The arbitrium system controls prophage induction

Abstract

Some Bacillus-infecting bacteriophages use a peptide-based communication system, termed arbitrium, to coordinate the lysis-lysogeny decision. In this system, the phage produces AimP peptide during the lytic cycle. Once internalized by the host cell, AimP binds to the transcription factor AimR, reducing aimX expression and promoting lysogeny. Although these systems are present in a variety of mobile genetic elements, their role in the phage life cycle has only been characterized in phage phi3T during phage infection. Here, using the B. subtilis SPβ prophage, we show that the arbitrium system is also required for normal prophage induction. Deletion of the aimP gene increased phage reproduction, although the aimR deletion significantly reduced the number of phage particles produced after prophage induction. Moreover, our results indicated that AimR is involved in a complex network of regulation and brought forward two new players in the SPβ lysis-lysogeny decision system, YopN and the phage repressor YopR. Importantly, these proteins are encoded in an operon, the function of which is conserved across all SPβ-like phages encoding the arbitrium system. Finally, we obtained mutant phages in the arbitrium system, which behaved almost identically to the wild-type (WT) phage, indicating that the arbitrium system is not essential in the laboratory but is likely beneficial for phage fitness in nature. In support of this, by possessing a functional arbitrium system, the SPβ phage can optimize production of infective particles while also preserving the number of cells that survive after prophage induction, a strategy that increases phage persistence in nature.

Keywords: AimP; AimR; SOS response; SPβ phages; bacteriophage; lysis; lysis/lysogeny; lysogeny; phi3T; repressor.

Graphical abstract

Figure 1

Effect of aimR and aimP mutations on phage titer (A) 168 Δ6 strains lysogenic for phage SPβ WT, ΔaimR, and ΔaimP were MC induced (0.5 μg/mL), and the number of resulting phages were quantified by titering using 168 Δ6 as the recipient strain. The results are represented as the plaque-forming units (PFUs) mL⁻¹. The means and SDs are presented (n = 4). An ordinary one-way ANOVA of transformed data was performed to compare mean differences between SPβ WT, ΔaimR, and ΔaimP titers. Adjusted p values were as follows: SPβ ΔaimR ^∗∗∗∗p ≤ 0.0001; SPβ ΔaimP ^∗p = 0.0115. (B) 168 Δ6 strains lysogenic for phages phi3T WT, ΔaimR, and ΔaimP were MC induced (0.5 μg/mL), and the number of resulting phages were quantified by titering using 168 Δ6 as the recipient strain. The results are represented as PFUs/mL⁻¹. The means and SDs are presented (n = 3). An ordinary one-way ANOVA of transformed data was performed to compare mean differences between SPβ WT, ΔaimR, and ΔaimP titers. Adjusted p values were as follows: SPβ ΔaimR ^∗∗∗∗p ≤ 0.0001; SPβ ΔaimP ^∗∗p = 0.0058. (C) Strain 168 lysogenic for phages SPβ WT, ΔaimR, and ΔaimP were MC induced (0.5 μg/mL), and the number of resulting phages was quantified by titering using 168 Δ6 as the recipient strain. The results are represented as PFUs/mL⁻¹. The means and SDs are presented (n = 3). An ordinary one-way ANOVA of transformed data was performed to compare mean differences between SPβ WT, ΔaimR, and ΔaimP titers. Adjusted p values were as follows: SPβ ΔaimR ^∗∗∗∗p ≤ 0.0001; SPβ ΔaimP ns, not significant.

Figure 2

Phage replication of SPβ WT, ΔaimR, and ΔaimR complemented (A) Strains Δ6 lysogenic for phages SPβ WT, ΔaimR, and ΔaimR complemented with aimR_SPβ were MC induced (0.5 μg/mL), and 1 mL of each culture at different time points after induction was collected. Samples were loaded in a 0.7% agarose gel, Southern blotted, and probed for phage DNA. (B) Strains 168 lysogenic for phages SPβ WT, ΔaimR, and ΔaimR complemented with aimR_SPβ were MC induced (0.5 μg/mL), and 5 mL of each culture at different time points after induction was collected. Samples were loaded in a 0.7% agarose gel, Southern blotted, and probed for phage SPβ DNA.

Figure 3

Schematic representation of the SPβ and phi3T arbitrium and operon genetic layout Diagram shows the genetic organization of the arbitrium genes, aimR and aimP, followed by the operon directly downstream. Colors denote putative functions according to BLAST results; orange, arbitrium genes; light yellow, sRNAs; gray, unknown function; navy blue, HTH_XRE domain; green, integrase domain; purple, ParB domain; light blue, putative repressor. The mutations obtained during the evolution experiments are marked. Shown was created with BioRender.com. See also Figures S1 and S7 and Tables S1 and S2.

Figure 4

Titer and lysogenization of SPβ WT, ΔaimR, ΔaimP, and evolved phages Strains lysogenic for phages SPβ WT, ΔaimR, and evolved aimR phages were MC induced (0.5 μg/mL). (A) The number of resulting phages were quantified using 168 Δ6 as the recipient strain. The results are represented as PFUs mL⁻¹. The means and SDs are presented (n = 3). An ordinary one-way ANOVA of transformed data was performed to compare mean differences between titers. Adjusted p values were as follows: SPβ ΔaimR ^∗∗∗∗p ≤ 0.0001; YopN^L90S and YopN^L46P ns; YopN^A49∗∗p = 0.0324. (B) The number of resulting lysogens were quantified using 168 Δ6 as the recipient strain. The results are represented as colony-forming units (CFUs) mL⁻¹ normalized by PFUs per milliliter and represented as the log CFU of an average phage titer (1 × 10⁹ PFUs). The means and SDs are presented (n = 3). An ordinary one-way ANOVA of transformed data was performed to compare mean differences in lysogenization. Adjusted p values were as follows: SPβ ΔaimR ^∗∗∗∗p ≤ 0.0001; YopN^L46P and YopN^A49∗ ns. See also Figure S3 and Table S1.

Figure 5

Titer of SPβ WT, ΔaimR, ΔyopN, and double mutant ΔaimR-yopN Strains lysogenic for phages SPβ WT, ΔaimR, ΔyopN, and ΔaimR-yopN were MC induced (0.5 μg/mL). The number of resulting phages were quantified using 168 Δ6 as the recipient strain. The results are represented as PFUs mL⁻¹. The means and SDs are presented (n = 5). An ordinary one-way ANOVA of transformed data was performed to compare mean differences between titers. Adjusted p values were as follows: SPβ WT versus SPβ ΔaimR ^∗∗∗∗p ≤ 0.0001; SPβ ΔyopN ns; SPβ ΔyopN ΔaimR ^∗∗p = 0.0010. SPβ ΔaimR versus SPβ ΔyopN ΔaimR ^∗p = 0.0351. See also Figures S2, S3, and S5.

Figure 6

Titer and lysogenization of SPβ WT and yopR mutant Strains lysogenic for phages SPβ WT and SPβ amyE::Pspank-YopR yopR::ermR were MC induced (0.5 μg/mL). (A) The number of resulting phages were quantified using 168 Δ6 or 168 Δ6 amyE::Pspank-YopR as the recipient strain. The results are represented as PFUs mL⁻¹. The means and SDs are presented (n = 3). (B) The number of resulting lysogens were quantified using 168 Δ6 or 168 Δ6 amyE::Pspank-YopR as the recipient strain. The results are represented as CFUs mL⁻¹. The means and SDs are presented (n = 3). An ordinary one-way ANOVA of transformed data was performed to compare mean differences between SPβ lysogen titers obtained using 168 Δ6 or 168 Δ6 amyE::Pspank-YopR recipient strains (adjusted p = 0.0171). (C) The lysates were titered using 168 Δ6 as the recipient strain. The resulting plaque morphologies were photographed. Shown was created with BioRender.com. See also Figure S5.

Figure 7

Growth curves of SPβ WT, ΔaimR, and ΔaimP after MC induction Strains lysogenic for phages SPβ WT, ΔaimR, and ΔaimP were MC induced (0.5 μg/mL). Optical density 600 nm (OD_600nm) was monitored over time, and cells were collected at time points 0, 2, 4, 6, and 8 h. The means and SDs are presented (n = 3). A two-way ANOVA was performed to compare mean differences in OD_600nm values. Adjusted p values were as follows: time 4 h SPβ ΔaimR ^∗∗∗∗p ≤ 0.0001, SPβ ΔaimP ^∗∗p = 0.0077; time 6 h SPβ ΔaimR ^∗∗∗∗p ≤ 0.0001, SPβ ΔaimP ^∗∗p = 0.0085; time 8 h SPβ ΔaimR ^∗∗∗∗p ≤ 0.0001, SPβ ΔaimP ^∗p = 0.0226.