Fluctuations in spo0A transcription control rare developmental transitions in Bacillus subtilis

Abstract

Phosphorylated Spo0A is a master regulator of stationary phase development in the model bacterium Bacillus subtilis, controlling the formation of spores, biofilms, and cells competent for transformation. We have monitored the rate of transcription of the spo0A gene during growth in sporulation medium using promoter fusions to firefly luciferase. This rate increases sharply during transient diauxie-like pauses in growth rate and then declines as growth resumes. In contrast, the rate of transcription of an rRNA gene decreases and increases in parallel with the growth rate, as expected for stable RNA synthesis. The growth pause-dependent bursts of spo0A transcription, which reflect the activity of the spo0A vegetative promoter, are largely independent of all known regulators of spo0A transcription. Evidence is offered in support of a "passive regulation" model in which RNA polymerase stops transcribing rRNA genes during growth pauses, thus becoming available for the transcription of spo0A. We show that the bursts are followed by the production of phosphorylated Spo0A, and we propose that they represent initial responses to stress that bring the average cell closer to the thresholds for transition to bimodally expressed developmental responses. Measurement of the numbers of cells expressing a competence marker before and after the bursts supports this hypothesis. In the absence of ppGpp, the increase in spo0A transcription that accompanies the entrance to stationary phase is delayed and sporulation is markedly diminished. In spite of this, our data contradicts the hypothesis that sporulation is initiated when a ppGpp-induced depression of the GTP pool relieves repression by CodY. We suggest that, while the programmed induction of sporulation that occurs in stationary phase is apparently provoked by increased flux through the phosphorelay, bet-hedging stochastic transitions to at least competence are induced by bursts in transcription.

Figure 1. Growth and transcription from spoIIG, spo0A, and abrB promoter fusions to luc in DSM medium.

(A) The gray curve depicts growth measured by optical density at 600 nm. The gray line connects the OD points, which are not visible, but were taken every 1.5 minutes. The arrows point to inflection points in the growth curve. The open triangles show relative luminescence readings corrected for OD for the spo0A promoter and the heavy line shows the same for the spoIIG promoter. The ordinate scale to the left (×10⁵) refers to the spoIIG promoter. The growth curves of these two strains are nearly identical (not shown). T₀ is indicated by the downward pointing arrow. (B) Growth rate and transcription from the abrB promoter. Strains expressing luc from a promoter fusion to abrB were grown in DSM. The open circles indicate results for the spo0A ⁺ background and the solid gray squares for a spo0A strain. The growth rate was numerically determined and is indicated by the gray line. Slightly negative values for growth rate indicate some lysis that occurred after T₀. T₀ is indicated by a downward pointing arrow for each strain, black for the wild type and grey for the spo0A mutant.

Figure 2. Comparison of spo0A and rrnB transcription in DSM.

(A) The luminescence data from Figure 1 for spo0A from the start of the experiment until T₀ are replotted (open triangles) together with numerically determined values for the growth rate (gray line). (B) Luminescence data from a luciferase fusion to the promoter of rrnB is plotted for the first four hours of growth, together with the numerically determined values for the growth rate (gray line).

Figure 3. Diauxie and expression from the spo0A promoter.

A strain expressing luc from the spo0A promoter (PP530) was grown in S7 minimal medium supplemented with (A) 0.5 mg/ml glucose+2.0 mg/ml arabinose, (B) 1 mg/ml glucose+1.5 mg/ml arabinose and (C) 2.5 mg/ml glucose with no arabinose. The open triangles show Pspo0A-luc expression and the thick line shows growth.

Figure 4. Effects of regulatory gene mutations on spo0A transcription.

In panels A–C the heavy lines with open triangles show luminescence during growth in DSM from the Pspo0A promoter. The gray lines show results for knockout mutants in sigH (A), spo0A (B) and codY (C). The black and gray arrows indicate T₀ for the wild type and mutant strains respectively. In the case of the spo0A mutant, T₀ is reached slightly earlier in accordance with the slightly earlier occurrence of the major bursts.

Figure 5. Bursts in spo0A transcription result in comK bursts.

(A) The heavy black line shows expression of spo0A-luc during growth in DSM. The light gray line shows the expression of comK-luc in DSM and the medium gray line shows the expression of comK-luc in the spo0A mutant background. (B) A selected microscope field visualized by merging a DIC and a fluorescent image, revealing a single comK-gfp-expressing cell among several hundred. The sample was taken from the strain BD5486 (Pspo0A::luc cm, PcomK-gfp::tet) growing in the plate reader immediately after the second burst in spo0A transcription.

Figure 6. Effects of the AA to GG mutation.

(A) Sequence of the promoter region of spo0A showing the location of the initiating residues. (B) The dark gray line shows transcription from the Pv promoter of spo0A when the AA initiating residues have been changed to GG. The black line shows the same for the wild-type strain and the light gray line shows growth rates for the mutant.

Figure 7. Transcription from a core spo0A promoter.

(A) The sequence of the synthetic core promoter is shown including the start site for transcription. The ribosomal binding site sequence from spoVG was inserted downstream from +1. (B) A strain with P_spo0A-core-luc inserted at amyE was grown in DSM. Light output (dark gray) and growth rate (light gray) were monitored.

Figure 8. Effects of ppGpp synthase mutations on spo0A transcription.

In both panels the black lines show spo0A expression in a wild-type background and the gray lines show results from mutant strains, as follows: (A) relA (B) relA yjbM ywaC.