement for regulation of kinetically similar promoters.

FIG. 1.

Effect of dksA on growth phase-dependent regulation of Pfis. (A) Schematic representation of the fis operon, consisting of two genes (dusB and fis) that are transcribed by a single promoter (Pfis). The approximate annealing region of the oligonucleotide oRO109 used in the primer extension assays is represented with an arrow. Beneath is the sequence of the minimal Pfis promoter. The operationally defined −10 and −35 promoter sequences (57) are boxed, and the major and minor transcription start sites are indicated with a large and small arrow, respectively. (B) Effect of DksA on GPDR of Pfis. Primer extension assays were performed to detect Pfis transcripts synthesized from the chromosome using total RNA obtained from RO1275 (WT) or RO1276 (ΔdksA) at various times of outgrowth from sta- tionary phase in LB medium. The signals corresponding to the two fis mRNA start sites (+1C and −2G) and for the bla mRNA produced in these strains are shown with arrows. (C) Effect of dksA on the fis mRNA expression pattern. Primer extension results similar to those shown in panel B were quantified by phosphorimaging, normalized to the amount of cells used (as determined by OD₆₀₀ readings) in each reaction, and plotted relative to the maximum value in RO1275, which was assigned a value of 100. The results from two experiments were averaged, with standard deviations being within 30% or less of the values. Relative fis mRNA levels are shown for RO1275 (dksA⁺) (○) and RO1276 (dksA) (•); growth curves based on OD₆₀₀ measurements are shown for dksA⁺ (▵) and dksA (▴) strains. (D) Effect of dksA on β-galactosidase activity from a Pfis-lacZ fusion during growth in LB medium. Saturated cultures of RO1261 (dksA⁺) (○, ▵) and RO1265 (dksA) (•, ▴) were diluted to an OD₆₀₀ of 0.06 in LB and grown at 31°C. β-Galactosidase assays (○, •) were performed at the indicated times during growth. Results are averages from three independent cultures; standard deviations were within 14% of the average values. Growth curves for both strains (▵, ▴) are shown.

FIG. 2.

Effect of dksA on growth rate-dependent control of Pfis. (A) Transcription activity from Pfis was monitored by β-galactosidase assays in strains RO1261 (WT for dksA; open symbols) and RO1265 (ΔdksA; filled symbols) grown in different media at 31°C to generate different growth rates. Growth media used were (□, ▪) M9 salts, 0.4% glucose, and 40 μg/ml each of methionine, aspartic acid, and threonine, (▵, ▴) M9 salts, 0.4% glycerol, and 0.4% Casamino Acids, (○, •) M9 salts, 0.4% glucose, and 0.4% Casamino Acids, and (⋄, ⧫) LB. Results are averages of three independent cultures, with standard deviations indicated by error bars. (B) The results from panel A are displayed in a bar graph to compare the effects of dksA on Pfis transcription in each growth medium used. Open bars, RO1261; filled bars, RO1265.

FIG. 3.

Effect of dksA on stringent control of Pfis. (A) Primer extension assays of fis mRNA under starved and nonstarved conditions. At the indicated times after SH addition, samples of the SH-treated and untreated (control) cell cultures were harvested and used to prepare total RNA. Primer extension reactions were performed to detect both the fis and bla mRNA signals. The two Pfis transcript signals correspond to those initiating at +1C and −2G; two prominent transcript signals were also detected for the bla transcript. Strains used in this experiment were RO1275 (WT for dksA), RO1276 (ΔdksA), and RO1279 (ΔrelA ΔspoT). Duplicate reactions similar to those represented in panel A for (B) dksA WT (WT), (C) ΔdksA, and (D) ΔrelA ΔspoT strains were quantified by phosphorimaging and averaged and are shown relative to the values at time zero in each set, which were assigned a value of 1.0. Error bars indicate standard deviations. Open circles, relative fis transcript levels in the control cultures (-SH); filled circles, relative fis transcript levels in the serine hydroxamate-treated cultures (+SH).

FIG. 4.

Effect of ppGpp and DksA on Pfis transcription in vitro. Duplicate reactions of multiple round in vitro transcription assays were performed with pRO362 in the absence or presence of 4 μM purified DksA and 200 μM ppGpp, as indicated. Transcripts from Pfis and the RNA-1 promoter (derived from the plasmid) are indicated. The Pfis signals from two independent reactions were averaged and are shown as a fraction of the Pfis activity in the absence of DksA and ppGpp.

FIG. 5.

Effect of DksA on the stability of RNAP complexes with Pfis. Pfis complexes with RNAP were preformed with supercoiled pRO362 in the presence (▿) or absence (•) of 2 μM DksA in transcription buffer containing 100 mM KCl as described in Materials and Methods. The complexes were challenged with heparin and assayed for single-round transcription at various times thereafter. Electrophoretically separated transcripts were quantified by phosphorimaging and are shown relative to the amount present prior to the heparin addition.

FIG. 6.

Effect of DksA on promoter sensitivity to CTP. Multiple-round in vitro transcription assays were performed with supercoiled pRO362 in transcription buffer containing 200 mM KCl and various concentrations of ATP (○) or CTP (•, ▿), in the absence (○, •) or presence (▿) of 6 μM DksA, as described in Materials and Methods. fis transcripts were electrophoretically separated and quantified by phosphorimaging and are shown relative to the maximal levels obtained at the highest NTP concentrations used.