Structural Basis of the Arbitrium Peptide–Aimr Communication System in the Phage Lysis–Lysogeny Decision

Articles https://doi.org/10.1038/s41564-018-0239-y

Structural basis of the arbitrium peptide–AimR communication system in the phage lysis–lysogeny decision

Qiang Wang1,4, Zeyuan Guan1,4, Kai Pei1, Jing Wang1, Zhu Liu1, Ping Yin1, Donghai Peng2 and Tingting Zou 3*

A bacteriophage can replicate and release virions from a host cell in the lytic cycle or switch to a lysogenic process in which the phage integrates itself into the host genome as a prophage. In Bacillus cells, some types of phages employ the arbitrium communication system, which contains an arbitrium hexapeptide, the cellular receptor AimR and the lysogenic negative regulator AimX. This system controls the decision between the lytic and lysogenic cycles. However, both the mechanism of molecular recognition between the arbitrium peptide and AimR and how downstream gene expression is regulated remain unknown. Here, we report crystal structures for AimR from the SPbeta phage in the apo form and the arbitrium peptide-bound form at 2.20 Å and 1.92 Å, respectively. With or without the peptide, AimR dimerizes through the C-terminal capping helix. AimR assembles a superhelical fold and accommodates the peptide encircled by its tetratricopeptide repeats, which is reminiscent of RRNPP family members from the quorum-sensing system. In the absence of the arbitrium peptide, AimR targets the upstream sequence of the aimX gene; its DNA binding activity is prevented following peptide binding. In summary, our findings provide a structural basis for peptide recognition in the phage lysis–lysogeny decision communication system.

uring the infection of a bacterial host, a temperate phage can The AimP protein consists of 43 amino acids that encode an either enter the lytic cycle or the lysogenic cycle. In the lytic N-terminal signal sequence and a C-terminal protease-cleavage Dcycle, multiple phage particles are produced and released into site. AimP is secreted by bacteria into the extracellular environment the environment, whereas in the lysogenic cycle, a phage incorpo- and is processed into a mature six-residue peptide called the arbi- rates its DNA into the host genome to establish a potentially ben- trium peptide. The arbitrium peptide concentration increases in the eficial relationship with the host1. The phage harnesses two types late infection stages as the number of phage particles increases. The of life-cycle transitions to survive under different conditions. arbitrium peptide can be taken up by other B. subtilis cells through Numerous studies have focused on the lysis–lysogeny decision of the bacterial oligopeptide permease transporter and binds to the bacteriophage lambda in Escherichia coli2–4. Some gene regulatory AimR receptor, inhibiting its DNA binding activity. As a result, the circuits in the phage genome, including the pL and pR lytic pro- lytic cycle is strongly suppressed and the phage switches to the lyso- moters, initiate early transcription and induce expression of the genic life cycle. This type of arbitrium communication system has lysogenic regulators CII and CIII5,6. In addition, intracellular signal- been identified in more than 100 Bacillus phages or prophages; arbi- ling factors play important roles in the lysis–lysogeny decision. For trium hexapeptide sequences are variable in the first three positions example, the accumulation of the lysogenic regulator CII protein but are more conserved in the last three positions (Supplementary promotes phage genome integration into the bacterial chromo- Fig. 1). Each hexapeptide binds specifically to the corresponding some7. However, few studies have focused on signal transduction in AimR receptor from the same phage to trigger the lysis-to-lysogeny other phage infection systems. transition. However, the molecular basis of the specific recogni- Recently, a communication system that contributes to the deci- tion between the arbitrium peptide and AimR remains unknown. sion process between lysis and lysogeny was characterized for SPbeta Moreover, the mechanism by which the arbitrium peptide regulates group phages during infection of Bacillus subtilis cells. This system, AimR DNA binding activity needs to be investigated. demonstrated using a phi3T phage, contains three phage genes: Here, we report the crystal structures of the SPbeta phage AimR aimP, encoding a small peptide as a signal molecule; aimR, encod- alone and in complex with the arbitrium peptide (GMPRGA) at ing a receptor for perceiving the signal peptide; and aimX, encod- resolutions of 2.20 Å and 1.92 Å, respectively. Both the apo and the ing a regulatory protein that inhibits the lysogenic process8–10. AimR peptide-bound forms dimerized via the C-terminal capping helix. includes a predicted N-terminal helix–turn–helix DNA binding AimR comprises four atypical TPR repeats that form a peptide motif followed by a tetratricopeptide repeat (TPR) domain, which accommodation pocket with other helices, which is reminiscent of is a typical motif found in intracellular peptide receptors of the other RRNPP family members. Together with biochemical analyses, RRNPP (named for the different sensors: Rgg, Rap, NprR, PlcR and we show that AimR directly recognizes the upstream sequence of PrgX) family in quorum-sensing systems11–14. During the lytic cycle, the aimX gene in vitro but abrogates its DNA binding activity fol- AimR binds upstream of the aimX gene, activating its transcription. lowing arbitrium peptide binding. This study provides a structural

1National Key Laboratory of Crop Genetic Improvement and National Centre of Plant Gene Research, Huazhong Agricultural University, Wuhan, China. 2State Key Laboratory of Agricultural Microbiology, Huazhong Agricultural University, Wuhan, China. 3College of Life Science and Technology, Huazhong Agricultural University, Wuhan, China. 4These authors contributed equally: Qiang Wang, Zeyuan Guan. *e-mail: [email protected]

Nature Microbiology | www.nature.com/naturemicrobiology Articles Nature Microbiology

a a 100 Å α4 N α5 N α2 9 37 Å L386 L386 α α6 α10 α3 α7 α1 L383 L384 L383 α8 α11 60 Å L384 α14 α12 CC L380 L380 18 13 α AimR α AimR A α16 B Helix 20 Helix 20 α20 AimRA AimRB 17 α15 α α19 b 44.1 kDa 88.0 kDa 50 Å 1.5 125 b 93.3 ± 1.7 kDa 100 1.0 75 47.6 ± 1.5 kDa

c(s) 50 0.5 25

0.0 Molecular mass (kDa ) 0 2 468 13 14 15 16 17 18 Sedimentation coefficient (S) Volume (ml)

Fig. 2 | Effect of the AimR α20 on dimerization. a, Close-up view of the AimR dimerization interface. The residues L380, L383, L384 and L386 Fig. 1 | Crystal structure of the apo form of AimR. a, The overall structure from protomer A (α20), as well as their counterparts from protomer B, of apo AimR. Dimeric AimR has the shape of an inverted trapezoid that is that are involved in hydrophobic interactions, are highlighted with sticks. 60 Å deep, with sides measuring 100 Å and 50 Å. The width of the inner b, Molecular mass of full-length AimR and the AimR ΔC mutant (residues cavity is approximately 37 Å. The two protomers are coloured from blue to 1–373), as measured by SV-AUC (left) and SLS (right). Full-length AimR is red along their length. The N terminus and C terminus of each protomer shown in black, and the AimR ΔC mutant in blue. The calculated molecular are indicated. The number of α-helices are sequentially named from the N masses are indicated above each panel. c(s) represents continuous terminus and are labelled on the cylinders. b, Surface features of the AimR (sedimentation coefficient distribution) analysis model. These experiments homodimer by electrostatic potential (blue, basic; red, acidic). Two regions were performed twice with equivalent results. Values in the right panel on the top side (indicated by the pink dashed outlines) are surrounded by represent mean ± s.d. positively charged amino acids. basis for the arbitrium communication system in phages during the of 20 α-helices, eight of which assemble into four atypical TPR lysis–lysogeny decision process. motifs: TPR1 (residues 186–226), TPR2 (residues 227–263), TPR3 (residues 295–327) and TPR4 (residues 328–361) (Supplementary Results Fig. 3). A Dali search using AimR as a query showed that 6 of 11 Crystal structure of apo AimR. To elucidate the mechanism of retrieved structures, comprising RapF, RapJ, NprR, PlcR, RapI and peptide recognition by the AimR protein, we determined the crystal RapH, with Z-scores higher than 13, were RRNPP family mem- structures of AimR with and without the AimP peptide. We sys- bers (Supplementary Table 2)15–20. These structural features are tematically screened the expression and peptide-binding activities consistent with previous analyses reporting that AimR belongs to of AimR homologues from Bacillus aerophilus, Bacillus sonorensis the RRNPP family, which is found in quorum-sensing systems of L12 and Bacillus phages phi3T and SPbeta. After numerous crystal- Gram-positive bacteria8. lization attempts, we crystallized the full-length apo and peptide- The C-terminal helix (α20), named the capping helix, mediates bound forms of AimR from the SPbeta group phage. The complex AimR dimerization (Fig. 1a and Fig. 2a) and has a solvent-accessible structure was determined using the single-wavelength anomalous area of approximately 810 Å2. Several residues from each capping diffraction method and was refined to a resolution of 1.92 Å. The helix, including L380, L383, L384 and L386, form a network of structure of the apo form was determined at a 2.20 Å resolution and van der Waals interactions (Fig. 2a). To investigate the oligomer- solved by molecular replacement using the coordinates of the AimP ization state of AimR in solution, we performed sedimentation peptide–AimR complex structure. The electron density of the N ter- velocity analytical ultracentrifugation (SV-AUC) and static light minus of AimR (amino acid residues 1–43) does not result in a well- scattering (SLS) analyses. AimR exhibited a molecular weight of built and refined model, suggesting that it is flexible. The diffraction approximately 88.0 kDa as measured by SV-AUC and 93.3 ± 1.7 kDa and refinement statistics are listed in Supplementary Table 1. by SLS; values that are close to the expected molecular mass of the The crystallographic asymmetric unit of the apo form of AimR dimer (90.6 kDa). In contrast, the molecular weight calculated for contains two molecules as a homodimer. The overall structure of the AimR ΔC mutant (residues 1–373; truncation of helix 20) was dimeric AimR appears as an inverted trapezoid that is 60 Å deep, approximately 44.1 kDa and 47.6 ± 1.5 kDa by SV-AUC and SLS, with sides measuring 100 Å and 50 Å. The distance between the two respectively (Fig. 2b). These results corroborated the dimeric struc- lysine residues, K145 of AimRA and K145 of AimRB, (in the inner ture observed and revealed the essential role of the capping helix. cavity) is approximately 37 Å (Fig. 1a). Analysis of the surface electrostatic potential of AimR revealed two positively charged patches Crystal structure of the AimP peptide-bound AimR. In the at the longer side that appear to be related to its DNA binding activ- complex crystals, two AimR molecules formed a homodimer in ity (Fig. 1b). The two protomers (AimRA and AimRB) are nearly each asymmetric unit; dimerization of these two molecules is also identical, with an average root mean square deviation of 1.469 Å mediated by the C-terminal capping helix (Fig. 3a). Although the over an average of 321 aligned Cα atoms (Supplementary Fig. 2). electron density of the AimR N-terminal residues 1–40 does not Each AimR forms a right-handed superhelical assembly comprised generate a well-built model, the overall structure of the arbitrium

Nature Microbiology | www.nature.com/naturemicrobiology Nature Microbiology Articles

a N N

α10

α12 α13 CC

Ala6th α18 Gly5th 14 Arg4th α Pro 17 3rd α Met2nd Gly AimR AimR 1st A B α15 α16 b c 88.2 kDa 92.3 ± 2.7 kDa 1.5 125 100 1.0 75

c(s) 50 0.5 25

0.0 Molecular mass (kDa) 0 2 4 68 13 14 15 16 17 18 Sedimentation coefficient (S) Volume (ml)

Fig. 3 | Crystal structure of the arbitrium peptide-bound AimR. a, Structure of dimeric arbitrium peptide GMPRGA-bound AimR. The GMPRGA peptides are shown as ball-and-stick models. b, The peptide-binding pocket. AimR is shown as an electrostatic surface, coloured by potential from blue (positive) to red (negative). The bound GMPRGA peptide is shown as a ball-and-stick model. c, The molecular mass of the GMPRGA peptide-bound, as measured by SV-AUC (left) and SLS (right). The calculated molecular masses are indicated above. The retention volume of 14.35 ml and the log[Mw] under the peak correspond to a homogenous sample of 92.3 ± 2.7 kDa (right), consistent with the data for dimeric AimR (91.8 kDa). c(s) represents continuous (sedimentation coefficient distribution) analysis model. These experiments were performed twice with equivalent results. Values in the right panel represent mean ± s.d. peptide-bound AimR has an inverted-trapezoid-like fold, similar the arbitrium peptide include a network of extensive hydrophobic to that of the apo form. The N-terminal portions of AimR open contacts and many hydrogen bonds (Supplementary Fig. 7)21. Close slightly, with a small increase in the width of the inner cavity inspection of the structures revealed that the peptide is primarily (~39 Å compared to ~37 Å in the apo AimR dimer) (Supplementary coordinated by nine residues through extensive hydrogen bond- Fig. 4). In each AimR protomer, four TPR repeats formed a pocket ing. Four residues are hydrogen bonded with oxygen atoms from with other helices (α7–9, α14 and α19) (Supplementary Fig. 3b) the peptide backbone: the residues Y159 of helix 7 and R228 of surrounding the arbitrium peptide GMPRGA, where the C termi- TPR1 in AimR contact the oxygen moieties of the sixth alanine; nus of the peptide is anchored inside the pocket, pointing to the N residue N202 of TPR1 interacts with the fifth glycine and N239 of terminus of AimR (Fig. 3a,b). TPR2 hydrogen bonds to the third proline. Two residues in TPR3, To further investigate the AimR dimer in the presence of the Q299 and E300, and the carbonyl oxygen of M296 interact with arbitrium peptide, we analysed the protein using both AUC and the nitrogen moiety of the first glycine. Moreover, three residues SLS analyses. As expected, the molecular mass of the arbitrium (N206, N329 and D360) specifically recognize the side chain of peptide–AimR complex (theoretical molecular mass of 91.8 kDa), the fourth arginine (Fig. 4a and Supplementary Fig. 7). Buttressing as derived from AUC, was approximately 88.2 kDa, indicative of a the hydrogen bonds, five hydrophobic residues from AimR (L205, homodimer (Fig. 3c, left). In the SLS analysis, the retention volume L242, F276, F362 and L363) are involved in van der Waals interac- of 14.35 ml and the log[Mw] under the peak correspond to a mass tions with the side chain of the second methionine (Fig. 4b,c and of 92.3 ± 2.7 kDa, confirming the dimeric state of the arbitrium Supplementary Fig. 7b). peptide–AimR complex in solution (Fig. 3c, right). Together with The importance of these residues in peptide coordination was the structural observation, these analyses strongly indicate that reinforced by mutational analysis. Some single point mutations, arbitrium peptide binding does not dissociate the AimR dimer, but including N206A, F276N, Q299A, E300A and N329A, exhib- might cause a subtle conformational change. Interestingly, AimR ited minor effects on peptide-binding activity. In contrast, other from the phi3T phage, a homologue that only shares 39% sequence single point mutations, including Y159A, L205N, R228A, L242N similarity to SPbeta, exists as a dimer in the absence of the arbitrium and D360A, showed reduced binding activity (Supplementary peptide but dissociates to a monomer following peptide binding8. Table 3). Importantly, a sequence alignment of eight AimRs from Both the size-exclusion chromatography and the crosslinking assays Bacillus phages revealed that three residues (Y159, R228, and suggest that there are at least two distinct oligomerization regula- D360) are highly conserved (Supplementary Fig. 1). Consistent tion mechanisms among these phages (Supplementary Fig. 5). with the conservation analysis, neither double mutant Y159A/ D360A nor R228A/D360A retained obvious peptide-binding activ- Peptide-binding mode. Following the assignment of most amino ity (Fig. 4c and Supplementary Table 3). The experimental appar- acids (residues 41–386) in the electron density map, strong electron ent dissociation constant (Kd) values of these mutants are listed in densities that are indicative of the arbitrium peptide (GMPRGA), Supplementary Table 3. In summary, these results demonstrated became clearly visible in the superhelical cavity of the AimR pro- that the residues Y159, R228 and D360 of AimR play an essential tein (Supplementary Fig. 6). The interactions between AimR and role in peptide binding.

Nature Microbiology | www.nature.com/naturemicrobiology Articles Nature Microbiology

a b

R228 Ala Y159

N202 Gly

N206 L205 Arg

N239

F362 D360 Pro N329 Met L242 L363 Gly

E300 Q299 F276

c Time (min) Time (min) Time (min) Time (min) 01020304050 01020304050 01020304050 01020304050 0.00 0.00 0.00 ) ) ) ) 0.00 –1 –1 –1 –1 –0.50 –0.20 –0.50 ( µ cal s ( µ cal s ( µ cal s ( µ cal s –1.00 –0.02 –0.40 –1.00 –1.50 ) ) ) ) 0.4 0.0 0.0 –1 –1 –1 –1 0.0 0.2 –5.0 –5.0 –5.0 0.0 –10.0 –10.0 –10.0 –0.2 –15.0 –0.4 –15.0 –15.0 –20.0 –0.6 –20.0 –20.0 –25.0 –0.8 Injectant (kcal mol Injectant (kcal mol Injectant (kcal mol –25.0 Injectant (kcal mol 0.0 0.5 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 Molar ratio Molar ratio Molar ratio Molar ratio K = 4.6 1.4 nM K = 16.4 10.4 nM K = 37.5 5.3 nM d ± d ± d ± R228A/D360A Wild type R228A D360A

Fig. 4 | Recognition of GMPRGA by AimR. a, Close-up view of the hydrogen bond interactions between the arbitrium peptide and the side chains of AimR residues. The hexapeptide is shown as a yellow ball-and-stick model, and the residues of AimR participating in hydrogen bond formation with the arbitrium peptide are shown as sticks, with nitrogen in blue and oxygen in red. Red dashed lines represent hydrogen bonds. b, Close-up view of hydrophobic interactions between the arbitrium peptide and AimR. The arbitrium peptide is shown as in a and the category of the amino acids is indicated. The residues on AimR participating in hydrophobic interactions are shown as grey sticks. c, Interaction of AimR mutants and GMPRGA peptide as detected by ITC. The ITC curves of the interaction of the AimR proteins (wild type, R228A, D360A, R228A/D360A) with the GMPRGA peptide are shown. The calculated Kd mean ± s.d. values are indicated. The top plots represent time, and the bottom plots represent molar ratio. These experiments were performed twice with equivalent results.

DNA binding activity of AimR inhibited by arbitrium peptide. we identified the 51 bp DNA segment responsible for AimR bind- Using a ChIP-seq assay, a previous study showed that AimR from ing: 5′ GGCCATCACTTAAATATTAGGTTTTAATAACATCTAG phage phi3T only coprecipitates with an intergenic DNA fragment TGATCAACTTCAAA 3′ (Supplementary Fig. 8b,c). AimR exhibits (from between the aimP and the aimX genes) when the medium notable DNA binding activity to this probe, with Kd = 101.5 ± 5.9 nM lacks the arbitrium peptide8. In the presence of the arbitrium pep- (Supplementary Fig. 9a). tide, the expression of aimX was reduced more than 20-fold. These We also examined whether the N-terminal motif or C-terminal observations suggest that AimR homologues are DNA binding capping helix of AimR are involved in DNA binding. N-terminal proteins that regulate transcriptional activation and that they lose truncation (residues 1–43) completely abolished the ability to target activity following peptide binding8. To test this model, we investi- DNA, whereas C-terminal truncation (residues 374–386) showed gated AimR DNA binding activity, with or without a bound peptide, a large decrease in DNA binding activity and the Kd could not be using electrophoretic mobility shift assays (EMSA). First, we incu- well fitted (Fig. 5a–c and Supplementary Fig. 9b). Based on our bated the AimR protein with a 653 bp DNA fragment (an upstream structure, two positively charged patches, constituting 14 lysine or sequence of aimX in the SPbeta phage genome 77469–78121) and arginine residues near the N terminus, were identified (Fig. 1b). characterized a 110 bp sequence that was strongly protected in The mutation of ten of these residues (K43D, K77D, K79D, R82D, DNase I footprinting assays (Supplementary Fig. 8a). Using EMSA, K111D, K113D, R117D, K137D, K141D and K145D) resulted in the

Nature Microbiology | www.nature.com/naturemicrobiology Nature Microbiology Articles acb Supplementary Fig. 11b). Furthermore, the AimR C mutation AimR AimR N (44–386) AimR C (1–373) Δ Δ Δ retains peptide binding activity (Supplementary Fig. 12b) and also responds to arbitrium peptide-triggered DNA dissociation (Fig. 5g). In contrast, the double mutant (R228A/D360A), which is unable to interact with the arbitrium peptide, retained similar DNA binding activity, even in the presence of the arbitrium peptide (Fig. 5h,i). Thus, these data confirmed that the arbitrium peptide disrupts the AimR DNA binding activity in vitro. * To investigate whether AimR and arbitrium peptide binding is essential for the phage lysis–lysogeny decision in vivo, we examined Lane 1234512 345 12345 the infection dynamics of the phages SPbeta and SPbeta ΔaimR . d GMPRGA AimR e GFPRGA AimRGf FGHGA AimR The wild type aimR gene and its different mutants were integrated into the amyE locus, using the Bacillus subtilis CU1065 genome as host strains. For wild-type phage SPbeta, the infected bacteria were lysed three hours after infection. When peptides were added into the culture, the arbitrium peptide (GMPRGA) protected bacteria from lysis by promoting phage lysogeny; however, the control peptide (GFGHGA) failed to prevent lysis (Fig. 6a), which is in agreement with previously reported observations8. B. subtilis CU1065 infected * with SPbeta ΔaimR exhibited lysogeny as anticipated (Fig. 6b). The lytic phenomenon was recovered by expressing AimR in B. subtilis Lane 1234512 345 12345 CU1065 (amyE::aimR) (Fig. 6c). However, in B. subtilis CU1065 g h i GMPRGA AimR GMPRGA (amyE::aimR R228A/D360A), the arbitrium peptide GMPRGA no AimR ΔC (1–373) (R228A/D360A) AimR (R228A/D360A) longer promoted lysogeny, confirming the essential role of these residues for peptide binding in vivo (Fig. 6d). Moreover, neither B. subtilis CU1065 (amyE::aimR ΔN 44–386) nor B. subtilis CU1065 (amyE::aimR ΔC 1–373), induced lysis after infection with SPbeta ΔaimR (Fig. 6e,f). Collectively, our in vivo and in vitro results indicate that AimR DNA binding activity, which is regulated by the arbitrium peptide, is essential for the lysis–lysogeny decision during SPbeta infection. * Discussion Lane 1234512345 12345 A previous study identified a phage communication system that regulates the lysis–lysogeny life cycle. Accumulation of the arbi- Fig. 5 | Effect of the arbitrium peptide on the DNA binding activity of trium peptide in a bacterial cell is recognized by the intracellular AimR. a–c, DNA binding activities of AimR (a), N-terminal-truncated AimR receptor AimR, which activates the phage lysogenic developmental (b) and C-terminal-truncated AimR (c) were examined. d–f, Effect of the pathway8. In this study, the high-resolution crystal structure of the arbitrium peptides GMPRGA (d), GFPRGA (e) and GFGHGA (f) on the arbitrium-peptide–AimR complex revealed the molecular basis of DNA binding activity of AimR. Residues in bold font indicate the sequences the specific peptide recognition by the AimR protein. The four TPR differences. g, Effect of the arbitrium peptide GMPRGA on the DNA repeats in AimR are atypical TPR folds, they form a concave pocket binding activity of the C-terminal-truncated AimR. h, DNA binding activity with other helices to accommodate the peptide GMPRGA. In con- of the AimR double mutant (R228A/D360A). i, Effect of the arbitrium trast to most hydrogen bonds (which arise from the backbone of the peptide GMPRGA on the DNA binding activity of the AimR double mutant arbitrium peptide), one amino acid, the arginine in position 4, uses (R228A/D360A). In a-i, EMSA experiments were performed twice and a its guanidinium group to form a hydrogen bond with three residues representative result is shown. The AimR protein concentrations in lanes (N206, N329 and D360). Thus, we speculated that the substitution 1–5 are 0, 0.008, 0.04, 0.2 and 1 μM, respectively. The concentration of the of similar amino acids in the peptide would make it difficult to dis- 51 bp DNA fragment was 20 nM. The black triangles represent increasing rupt the interactions. For example, the peptide GFPRGA from the amounts of AimR. AimR–DNA complex formation is indicated by a black prophage in B. sonorensis L12 showed comparable binding affinity arrowhead and free DNA by a black asterisk. to AimR in vitro and consequently terminated the DNA binding activity (Fig. 5e and Supplementary Fig. 11a), raising the possibility of potential cross-talk between different phages. complete abolishment of DNA binding (Supplementary Fig. 10). In this study, we determined that AimR targets a 51 bp DNA These results suggest that the N-terminal portion of AimR is crucial sequence located downstream of the aimP gene and that its DNA for DNA targeting and that dimer formation is essential for its DNA binding activity is lost in the presence of the arbitrium peptide. binding properties. Interestingly, the N-terminal motif of AimR (residues 1–43), which As anticipated, the addition of the arbitrium peptide (GMPRGA) is invisible in the crystal structure, is crucial for DNA binding leads to complete dissociation of AimR from DNA (Fig. 5d). As a (Fig. 5b). A previous report predicted some AimR homologues, for control, AimR was incubated with two peptides (GFPRGA and example, AimR from phi3T encodes an N-terminal helix–turn– GFGHGA, the residues in bold font indicate sequence differences) helix motif, although AimR from SPbeta does not appear to pos- from other species, to assess DNA binding activity. The first hexa- sess a typical DNA binding motif8. Our results suggest that both peptide, with only one alteration, has a binding affinity similar the N-terminal motif and the nearby positively charged residues are to that of the SPbeta cognate GMPRGA peptide (Supplementary responsible for DNA binding (Supplementary Fig. 10). To identify Fig. 11a) and disrupts the interaction of AimR with DNA (Fig. 5e). the DNA bases that interact with AimR, we simultaneously mutated However, the latter hexapeptide showed no detectable binding to five nucleotides and screened ten multi-site mutant DNA fragments AimR and did not disrupt the DNA–AimR complex (Fig. 5f and by EMSA. Unfortunately, no significant decrease in binding affinity

Nature Microbiology | www.nature.com/naturemicrobiology Articles Nature Microbiology

a b c SPbeta SPbeta (ΔaimR ) SPbeta (ΔaimR ) B. subtilis CU1065 B. subtilis CU1065 amyE::aimR 1.2 ck 1.2 ck 1.2 ck No peptide No peptide No peptide 1.0 GMPRGA 1.0 GMPRGA 1.0 GMPRGA GFGHGA GFGHGA GFGHGA 0.8 0.8 0.8

0.6 0.6 0.6

0.4 0.4 0.4

0.2 0.2 0.2 Absorbance at 600 nm Absorbance at 600 nm Absorbance at 600 nm

0.0 0.0 0.0 0 100 200 300 400 500 0 100 200 300400 500 0100 200300 400 500 Time (min) Time (min) Time (min) d e f SPbeta (ΔaimR ) SPbeta (ΔaimR ) SPbeta (ΔaimR ) amyE::aimR R228A/D360A amyE::aimR ΔN 44–386 amyE::aimR ΔC 1–373 1.2 ck 1.2 ck 1.2 ck No peptide No peptide No peptide 1.0 GMPRGA 1.0 GMPRGA 1.0 GMPRGA GFGHGA GFGHGA GFGHGA 0.8 0.8 0.8

0.6 0.6 0.6

0.4 0.4 0.4

0.2 0.2 0.2 Absorbance at 600 nm Absorbance at 600 nm Absorbance at 600 nm

0.0 0.0 0.0 0 100 200 300 400 500 0 100 200 300400 500 0100 200300 400500 Time (min) Time (min) Time (min)

Fig. 6 | Effect of the arbitrium peptides on the infection dynamics of SPbeta and SPbeta ΔaimR. a,b, Growth curves of the B. subtilis strain CU1065 infected with SPbeta (a) and SPbeta Δaim (b). c–f, Growth curves of SPbeta ΔaimR infection of B. subtilis strain CU1065 amyE::aimR (c), amyE::aimR R228A/D360A (d), amyE::aimR ΔN (e) and amyE::aimR ΔC (f). Infections were performed at a multiplicity of infection of 0.001 in LB media supplemented with no peptide, 1 μM synthesized GMPRGA peptide, or GFGHGA peptide. ck, no phage infection. Data represent an average of three biological replicates, each with three technical replicates. Error bars indicate the s.d. of at least three independent measurements (the points are the mean).

was observed (Supplementary Fig. 13a,b). Additionally, the mono- N-terminal domain to block the RapF-ComA interaction15. PapR meric mutant AimR (residues 1–376) showed markedly reduced peptide binding to PlcR leads to a conformational change within DNA binding activity, suggesting that the dimer conformation also the linker helix and moves the helix–turn–helix domain to acti- favours DNA binding. These results are limited but still enlighten- vate its DNA binding activity18,30. In this study, we did not observe ing. Studies of the structure of the DNA-bound AimR complex will the oligomerization state transition for AimR from SPbeta bound to shed further insights on the DNA recognition by AimR. the arbitrium peptide and slight conformational changes between The signal pathway for the phage lysis–lysogeny decision is rem- the apo and peptide-bound structures (the N-terminal motif iniscent of the quorum-sensing cell–cell communication system, missed) were observed. However, the overall structure of full-length which is known in Gram-positive bacteria to coordinate biofilm AimR became extended, as revealed by small-angle X-ray scattering formation, virulence and antibiotic resistance22,23. In the quorum- (SAXS) analysis (Supplementary Fig. 14a,b). Thus, we speculated sensing system, peptide pheromones (generally 5–10 amino acids) that peptide binding might allosterically lead to a conformational secreted by cells are perceived by RRNPP family receptors24–26. All change in the N-terminal motif (Supplementary Fig. 14c). Another RRNPP members are characterized by an N-terminal regulatory possibility is that the complex structure of DNA-bound AimR is target-binding domain followed by a TPR domain that is responsi- somewhat different from the DNA-free AimR structure and that ble for self-oligomerization and cognate signal peptide interaction. the peptide is likely to inhibit the conformational changes. Overall, Structural and biochemical analyses revealed that the TPR domain we added a new component to the working model of the RRNPP mediates arbitrium peptide recognition, homodimerization and family and provided insight into peptide-signal regulation in Gram- DNA association15,17,18,27–35. Despite the structural element similari- positive bacteria. ties among these RRNPP proteins, the molecular basis for downstream gene regulation by the receptors is distinct. In some cases, Methods the peptides induce oligomerization of the receptors. Apo NprR Bacterial strains, plasmids and phage. Te B. subtilis strains 168 (BGSCID: 1A1) is a dimer with little DNA binding activity, whereas NprX peptide and CU1065 (BGSCID: 1A100) were obtained from the Bacillus Genetic Stock binding leads to the formation of an NprR tetramer and activates Center (BGSC). Te pDG1730 and pDG780 plasmids were gifs from M. Sun and downstream gene expression17. Without the peptide, PrgX exists D. Peng (Huazhong Agricultural University). Wild-type or mutant aimR genes with their own promoter were cloned into the pDG1730 vector. Tese plasmids as a tetramer. The activating peptide cCF10, but not the inhibi- were transformed into B. subtilis CU1065 and the modifed genes were integrated 28,29 tory peptide iCF10, destabilizes the PrgX tetramer . The AimR into the amyE locus of the B. subtilis genome. Competent cells were prepared receptor from the phage phi3T changes its oligomeric state from following the classical nutritional downshif method, as previously described36. dimer to inactive monomer after binding the arbitrium peptide Phage SPbeta was induced from B. subtilis strain 168 using 0.5 μg ml−1 Mitomycin (Supplementary Fig. 5), which inhibits aimX transcription and C (MedChemExpress) for 3 h. Te lysate was fltered through a 0.45 μm membrane flter (Millipore), purifed and amplifed in B. subtilis strain CU106537. results in lysogeny8. On the other hand, some receptors undergo conformational changes to expose, or bury, the regulatory target Protein expression and purification. The codon-optimized complementary binding domain. After binding the PhrF peptide, RapF relocates its DNA of full-length aimR from SPbeta (GenBank: NC_001884) was subcloned

Nature Microbiology | www.nature.com/naturemicrobiology Nature Microbiology Articles into the pET21b vector (Invitrogen) with a C-terminal 8×His tag. The AimR Size-exclusion chromatography assay. The oligomeric states of AimRs were mutants were generated with a standard two-step PCR-based strategy. All of determined by size-exclusion chromatography. AimR proteins (10 μM) from the proteins were overexpressed in the E. coli strain BL21 (DE3), cultured in SPbeta and phi3T were purified and incubated for 30 min with: GMPRGA peptide Luria-Bertani medium. AimR expression was induced by the addition of 0.2 mM (20 μM), SAIRGA peptide (20 μM), or without peptide. Samples were then injected isopropyl-β-D-thiogalactopyranoside at an absorbancy (600 nm) of 1.1. The cells into a Superdex 200 Increase 10/30 GL column equilibrated with buffer containing were harvested 16 h after induction at 16 °C, resuspended in a buffer containing 150 mM NaCl, 20 mM Tris-HCl (pH 8.0) and 2 mM DTT. Data analysis was carried 25 mM Tris-HCl (pH 8.0) and 150 mM NaCl, and disrupted using a homogenizer out using GraphPad Prism7. (JNBIO). The cell debris was removed via centrifugation at 23,000g for 1 h at 4 °C. The supernatant was collected and loaded onto Ni2+-nitrilotriacetate affinity resin Crosslinking assay. The oligomeric states of AimR, AimR (R228A/D360A) and (Ni-NTA, Qiagen), followed by washing with 25 mM Tris-HCl (pH 8.0), 150 mM peptide-bound AimR from SPbeta and phi3T were characterized by chemical NaCl and 15 mM imidazole. AimR was eluted using 25 mM Tris-HCl (pH 8.0) and crosslinking using an amine-group specific and water-soluble crosslinker 250 mM imidazole. The eluted protein was further purified using anion-exchange bis(sulfosuccinimidyl)suberate (BS3; ThermoFisher Scientific). AimRs or peptide- chromatography (Source 15Q 10/100, GE Healthcare). The elution peak was bound AimR (0.5 mg ml−1) were mixed with five different BS3 concentrations subjected to size-exclusion chromatography (Superdex-200 Increase 10/300, GE (0, 1.25, 2.5, 5 and 10 mM) in buffer containing 150 mM NaCl and 25 mM HEPES Healthcare) in a buffer containing 25 mM Tris-HCl (pH 8.0), 150 mM NaCl and (pH 7.5) and incubated at room temperature for 30 min. Reactions were quenched 5 mM dithiothreitol (DTT). The peak fractions were collected for crystallization by the addition of 50 mM Tris-HCl (pH 7.4) for 15 min and then analysed on 12% and assays. SDS-polyacrylamide gel electrophoresis.

Crystallization. For the apo form of AimR and the peptide-bound AimR complex, DNase I footprinting assay. A 653 bp DNA probe (77469–78121) between the the purified protein was concentrated to ~150 μM (~6.6 mg ml−1), incubated on aimR and aimX open reading frames was amplified from the SPbeta genome using ice for 30 min with or without 1.5 mM peptide GMPRGA (GLS Biochem), and PCR with specific fluorescence-labelled primers (Supplementary Table 4, primer crystallized in hanging drops at 18 °C using a mixture of 1 µl of the sample and 1 and primer 2) and purified with a Nucleic Acid Purification kit (Tiangen). The 1 µl of the crystallization buffer. Crystals of the apo and complex forms appeared DNA probes and AimR protein (0.8 μM ml−1) were mixed and treated with DNase under the same conditions in the presence of 10% polyethylene glycol 8000, 0.1 M I (Fermentas, 1.2 U ml−1) at 25 °C for 5 min. The reaction was stopped by adding sodium cacodylate (pH 6.0) and 0.2 M NaCl. AimR crystals grew better when the 0.25 M EDTA and incubating in a water bath at 75 °C for 15 min. The digested polyethylene glycol 8000 concentration was increased to 12%. The crystals of the DNA fragments were purified with the Nucleic Acid Purification kit (Axygen) and GMPRGA peptide–AimR complex were further optimized in 9% polyethylene eluted in 50 μl distilled water. The purified DNA was sequenced. glycol 8000, 0.1 M sodium cacodylate, pH 6.1, 0.2 M NaBr and 5 mM DTT. The crystals grew to full size in three days. They were flash-frozen in liquid nitrogen EMSA. DNA fragments were amplified from the SPbeta genome using PCR with and cryoprotected by adding glycerol to a final concentration of 30%. The apo form specific fluorescence labelled primers (Supplementary Table 4, primer 1–14) for crystals were diffracted to 2.20 Å and the complex-form crystals were diffracted to screening of the minimal DNA binding fragment size. DNA probes were annealed 1.92 Å. with FAM-labelled primers (Supplementary Table 4, primer 15 and primer 16) in boiling water to generate the 51 bp DNA fragment. The labelled probes were Data collection and structure determination. All of the diffraction data were incubated with 0.008, 0.04, 0.2 and 1 μM AimR proteins or peptide-bound AimR collected at the Shanghai Synchrotron Research Facility (SSRF) on beamline complex (molar ratio = 10:1) in EMSA buffer (25 mM Tris-HCl (pH 8.0); 5 mM −1 −1 BL17U or BL19U and then integrated and processed using the HKL2000 program MgCl2; 5 mM DTT; 0.1 mg ml BSA; 10% glycerol; 150 mM NaCl and 50 μg ml suite38–40. Further data processing was performed using the CCP4 suite41. The salmon sperm DNA) at 4 °C for 15 min. The reactions were resolved on 6% native structure of GMPGRA peptide-bound AimR was determined using the single- acrylamide gels (37.5:1 acrylamide:bis-acrylamide) in 0.5 × Tris-Boric acid buffer at wavelength anomalous diffraction method with selenomethionine-labelled 150 V for 2.5 h. Images of the gels were obtained using FLA5100 (Fujifilm). crystals. The structure of the apo form of AimR was determined through molecular replacement with the peptide-bound structure as the search model Construction of aimR phage deletion strain. Long flanking homology PCR was using the program PHASER42. All of the structures were iteratively built using used to disrupt the aimR gene in B. subtilis strain 16845. Three fragments were COOT and refined using the PHENIX program43,44. Data collection and structure prepared as follows: (1) The upstream region flanking aimR (700 bp upstream to refinement statistics are summarized in Supplementary Table 1. All of the figures aimR start codon) was amplified from B. subtilis 168; (2) The kanamycin resistance were generated using PyMOL (http://www.pymol.org/). cassette was amplified from plasmid pDG780; (3) The downstream region flanking aimR (700 bp downstream to aimR start codon) was amplified from B. subtilis 168. Isothermal titration calorimetry. Isothermal titration calorimetry (ITC) The three fragments were joined using Q5 High-Fidelity DNA Polymerase (NEB) experiments were performed at 25 °C using Auto-iTC200 titration calorimetry and transformed into the B. subtilis 168 competent cells. The aimR deletion mutant (MicroCal). To assess the peptide-binding affinity of proteins, GMPRGA (100 μM) was selected for kanamycin resistance and was confirmed via sequencing. The was dissolved in 40 μl reaction buffer (25 mM Tris-HCl (pH 8.0); 150 mM NaCl) SPbeta ΔaimR phage was induced using Mitomycin C. and titrated against 200 μl 20 μM wild-type AimR protein or the mutants in the same buffer. The first injection (0.5 μl) was followed by 19 injections of 2 μl. The Growth dynamics of SPbeta-infected B. subtilis. Overnight cultures of heat of dilution was measured by injecting GMPRGA into the buffer alone in each B. subtilis CU1065 and aimR-overexpressing strains were diluted 1:100 in Luria experiment. The values were subtracted from the experimental curves before data Bertani media and incubated at 28 °C with shaking until the suspension reached analysis. The stirring rate was 750 r.p.m. The MicroCal Origin software, which was absorbancy at 600 nm of 0.1. Next, 1 μM synthesized peptides were added and supplied with the instrument, was used to determine the site-binding model that incubated for 1 h. The bacteria were infected with the SPbeta phage or the SPbeta produced a good fit (low × 2 value) for the resulting data. ΔaimR phage at a multiplicity of infection of 0.001. Absorbancy at 600 nm was measured every 30 min using a spectrophotometer (722 N, Jingke Instrument). Analytical ultracentrifugation. The oligomerizations of wild type AimR, All the growth dynamics curves were drawn using Origin 8.0. arbitrium peptide-bound AimR and C-terminal-deleted AimR were investigated using AUC experiments, which were performed in a Beckman Coulter XL-I SAXS measurements. For SAXS experiments, the AimR protein was prepared analytical ultracentrifuge using two-channel centrepieces. AimR, GMPRGA at 100 μM in a buffer containing 150 mM NaCl and 25 mM Tris-HCl pH 8.0 with peptide-AimR (molar ratio = 10:1) and AimR ΔC truncation (residues 1–373) were or without 200 μM peptide. Mutant AimR (R228A/D360A) and wild-type AimR in a solution of 25 mM Tris-HCl (pH 8.0) and 150 mM NaCl. Data were collected were similarly treated. SAXS data for the protein solutions were collected at the via absorbance detection at 18 °C for proteins at a concentration of 1 mg ml−1 SSRF using the BL19U2 beamline at room temperature. For each measurement, 20 and rotor speed of 45,000 r.p.m. The SV-AUC data were globally analysed using consecutive frames with 1 s exposure time were recorded, averaged after checking the SEDFIT program and were fitted to a continuous c(s) distribution model to that there was no difference between the first and last frames of the SAXS data determine the molecular mass of each protein. Data analysis was carried out using and processed using the RAW software (bioxtasraw.sourceforge.net). Similarly, the GraphPad Prism7. background data were recorded using sample buffer and were subtracted from the protein patterns. Static light scattering experiment. Wild-type AimR, arbitrium peptide- bound AimR and C-terminal-deleted AimR (approximately 1 mg ml−1) were Reporting summary. Further information on research design is available in the independently loaded onto a Superdex 200 increased 10/300 column connected Nature Research Reporting Summary linked to this article. to a HELEOS multi-angle light scattering instrument (WYATT Technology). Next, 5 mg ml−1 AimR, GMPRGA peptide-AimR (molar ratio = 10:1) and AimR ΔC mutant (residues 1–373) were eluted in a buffer of 25 mM Tris-HCl Data availability (pH 8.0) and 150 mM NaCl at a flow rate of 0.5 ml min−1. Each fraction was Coordinates for the atomic structures have been deposited in the RCSB Protein automatically analysed using multi-angle light scattering. The figure was drawn Data Bank under PDB codes 5XYB and 5Y24. The data that support the findings of using Origin 8.0. this study are available from the corresponding author on request.

Nature Microbiology | www.nature.com/naturemicrobiology Articles Nature Microbiology

Received: 20 November 2017; Accepted: 6 August 2018; 29. Kozlowicz, B. K. et al. Molecular basis for control of conjugation by bacterial Published: xx xx xxxx pheromone and inhibitor peptides. Mol. Microbiol. 62, 958–969 (2006). 30. Declerck, N. et al. Structure of PlcR: Insights into virulence regulation and evolution of quorum sensing in Gram-positive bacteria. Proc. Natl Acad. Sci. References USA 104, 18490–18495 (2007). 1. Oppenheim, A. B., Kobiler, O., Stavans, J., Court, D. L. & Adhya, S. Switches 31. Bouillaut, L. et al. Molecular basis for group-specifc activation of the in bacteriophage λ development. Annu. Rev. Genet. 39, 409–429 (2005). virulence regulator PlcR by PapR heptapeptides. Nucleic Acids Res. 36, 2. Golding, I. Single-cell studies of phage λ: hidden treasures under Occam’s 3791–3801 (2008). rug. Annu. Rev. Virol. 3, 453–472 (2016). 32. Baker, M. D. & Neiditch, M. B. Structural basis of response regulator 3. Zeng, L. et al. Decision making at a subcellular level determines the outcome inhibition by a bacterial anti-activator protein. PLoS Biol. 9, e1001226 (2011). of bacteriophage infection. Cell 141, 682–691 (2010). 33. Parashar, V., Aggarwal, C., Federle, M. J. & Neiditch, M. B. Rgg protein 4. Golding, I. Decision making in living cells: lessons from a simple system. structure-function and inhibition by cyclic peptide compounds. Proc. Natl Annu. Rev. Biophys. 40, 63–80 (2011). Acad. Sci. USA 112, 5177–5182 (2015). 5. Echols, H. & Green, L. Establishment and maintenance of repression by 34. Makthal, N. et al. Structural and functional analysis of RopB: a major bacteriophage λ: the role of the cI, cII, and c3 proteins. Proc. Natl Acad. Sci. virulence regulator in Streptococcus pyogenes. Mol. Microbiol. 99, USA 68, 2190–2194 (1971). 1119–1133 (2016). 6. Kobiler, O. et al. Quantitative kinetic analysis of the bacteriophage λ genetic 35. Talagas, A. et al. Structural insights into streptococcal competence network. Proc. Natl Acad. Sci. USA 102, 4470–4475 (2005). regulation by the cell-to-cell communication system ComRS. PLoS Pathog. 7. Hoyt, M. A., Knight, D. M., Das, A., Miller, H. I. & Echols, H. Control 12, e1005980 (2016). of phage lambda development by stability and synthesis of cII protein: 36. Deng, Y. et al. ApnI, a transmembrane protein responsible for subtilomycin role of the viral cIII and host hfA, himA and himD genes. Cell 31, immunity, unveils a novel model for lantibiotic immunity. Appl. Environ. 565–573 (1982). Microbiol. 80, 6303–6315 (2014). 8. Erez, Z. et al. Communication between viruses guides lysis–lysogeny 37. Zahler, S. A., Korman, R. Z., Rosenthal, R. & Hemphill, H. E. Bacillus subtilis decisions. Nature 541, 488–493 (2017). bacteriophage SPbeta: localization of the prophage attachment site, and 9. Davidson, A. R. Virology: phages make a group decision. Nature 541, specialized transduction. J. Bacteriol. 129, 556–558 (1977). 466–467 (2017). 38. Wang, Q. S. et al. Te macromolecular crystallography beamline of SSRF. 10. Hynes, A. P. & Moineau, S. Phagebook: the social network. Mol. Cell 65, Nucl. Sci. Tech. 26, 12–17 (2015). 963–964 (2017). 39. Wang, Q.-S. et al. Upgrade of macromolecular crystallography beamline 11. Perez-Pascual, D., Monnet, V. & Gardan, R. Bacterial cell-cell BL17U1 at SSRF. Nucl. Sci. Tech. 29, 68 (2018). communication in the host via RRNPP peptide-binding regulators. 40. Otwinowski, Z. & Minor, W. Processing of X-ray difraction data collected in Front. Microbiol. 7, 706 (2016). oscillation mode. Methods Enzymol. 276, 307–326 (1997). 12. Do, H. & Kumaraswami, M. Structural mechanisms of peptide recognition 41. Winn, M. D. et al. Overview of the CCP4 suite and current developments. and allosteric modulation of gene regulation by the RRNPP family of Acta Crystallogr. D 67, 235–242 (2011). quorum-sensing regulators. J. Mol. Biol. 428, 2793–2804 (2016). 42. McCoy, A. J. et al. Phaser crystallographic sofware. J. Appl. Crystallogr. 40, 13. Rocha-Estrada, J., Aceves-Diez, A. E., Guarneros, G. & de la Torre, M. 658–674 (2007). Te RNPP family of quorum-sensing proteins in Gram-positive bacteria. 43. Adams, P. D. et al. PHENIX: a comprehensive Python-based system for Appl. Microbiol. Biotechnol. 87, 913–923 (2010). macromolecular structure solution. Acta Crystallogr. D 66, 213–221 (2010). 14. Neiditch, M. B., Capodagli, G. C., Prehna, G. & Federle, M. J. Genetic and 44. Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. structural analyses of RRNPP intercellular peptide signaling of Gram-positive Acta Crystallogr. D 60, 2126–2132 (2004). bacteria. Annu. Rev. Genet. 51, 311–333 (2017). 45. Wach, A. PCR-synthesis of marker cassettes with long fanking homology 15. Gallego del Sol, F. & Marina, A. Structural basis of Rap phosphatase regions for gene disruptions in S. cerevisiae. Yeast 12, 259–265 (1996). inhibition by Phr peptides. PLoS Biol. 11, e1001511 (2013). 16. Parashar, V., Jefrey, P. D. & Neiditch, M. B. Conformational change-induced repeat domain expansion regulates Rap phosphatase quorum-sensing signal Acknowledgements receptors. PLoS Biol. 11, e1001512 (2013). We thank the staff of the BL17U1/BL19U1/BL19U2 beamline of the NCPSS at the 17. Zouhir, S. et al. Peptide-binding dependent conformational changes regulate Shanghai Synchrotron Radiation Facility for assistance during data collection and the transcriptional activity of the quorum-sensor NprR. Nucleic Acids Res. 41, research associates at the Center for Protein Research, Huazhong Agricultural University, 7920–7933 (2013). for technical support. We thank Z. Dong and Y. Deng for helpful discussions and gene 18. Grenha, R. et al. Structural basis for the activation mechanism of the PlcR disruption in prophages. We also thank Q. Tang for help with the DNA I footprinting virulence regulator by the quorum-sensing signal peptide PapR. Proc. Natl assay. This work was supported by funds from the National Key R&D Program of China Acad. Sci. USA 110, 1047–1052 (2013). (2018YFA0507700), the National Natural Science Foundation of China (grant no. 19. Parashar, V., Mirouze, N., Dubnau, D. A. & Neiditch, M. B. Structural basis of 81601742 for T.Z.) and the Fundamental Research Funds for the Central Universities response regulator dephosphorylation by Rap phosphatases. PLoS Biol. 9, (Program No. 2662014BQ028 for T.Z. and 2662017PY031 for P.Y.). e1000589 (2011). 20. Holm, L., Kaariainen, S., Rosenstrom, P. & Schenkel, A. Searching protein Author contributions structure databases with DaliLite v.3. Bioinformatics 24, 2780–2781 (2008). Q.W., Z.G., P.Y., Z.L., D.P. and T.Z. designed the project and analysed data. Q.W. and 21. Krissinel, E. & Henrick, K. Inference of macromolecular assemblies from Z.G. contributed to protein production and crystallization. Q.W. and K.P. performed crystalline state. J. Mol. Biol. 372, 774–797 (2007). EMSA, AUC, SLS, ITC, size-exclusion chromatography, and crosslinking and DNase 22. Waters, C. M. & Bassler, B. L. Quorum sensing: cell-to-cell communication in I footprinting assays. Q.W., J.W. and T.Z performed gene disruption and in vivo bacteria. Annu. Rev. Cell Dev. Biol. 21, 319–346 (2005). experiments. Z.L. performed SAXS measurements. Q.W., Z.G. and T.Z. wrote the paper. 23. Bassler, B. L. & Losick, R. Bacterially speaking. Cell 125, 237–246 (2006). All authors contributed to editing of the manuscript. 24. Perego, M. Forty years in the making: understanding the molecular mechanism of peptide regulation in bacterial development. PLoS Biol. 11, Competing interests e1001516 (2013). The authors declare no competing interests. 25. Monnet, V. & Gardan, R. Quorum-sensing regulators in Gram-positive bacteria: ‘cherchez le peptide’. Mol. Microbiol. 97, 181–184 (2015). 26. Monnet, V., Juillard, V. & Gardan, R. Peptide conversations in Gram-positive Additional information bacteria. Crit. Rev. Microbiol. 42, 339–351 (2016). Supplementary information is available for this paper at https://doi.org/10.1038/ 27. Core, L. & Perego, M. TPR-mediated interaction of RapC with ComA inhibits s41564-018-0239-y. response regulator-DNA binding for competence development in Bacillus subtilis. Mol. Microbiol. 49, 1509–1522 (2003). Reprints and permissions information is available at www.nature.com/reprints. 28. Shi, K. et al. Structure of peptide sex pheromone receptor PrgX and PrgX/ Correspondence and requests for materials should be addressed to T.Z. pheromone complexes and regulation of conjugation in Enterococcus faecalis. Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in Proc. Natl Acad. Sci. USA 102, 18596–18601 (2005). published maps and institutional affiliations.

Nature Microbiology | www.nature.com/naturemicrobiology nature research | life sciences reporting summary

Corresponding author(s): Tingting Zou Initial submission Revised version Final submission Life Sciences Reporting Summary Nature Research wishes to improve the reproducibility of the work that we publish. This form is intended for publication with all accepted life science papers and provides structure for consistency and transparency in reporting. Every life science submission will use this form; some list items might not apply to an individual manuscript, but all fields must be completed for clarity. For further information on the points included in this form, see Reporting Life Sciences Research. For further information on Nature Research policies, including our data availability policy, see Authors & Referees and the Editorial Policy Checklist.

` Experimental design 1. Sample size Describe how sample size was determined. Trial experiments done previously were used to determine sample size with adequate statistical power. 2. Data exclusions Describe any data exclusions. No data were excluded from the analyses. 3. Replication Describe whether the experimental findings were All attempts at replicating the results were successful. reliably reproduced. 4. Randomization Describe how samples/organisms/participants were B. subtilis CU1065 or aimR-overexpressing strains were randomly assigned to allocated into experimental groups. groups for in vivo experiments. 5. Blinding Describe whether the investigators were blinded to The investigators were not completely blinded during the experiment. group allocation during data collection and/or analysis. Note: all studies involving animals and/or human research participants must disclose whether blinding and randomization were used.

6. Statistical parameters For all figures and tables that use statistical methods, confirm that the following items are present in relevant figure legends (or in the Methods section if additional space is needed). n/a Confirmed

The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement (animals, litters, cultures, etc.) A description of how samples were collected, noting whether measurements were taken from distinct samples or whether the same sample was measured repeatedly A statement indicating how many times each experiment was replicated The statistical test(s) used and whether they are one- or two-sided (note: only common tests should be described solely by name; more complex techniques should be described in the Methods section) A description of any assumptions or corrections, such as an adjustment for multiple comparisons The test results (e.g. P values) given as exact values whenever possible and with confidence intervals noted A clear description of statistics including central tendency (e.g. median, mean) and variation (e.g. standard deviation, interquartile range)

Clearly defined error bars June 2017

See the web collection on statistics for biologists for further resources and guidance.

1 ` Software nature research | life sciences reporting summary Policy information about availability of computer code 7. Software Describe the software used to analyze the data in this HKL2000 package, PHENIX 1.14rc3-3206-000, CCP4 interface 7.0.060, COOT, study. PyMOL, SEDFIT software, RAW software, Origin8.0, MicroCal Origin software, GraphPad Prism7 and Quantity One.

For manuscripts utilizing custom algorithms or software that are central to the paper but not yet described in the published literature, software must be made available to editors and reviewers upon request. We strongly encourage code deposition in a community repository (e.g. GitHub). Nature Methods guidance for providing algorithms and software for publication provides further information on this topic.

` Materials and reagents Policy information about availability of materials 8. Materials availability Indicate whether there are restrictions on availability of There are no restrictions on availability of unique materials used in the study. unique materials or if these materials are only available for distribution by a for-profit company. 9. Antibodies Describe the antibodies used and how they were validated No antibodies were used. for use in the system under study (i.e. assay and species). 10. Eukaryotic cell lines a. State the source of each eukaryotic cell line used. No eukaryotic cell lines were used.

b. Describe the method of cell line authentication used. No eukaryotic cell lines were used.

c. Report whether the cell lines were tested for No eukaryotic cell lines were used. mycoplasma contamination.

d. If any of the cell lines used are listed in the database No commonly misidentified cell lines were used. of commonly misidentified cell lines maintained by ICLAC, provide a scientific rationale for their use.

` Animals and human research participants Policy information about studies involving animals; when reporting animal research, follow the ARRIVE guidelines 11. Description of research animals Provide details on animals and/or animal-derived No animals were used. materials used in the study.

Policy information about studies involving human research participants 12. Description of human research participants Describe the covariate-relevant population The study did not involve human research participants. characteristics of the human research participants. June 2017