The RapP-PhrP quorum-sensing system of Bacillus subtilis strain NCIB3610 affects biofilm formation through multiple targets, due to an atypical signal-insensitive allele of RapP

Abstract

The genome of Bacillus subtilis 168 encodes eight rap-phr quorum-sensing pairs. Rap proteins of all characterized Rap-Phr pairs inhibit the function of one or several important response regulators: ComA, Spo0F, or DegU. This inhibition is relieved upon binding of the peptide encoded by the cognate phr gene. Bacillus subtilis strain NCIB3610, the biofilm-proficient ancestor of strain 168, encodes, in addition, the rapP-phrP pair on the plasmid pBS32. RapP was shown to dephosphorylate Spo0F and to regulate biofilm formation, but unlike other Rap-Phr pairs, RapP does not interact with PhrP. In this work we extend the analysis of the RapP pathway by reexamining its transcriptional regulation, its effect on downstream targets, and its interaction with PhrP. At the transcriptional level, we show that rapP and phrP regulation is similar to that of other rap-phr pairs. We further find that RapP has an Spo0F-independent negative effect on biofilm-related genes, which is mediated by the response regulator ComA. Finally, we find that the insensitivity of RapP to PhrP is due to a substitution of a highly conserved residue in the peptide binding domain of the rapP allele of strain NCIB3610. Reversing this substitution to the consensus amino acid restores the PhrP dependence of RapP activity and eliminates the effects of the rapP-phrP locus on ComA activity and biofilm formation. Taken together, our results suggest that RapP strongly represses biofilm formation through multiple targets and that PhrP does not counteract RapP due to a rare mutation in rapP.

FIG 1

The proposed RapP interaction network. The scheme illustrates the RapP network in B. subtilis 3610 based on previous results (black arrows), including the additional interactions (gray arrows) which are proposed on the basis of results reported in this work. RapP (red) is a repressor of Spo0F activity and, directly or indirectly, represses ComA in an Spo0F-independent manner. RapP is repressed by the mature PhrP peptide (A)DRAAT, produced by secretion and subsequent cleavage of PhrP (the three-colored rectangle represents a signal peptide, an extracellularly cleaved prepeptide, and a mature peptide). ComA is activated by the ComX~P QS system (blue) and directly controls the transcription of several operons, including the srfA operon. The exopolysaccharide (eps) operon is indirectly controlled by ComA. Spo0A, the master regulator of sporulation and biofilm formation, indirectly regulates the expression of the eps operon and of the phrP gene. Spo0F is part of the phosphorelay that activates Spo0A and is a known target of many Rap proteins. rapP of strain NCIB3610 encodes an atypical substitution that prevents its repression by the PhrP signaling peptide.

FIG 2

Transcriptional regulation of rapP and phrP promoters. Cells were grown in SMM, and their YFP fluorescence was measured as a function of time as described in Materials and Methods. Time zero is defined as the time when the OD equals 1.5. (A) YFP levels of wild-type 3610 (AES1401, open circle) and the ΔcomA 3610 (AES1452, filled circle) strains carrying a P_rapP-3×YFP reporter. (B) YFP levels in the wild-type strain 3610 background carrying a P_phrP-3×YFP reporter with the following genetic modifications: wild type (AES1411, open circles), Δspo0A (AES1714, filled circles), ΔabrB (AES1668, filled squares), and ΔsigH (AES1413, open squares). Expression levels in this and the following figures are given as the ratio between the mean expression levels of the designated genotype to the autofluorescence of the wild-type PY79. We find autofluorescence to be very stable when cells are grown in minimal medium. Error bars represent the standard error of the mean for at least three independent experiments.

FIG 3

RapP effect on ComA reporters in various genetic backgrounds. Promoter activity was measured in each genotype with (+) or without (−) a P_hs-rapP construct. (A) Shown are the YFP expression levels of four strains carrying the P_rapA-3×YFP reporter in the following backgrounds: wild-type PY79 (AES1419), P_hs-rapP (AES1444), ΔcomA (AES1502), and P_hs-rapP ΔcomA (AES1503). (B) YFP levels were measured in a PY79 P_srfA-3×YFP background without (−) and with (+) an IPTG-inducible rapP construct. The following genetic backgrounds are used: wild-type PY79 (AES1334), P_hs-rapP (AES1379), Δspo0A (AES1472), Δspo0A P_hs-rapP (AES1874), ΔdegU (AES1875), ΔdegU P_hs-rapP (AES1606), ΔdegU Δspo0A (AES1876), and ΔdegU Δspo0A P_hs-rapP (AES1877). (C) The experiment is the same as that described for panel B but in a background of a plasmid-free derivative of strain 3610. The following genotypes were used: the wild-type plasmid-free strain (strain AES1605), P_hs-rapP strain (AES2472), Δspo0A strain (AES2436), and Δspo0A P_hs-rapP strain (AES2437). All measurements were performed at an OD₆₀₀ of 1.5. All genetic backgrounds with the P_hs-rapP construct were induced with 10 μM IPTG. In this and the following figures, the difference between all pairs marked with an asterisk are statistically significant (t test, P < 0.05).

FIG 4

Dependence of eps expression on comA and rapP mutations. (A and B) The distribution of expression levels of YFP driven by the eps promoter as a function of time in strain AES1819 (PY79 P_eps-3×YFP) (A) and strain AES1820 (P_rapP-rapP-phrP P_eps-3×YFP) (B). Note the difference between strains in the fraction of high-YFP-expressing cells (ON population) and how expression changes over time. (C) Shown are the fractions of the ON cells for strains that express the P_eps-3×YFP reporter in the following genetic backgrounds: AES1819 (PY79 wild-type), AES1820 (P_rapP-rapP-phrP), AES1827 (ΔcomA), and AES1828 (ΔcomA P_hs-rapP). Inducible constructs were induced with 10 μM IPTG.

FIG 5

rapP³⁶¹⁰ codes for an N236T substitution mutation compared to the consensus, which renders it insensitive to the PhrP peptide. (A) Sequence alignment of amino acids 226 to 250 from RapP³⁶¹⁰ with the sequences of other RapP homologues and with other Rap proteins found in strain 3610. Asparagine residue 236 is conserved in all Rap proteins except for the product of rapP³⁶¹⁰, where it is replaced with threonine. Residue 236 of RapP is shown in boldface, and the aligned column is boxed for emphasis. (B) The addition of an IPTG-inducible phrP gene suppresses the effect of rapP^T236N but not of rapP³⁶¹⁰ on P_srfA-3×YFP expression. Shown are YFP expression levels of a P_srfA-3×YFP reporter in strain PY79 with different backgrounds, where each genotype was measured with (+) or without (−) a P_hs-phrP construct. Genotypes are as follows: wild-type PY79 (AES1334), P_hs-phrP (AES1477), P_rapP-rapP³⁶¹⁰ (AES1380), P_rapP-rapP³⁶¹⁰ P_hs-phrP (AES1478), P_rapP-rapP^T236N (AES1678), and P_rapP-rapP^T236N P_hs-phrP (AES1709). Expression was measured at an OD₆₀₀ of 1.5, and IPTG was added at 10 μM when needed. (C) P_srfA-3×YFP expression was monitored in a P_rapP-rapP^T236N (AES1678) background as a function of the concentration of the externally added peptides. Three peptides were tested for their ability to repress rapP^T236N activity: the putative signaling peptides of PhrP (DRAAT or ADRAAT) and the hexapeptide signal of PhrH (TDRNTT).

FIG 6

Replacement of rapP³⁶¹⁰ by rapP^T236N eliminates the phenotypic effects of the pBS32 plasmid on ComA activity and biofilm formation. (A) Biofilm formation by the 3610 plasmid-free strain (DS2569), 3610 rapP^T236N-phrP strain (AES1707), and strain 3610 (rapP³⁶¹⁰-phrP, AES1109). (B) YFP expression levels of P_srfA-3×YFP as a function of optical density during growth in SMM for the following backgrounds: 3610 plasmid-free strain (AES1605), 3610 rapP^T236N-phrP strain (AES1873), and strain 3610 (rapP³⁶¹⁰-phrP, AES1336).