Synthetic growth phenotypes of Escherichia coli lacking ppGpp and transketolase A (tktA) are due to ppGpp-mediated transcriptional regulation of tktB

Abstract

Many physiological adjustments to nutrient changes involve ppGpp. Recent attempts to deduce ppGpp regulatory effects using proteomics or gene profiling can rigorously identify proteins or transcripts, but the functional significance is often unclear. Using a random screen for synthetic lethals we found a ppGpp-dependent functional pathway that operates through transketolase B (TktB), and which is 'buffered' in wildtype strain by the presence of an isozyme, transketolase A (TktA). Transketolase activity is required in cells to make erythrose-4-phosphate, a precursor of aromatic amino acids and vitamins. By studying tktB-dependent nutritional requirements as well as measuring activities using PtalA-tktB'-lacZ transcriptional reporter fusion, we show positive transcriptional regulation of the talA-tktB operon by ppGpp. Our results show the existence of RpoS-dependent and RpoS-independent modes of positive regulation by ppGpp. Both routes of activation are magnified by elevating ppGpp levels with a spoT mutation (spoT-R39A) defective in hydrolase but not synthetase activity or with the stringent suppressor mutations rpoB-A532Delta or rpoB-T563P in the absence of ppGpp.

Figure 1

Growth properties of transketolase mutants. A. Strains on plate: tktA::Tp (CF13942), ppGpp⁰ i.e., relA256 spoT212 (CF10237), ppGpp⁰ tktA::Tp (CF13926) and tktA tktB (CF13927); B. LB agar after 18 hours; C. Minimal media with glucose and casamino acids after 36 hrs; and D. Minimal media with glucose casamino acids, tryptophan and pyridoxine after 36 hrs. All incubations were at 37°C.

Figure 2

Pathways for biosynthesis of the intermediary metabolite D-erythrose-4-phosphate and amino acids and vitamins derived from it. The enzymes involved at each step are indicated by gene names that encode them; broken arrows represent multiple steps in the pathway; abbreviations, Xly-5-P : D-xylulose-5-phosphate; Rib-5-P : D-ribose-5-phosphate; Gly-3-P : D-Glyceraldehyde-3-phosphate; Sedohep-7-P : Sedoheptulose-7-phosphate; Fruc-6-P : D-fructose-6-phosphate and Ery-4-P : D-erythrose-4-phosphate.

Figure 3

Schematic representation of the genomic neighborhoods of tktA and tktB genes and the DNA segments in each lacZ operon fusion. A. tktA and proximal ORF’s; B. tktB and proximal ORF’s. Open-reading frames are represented by thick filled arrows; P1 and P2 refer to promoters characterized in the intergenic region (Lancour and Landini, 2004); Fusions A, B and C refer to the transcriptional fusions described in materials and methods. Horizontal bracketed lines refer to DNA segments present in each fusion; fusions A and C have identical start points upstream of the talA coding sequence and fusions B and C have identical end points within tktB; the contents of each fusion are described in materials and methods.

Figure 4

Stringent rpoB suppressor mutations increase RpoS protein levels of exponentially growing cells. Extracts were made from LB grown cells taken at different stages of growth. Extracts from cells equivalent to 0.1 A₆₀₀ were used for immunoblotting with anti-RpoS antibody. Lanes, 1, 4 & 7 have extracts from ppGpp⁰ strains CF14276; lanes 2, 5 & 8 from the rpoBT563P derivative CF14278 and lanes 3, 6 & 9 from the rpoBA532Δ derivative CF14277. Log, early stationary phase and stationary phase correspond to A₆₀₀ values of 0.6–0.8; 2–2.2; and 3.5–3.6, respectively.

Figure 5

Regulation of tktB transcription – the role of ppGpp, RpoS and stringent rpoB mutations. tktB transcription was monitored during growth in LB with the talA-tktB′::lacZ fusion C. Strains used are, CF14213 and CF14241 (columns A & B); CF14214 and CF14242 (columns C & D)); CF15008 and CF15023 (columns E & F); CF14277 and CF14281 (columns G & H); CF14278 and CF14280 (columns I & J); For each culture activity was measured in log phase (A₆₀₀ 0.5–0.6), early stationary phase (A₆₀₀ 1.5–2) and stationary phase (A₆₀₀ 2.5–3.5). β-galactosidase specific activities are the mean of three independent experiments expressed in Miller units. In the data table, but not the bar graph, values are rounded to the nearest whole number. a – standard deviation (S.D).

Figure 6

Effect of over-production of dksA on talA-tktB′::lacZ expression. β-galactosidase specific activities are plotted against A₆₀₀ during growth in LB in the presence of plasmid pJK537 or the vector control pBR322 in wild-type (CF14213) or ppGpp⁰ strains (CF14314) (A); the rpoS mutant derivatives of wildtype (CF14241) or ppGpp⁰ strains (CF14242)(B); the activities plotted are mean of three independent experiments expressed in Miller units.

Figure 7

A model for the transcriptional regulation observed in this study for the talA-tktB operon. Two known promoters P1 and P2 upstream of talA are represented by a vertical line. Horizonal bars indicate the talA-tktB operon. Excess DksA refers to over-expression of the protein using plasmid pJK537. Continuous line arrows refer to activation signals; broken line arrows refer to activation signals observed only in the absence of ppGpp.