Analysis of (p)ppGpp metabolism and signaling using a dynamic luminescent reporter

Abstract

As rapidly growing bacteria begin to exhaust nutrients, their growth rate slows, ultimately leading to the non-replicative state of quiescence. Adaptation to nutrient limitation requires widespread metabolic remodeling that is in part mediated by the phosphorylated nucleotides guanosine tetra- and penta-phosphate, collectively (p)ppGpp. We have developed a novel reporter of (p)ppGpp abundance in the Gram-positive bacterium Bacillus subtilis based on the recent identification of a riboswitch that binds (p)ppGpp and modulates transcription via regulation of a transcriptional terminator. Placement of an unstable reporter, firefly luciferase, downstream of the riboswitch allows for sensitive and dynamic assessment of (p)ppGpp. We first confirm that the reporter accurately reflects (p)ppGpp abundance in a variety of well-established conditions. We then proceed to use it to demonstrate the physiological importance of several mechanisms of regulation of (p)ppGpp metabolism previously observed only in vitro including allosteric interactions between (p)ppGpp synthesis enzymes and the hydrolytic activity of a (p)ppGpp synthetase. (p)ppGpp signaling has been implicated in the regulation of gene expression, and we demonstrate a close temporal association between gene expression and (p)ppGpp abundance, indicating a rapid, and therefore likely direct mechanism of (p)ppGpp dependent gene activation. Thus, this reporter provides a new, comprehensive analysis of (p)ppGpp signaling in vivo and offers the potential ability to sensitively monitor the temporal dynamics of (p)ppGpp abundance under diverse environmental conditions.

Fig 1. RsFluc (p)ppGpp-sensitive firefly luciferase reporter.

A, schematic of the (p)ppGpp reporter RsFluc depicting the inducible promoter P_hyperspank fused to the (p)ppGpp-senstive riboswitch of D. hafniense ilvE followed by the gene encoding firefly luciferase. B, luminescence (RLU/OD₆₀₀) of RsFluc (blue; JDB4496), mutant RsFluc^mut (green; JDB4631) in wildtype backgrounds and RsFluc in a ∆relA∆sasA∆sasB background (red; JDB4512), C) ppGpp concentration determined by HPLC-MS (black closed circles) and corresponding reporter luminescence corrected for (p)ppGpp⁰ luminescence by subtraction (blue open circles). D, luminescence per OD₆₀₀ of RsFluc (black), RsFluc^natA (purple, JDB4730) and RsFluc^livK (cyan, JDB4731). Shown are representative examples of at least three biological replicates, each comprising three technical replicates.

Fig 2. RsFluc reporter response to stimulation of (p)ppGpp synthesis.

Luminescence (RLU/OD₆₀₀) of: A, RsFluc (black; JDB4496) and RsFluc^mut (blue; JDB4631) after amino acid downshift at T₆₀ (dashed line); B, RsFluc (black) and RsFluc^mut (blue) after 0 and 60 mins of 50 ng/mL mupirocin treatment. Significance was determined by two-tailed t test: p-value < .05 for RsFluc, no significant difference for RsFluc^mut; C, RsFluc grown in defined minimal media containing either 2.5 mg/mL of glucose (black) or 0.5 mg/mL glucose and 2.0 mg/mL arabinose (blue). Respective OD₆₀₀ measurements are shown in grey and light blue, respectively, and D) RsFluc (black) and a P_purE luciferase promoter fusion in wildtype (orange; JDB4803) and (p)ppGpp⁰ (purple; JDB4804) backgrounds. Shown are representative examples of at least three biological replicates, each comprising three technical replicates.

Fig 3. Relative contributions of (p)ppGpp synthetases to dynamics of RsFluc activity.

Luminescence (RLU/OD₆₀₀) of: A, RsFluc in wildtype (black, JDB4496), ∆relA∆sasA∆sasB (red, JDB4512), ∆sasA (green, JDB4515), ∆sasB (orange, JDB4516), and relA-D264G (blue, JDB4741) backgrounds and B), wildtype (black), ∆sasA (green), ∆sasB (orange), ∆sasA∆sasB (brown, JDB4511), and relA-D264G (blue) strains after amino acid downshift at T₆₀ (dashed line). Shown are representative examples of at least three biological replicates, each comprising three technical replicates.

Fig 4. RsFluc activity coordinated by allosteric network.

Luminescence (RLU/OD₆₀₀) of: A, RsFluc in wildtype (black, JDB4496), relA-Y200A (fuschia, JDB4528), and sasB-F42A (blue, JDB4711) strains and B, RsFluc in wildtype (black), relA-Y200A (fuschia), and sasB-F42A (blue) strains after amino acid downshift (dashed line). Shown are representative examples of at least three biological replicates, each comprising three technical replicates.

Fig 5. RelA hydrolase activity contributes to dynamics of RsFluc activity.

A, schematic of bacitracin-inducible relA constructs. B) luminescence (RLU/OD₆₀₀) of RsFluc in wildtype (black, JDB4496), ∆relA∆sasA∆sasB with P_liaI-relA (blue, JDB4675), ∆relA∆sasA∆sasB with P_liaI-relA-D78A (red, JDB4676) and ∆relA∆sasA∆sasB (gray, JDB4512) strains. RelA expression induced by 5 µg/mL bacitracin at T₉₀ (dashed line). Shown is a representative example of at least three biological replicates, each comprising three technical replicates.

Fig 6. RsFluc activity and amino acid availability.

Luminescence (RLU/OD₆₀₀) of: A, RsFluc in prototroph (JDB4656) cultured in S7 supplemented with 0.005% (blue), 0.01% (black), and 0.02% (green) casamino acids and B, RsFluc (black) and RsFluc^mut (blue) in prototroph background without amino acid supplementation. Shown are representative examples of at least three biological replicates, each comprising three technical replicates.

Fig 7. Correlation of amino acid biosynthetic gene expression and RsFluc activity.

Luminescence (RLU/OD₆₀₀) of: A, RsFluc (black, JDB4496), P_serA-luc reporter (blue, JDB4759), P_ilvB-luc reporter (green, JDB4792), and P_metE-luc reporter (red, JBD4798) in the wildtype background and B, reporters as A in the relA-D264G genetic background. Shown are representative examples of at least three biological replicates, each comprising three technical replicates.

Fig 8. Temporal relationship between protein synthesis and RsFluc activity.

A, luminescence (RLU/OD₆₀₀) of RsFluc (blue, JDB4496) and growth rate (black). Growth rate/hour (µ) at a given time (t) is defined as log₂[OD₆₀₀(t)-OD₆₀₀ (t_-60)]. B, representative composite (fluorescent, phase) images of cells at TP1 and TP2 (see S11 Fig) in either wildtype (JDB4486) or ∆relA∆sasA∆sasB (‘(p)ppGpp⁰’; JDB4512) backgrounds. C) and D) quantification of images from B.