Rap-Phr Systems from Plasmids pAW63 and pHT8-1 Act Together To Regulate Sporulation in the Bacillus thuringiensis Serovar kurstaki HD73 Strain

The life cycle of Bacillus thuringiensis in insect larvae is regulated by quorum-sensing systems of the RNPP family. After the toxemia caused by Cry insecticidal toxins, the sequential activation of these systems allows the bacterium to trigger first a state of virulence (regulated by PlcR-PapR) and then a necrotrophic lifestyle (regulated by NprR-NprX); ultimately, sporulation is controlled by the Rap-Phr systems. Our study describes a new rap-phr operon carried by a B. thuringiensis plasmid and shows that the Rap protein has a moderate effect on sporulation. However, this system, in combination with another plasmidic rap-phr operon, provides effective control of sporulation when the bacteria develop in the cadavers of infected insect larvae. Overall, this study highlights the important adaptive role of the plasmid Rap-Phr systems in the developmental fate of B. thuringiensis and its survival within its ecological niche.

ABSTRACT

Bacillus thuringiensis is a Gram-positive spore-forming bacterium pathogenic to various insect species. This property is due to the Cry toxins encoded by plasmid genes and mostly produced during sporulation. B. thuringiensis contains a remarkable number of extrachromosomal DNA molecules and a great number of plasmid rap-phr genes. Rap-Phr quorum-sensing systems regulate different bacterial processes, notably the commitment to sporulation in Bacillus species. Rap proteins are quorum sensors acting as phosphatases on Spo0F, an intermediate of the sporulation phosphorelay, and are inhibited by Phr peptides that function as signaling molecules. In this study, we characterize the Rap63-Phr63 system encoded by the pAW63 plasmid from the B. thuringiensis serovar kurstaki HD73 strain. Rap63 has moderate activity on sporulation and is inhibited by the Phr63 peptide. The rap63-phr63 genes are cotranscribed, and the phr63 gene is also transcribed from a σ^H-specific promoter. We show that Rap63-Phr63 regulates sporulation together with the Rap8-Phr8 system harbored by plasmid pHT8_1 of the HD73 strain. Interestingly, the deletion of both phr63 and phr8 genes in the same strain has a greater negative effect on sporulation than the sum of the loss of each phr gene. Despite the similarities in the Phr8 and Phr63 sequences, there is no cross talk between the two systems. Our results suggest a synergism of these two Rap-Phr systems in the regulation of the sporulation of B. thuringiensis at the end of the infectious cycle in insects, thus pointing out the roles of the plasmids in the fitness of the bacterium.

IMPORTANCE The life cycle of Bacillus thuringiensis in insect larvae is regulated by quorum-sensing systems of the RNPP family. After the toxemia caused by Cry insecticidal toxins, the sequential activation of these systems allows the bacterium to trigger first a state of virulence (regulated by PlcR-PapR) and then a necrotrophic lifestyle (regulated by NprR-NprX); ultimately, sporulation is controlled by the Rap-Phr systems. Our study describes a new rap-phr operon carried by a B. thuringiensis plasmid and shows that the Rap protein has a moderate effect on sporulation. However, this system, in combination with another plasmidic rap-phr operon, provides effective control of sporulation when the bacteria develop in the cadavers of infected insect larvae. Overall, this study highlights the important adaptive role of the plasmid Rap-Phr systems in the developmental fate of B. thuringiensis and its survival within its ecological niche.

INTRODUCTION

Bacillus thuringiensis belongs to the Bacillus cereus group of Gram-positive, rod-shaped, spore-forming bacteria and is distinguished from the other closely related species of this group by the production of crystal inclusions toxic to the larvae of various insects (1). Due to this entomopathogenic property, B. thuringiensis is widely used as a biopesticide to control agricultural pests, mainly lepidopteran and coleopteran insects, or human disease vectors, such as mosquitoes (2, 3). Crystal inclusions consist of insecticidal proteins encoded by cry and cyt genes located on plasmids and generally transcribed by sporulation sigma factors (4). B. thuringiensis strains have been shown to carry a complex plasmid pattern, with as many as 17 different extrachromosomal elements with sizes ranging from 2 to 600 kb (5–7). Because of their biotechnological relevance, plasmids harboring cry genes have been the most studied (8–10). Recent studies have shown that a number of Bacillus plasmids harbor genes encoding various Rap-Phr systems (11–16). Rap-Phr quorum-sensing systems were first described in Bacillus subtilis, in which they regulate various processes, such as sporulation, competence, transfer of mobile genetic elements, production of proteases, and biofilm formation (17).

The Rap proteins, which belong to the RNPP family of quorum-sensing regulators from Gram-positive bacteria, consist of a response regulator with TPR (tetratricopeptide repeat) domains forming a hydrophobic pocket able to bind the signaling peptide. This binding induces a conformational change and modulates regulator activities (17–19). The infectious cycle of B. thuringiensis in the insect is sequentially regulated by three RNPP regulators and their cognate peptides (20): first, PlcR-PapR regulates the virulence stage; next, NprR-NprX controls the necrotrophic stage and the transition to sporulation; and ultimately, Rap-Phr controls the initiation of the sporulation process.

The commitment to sporulation is regulated by the phosphorylation state of the major response regulator Spo0A (21, 22). Different signals, such as nutritional deprivation, are recognized by sporulation kinases, which are able to autophosphorylate (23). These kinases phosphorylate Spo0F, which is used as a substrate by the phosphotransferase Spo0B to phosphorylate Spo0A (24). Response regulator aspartate phosphatases (Rap) inhibit this signal transduction pathway by dephosphorylating the Spo0F-P response regulator (25). Rap protein activity is inhibited by its cognate Phr peptide, which is translated in a premature form that needs to be secreted, processed, and reimported by oligopeptide permeases to be active (26–28). The mature Phr peptides contain 5, 6, or 7 amino acids (12, 27, 29, 30).

A number of Rap-Phr systems are found in bacteria of the B. cereus group, particularly in B. thuringiensis plasmids, and mainly have an effect on sporulation (31, 32). The pXO1 pathogenicity plasmid from B. anthracis hosts a Rap-Phr system that regulates the sporulation process (11). A recent study revealed that plasmid Rap-Phr systems of B. thuringiensis strain Bt8741 regulate sporulation, biofilm formation, spreading, and extracellular proteolytic activity (33). Moreover, the Rap8-Phr8 system of pHT8_1, a cryptic plasmid of B. thuringiensis serovar kurstaki strain HD73, inhibits sporulation and biofilm formation (12). The effect of this system on sporulation was particularly evident in bacteria developing in insect larvae, the presumed ecological niche of B. thuringiensis (12). The serovar kurstaki strain HD73 harbors six other plasmids, among which pHT77 and pAW63 also carry rap-phr genes (31, 34). The Rap-Phr system from pHT77 does not affect sporulation, and its function remains unknown (31). The pAW63 plasmid harbors a Rap-Phr system (35) whose involvement in sporulation has been predicted in silico (31). pAW63 is a theta-replicating conjugative plasmid of 72 kb (36, 37) that efficiently conjugates and mobilizes nonconjugative plasmids in food matrices and under adverse conditions (38, 39). Furthermore, pAW63 shares a common backbone with the pXO2 pathogenic plasmid of B. anthracis and with plasmid pBT9727 from the pathogenic B. thuringiensis serovar konkukian strain 97-27, including replication and transfer modules (35).

In this study, we characterized the Rap63-Phr63 quorum-sensing system from the pAW63 plasmid. Our results show that it acts together with Rap8-Phr8 to synergistically regulate sporulation in vitro and in vivo. However, despite important Phr sequence similarities, no cross talk was detected between these closely related Rap-Phr systems, revealing a high specificity of the Phr peptides for their cognate Rap proteins. These results highlight the importance of plasmid-borne Rap-Phr systems in improving the fitness of the bacteria under naturalistic conditions.

(This research was conducted partly by P. Cardoso in partial fulfillment of the requirements for a doctoral degree from the Université Paris-Saclay, France, and the Universidade Estadual de Londrina, Brazil [40]).

RESULTS

Transcriptional analysis of the rap and phr genes.

In B. subtilis, the rap and phr genes have been shown to form a transcriptional unit with a promoter located upstream of the rap gene (27). As generally described for these genes, the phr63 gene overlaps the rap63 gene by 1 bp. To determine whether the rap63 and phr63 genes are cotranscribed, reverse transcription-PCR (RT-PCR) was performed on RNA extracted 3 h after the onset of the stationary phase in HCT, a sporulation-specific medium (5). The result demonstrates that the rap63 and phr63 genes are cotranscribed from a promoter located upstream of the rap63 gene and that these two genes form a transcriptional unit (Fig. 1A and B).

FIG 1.

The rap63-phr63 transcription unit. (A) Schematic representation of the rap63-phr63 locus in the pAW63 plasmid. Arrowheads correspond to the primers used to amplify the three RT-PCR fragments (fragments A, B, and C). (B) RT-PCR experiment. Total RNA was extracted from a t3 culture in HCT medium at 37°C. Genomic DNA (1), RNA (2), and cDNA (3) were used as templates for PCR amplification, analyzed in a 1% agarose gel, and compared to molecular weight markers (M) (SmartLadder small fragments; Eurogentec). (C) Kinetics of rap63-phr63 expression. Shown are the β-galactosidase activities of the HD73 wild-type (wt), Δspo0A, and ΔsigH strains carrying P_rap63′-lacZ in HCT medium at 37°C. (D) Kinetics of phr63 expression. Shown are the β-galactosidase activities of the HD73 wild-type, Δspo0A, and ΔsigH strains carrying P_phr63′-lacZ in HCT medium at 37°C. Time zero corresponds to the entry into stationary phase. Error bars represent the standard errors of the means.

To monitor the expression of the rap63-phr63 operon, a 659-bp DNA fragment upstream of the rap63 gene was fused to the lacZ reporter gene on the low-copy-number plasmid pHT304-18Z. In the wild-type HD73 strain, transcription from this promoter region (P_rap63) started 2 h after the onset of the stationary phase (t2), increased sharply, and reached a maximum at t4 (Fig. 1C). In the sigH mutant strain, the expression level was higher and started earlier (t1) than in the wild-type strain. In the spo0A mutant strain, expression started at t2 but with a slope slightly lower than for the wild-type strain. These results demonstrate that expression of the rap63-phr63 operon is slightly activated by Spo0A and repressed by the sigma factor SigH. No Spo0A boxes and no SigH consensus sequences were found in the P_rap63 promoter region, suggesting an indirect effect of Spo0A and SigH on the transcription of the rap63-phr63 operon.

The transcription of phr genes is often regulated by an additional specific promoter situated upstream of the phr gene and inside the rap gene, and generally controlled by the alternative sigma factor sigma H (σ^H, or SigH) (27, 41). A 562-bp DNA region (P_phr63) upstream of phr63 was cloned into plasmid pHT304-18Z in order to determine whether this gene is transcribed from such a specific promoter. In the wild-type HD73 strain, expression from P_phr63 started during late-exponential growth and was activated from t2 to t4 (Fig. 1D). In the HD73 ΔsigH and Δspo0A mutant strains, P_phr63′-lacZ expression was significantly reduced, and the activation observed at t2 in the wild-type strain was lacking. Sequences corresponding to a –35 (GCAGGAATT) and a degenerated -10 (AAAGAAG) SigH consensus box (42) were identified in the P_phr63 promoter region. These sequences, predicted to be recognized by SigH, are located 142 bp upstream of the start codon of the phr gene. These results demonstrate that the phr63 gene is also transcribed from a specific promoter that might be regulated by the SigH factor. The highly SigH-dependent expression of phr63 may boost the production of signaling peptides to repress Rap63 activity in stationary phase. However, P_phr63′-lacZ expression was not abolished in the HD73 ΔsigH mutant strain (Fig. 1D), indicating that phr63 can also be transcribed during stationary phase from a SigH-independent promoter.

Rap63 negatively affects sporulation.

Our previous in silico analyses of Rap-Phr systems from the B. cereus group predicted that Rap63 could affect sporulation (31). To verify this prediction, rap63, rap63-phr63, and phr63 genes were introduced into plasmid pHT315xyl (43), a multicopy plasmid in which the expression of the cloned genes is under the regulation of a xylose-inducible promoter (P_xylA). These constructions were transformed into the B. thuringiensis HD73 wild-type strain. The control strain, bearing the empty pHT315xyl plasmid, sporulated efficiently (82% of spores) after 48 h at 30°C in HCT medium supplemented with xylose (Fig. 2; see also Table S1 in the supplemental material). In sharp contrast, the strain expressing the rap63 gene presented a reduced sporulation efficiency (38% versus 82%) and a 5-fold reduction in the production of heat-resistant spores (5.80E+07 versus 3.25E+08 in the control strain). This greater reduction is not due to an effect of rap63 overexpression on growth (Fig. S1) but reflects the low viability of the bacteria that do not sporulate. The strains expressing the rap63-phr63 or phr63 gene presented sporulation rates similar to that of the control strain (84% and 85%, respectively). Therefore, these results confirm that Rap63 inhibits sporulation moderately and that Rap63 activity is counteracted by its cognate Phr63 peptide.

FIG 2.

Rap63 negatively affects sporulation. Sporulation efficiency was measured in the HD73 control strain (315xyl) and in HD73 strains expressing rap63 (xyl_rap63), rap-phr63 (xyl_rap63-phr63), or phr63 (xyl_phr63) in HCT medium at 30°C. After 48 h of growth, the percentage of spores was calculated as 100 multiplied by the ratio between the number of heat-resistant spores per milliliter and the number of total viable cells per milliliter. Error bars represent standard errors of the means. Experimental values are detailed in Table S1. The letters “a” and “b” above the bars indicate significant differences in mean values (P < 0.001).

Rap63 delays the expression of Spo0A-regulated genes.

Rap proteins prevent the phosphorylation of Spo0A, and the phosphorylated form of Spo0A (Spo0A-P) activates the expression of several genes related to sporulation, such as the spoIIE gene, which is transcribed at the onset of stationary phase in B. subtilis (21) and B. thuringiensis (44). This gene can therefore be used as a reporter to measure the presence of Spo0A-P in the bacterial cell. The promoter region of spoIIE (P_spoIIE) was fused to the yfp fluorescent reporter gene (encoding yellow fluorescent protein [YFP]) and inserted into the pHT315xylΩrap63 and pHT315xylΩrap63-phr63 plasmids to yield plasmids pHT315xylΩrap63_P_spoIIE′-yfp and pHT315xylΩrap63-phr63_P_spoIIE′-yfp, respectively (see Table 2). In the control strain (harboring plasmid pHT315xyl_P_spoIIE′-yfp), the transcription of spoIIE started 2 h after the onset of the stationary phase (Fig. 3). In the strain harboring the rap63 gene, expression from P_spoIIE was strongly delayed, beginning around t5. When both Rap63 and Phr63 were produced, expression from P_spoIIE was restored to the same kinetics and level as in the control strain. These results demonstrate that Rap63 delays the expression of Spo0A-regulated genes and that Phr63 inhibits Rap63 activity. This is in accordance with the sporulation experiments, suggesting that the moderate effect of Rap63 on the control of the sporulation process might be due to the delayed transcription of Spo0A-regulated genes.

TABLE 2.

Plasmid constructions used in this study

Plasmid	Description
pHT315xylΩrap63 (xyl_rap63)	The rap63 gene was amplified using primer set Rap7557-F/Rap7557-R with B. thuringiensis HD73 genomic DNA as the template. The fragment was inserted between the BamHI and HindIII sites downstream of the inducible PxylA promoter carried by plasmid pHT315xyl (42).
pHT315xylΩphr63 (xyl_phr63)	The phr63 gene was amplified using primer set Phr7557-F/Phr7557-R with B. thuringiensis HD73 genomic DNA as the template. The fragment was inserted between the BamHI and HindIII sites of the pHT315xyl plasmid.
pHT315xylΩrap63-phr63 (xyl_rap63-phr63)	The rap-phr63 genes were amplified using primer set Rap7557-F/Phr7557-R with B. thuringiensis HD73 genomic DNA as the template. The fragment was inserted between the BamHI and HindIII sites of the pHT315xyl plasmid.
pHT315xylΩrap63-phr63_R3 (xyl_rap63-phr63_R3)	The rap63 gene, together with a truncated phr63 gene, was amplified using primer set Rap7557-F/Phr7557R3 with B. thuringiensis HD73 genomic DNA as the template. The fragment was inserted between the BamHI and HindIII sites of the pHT315xyl plasmid.
pHT304-18-Prap63′-lacZ	Primers Prom7557-F and Prom7557-R were used to amplify the promoter region of the rap63 gene, using B. thuringiensis HD73 genomic DNA as the template. The fragment of 659 bp was inserted between the HindIII and BamHI sites of pHT304-18Z (65).
pHT304-18-Pphr63′-lacZ	Primers Prom7557Phr-F and Prom7557Phr-R were used to amplify the promoter region of the phr63 gene, using B. thuringiensis HD73 genomic DNA as the template. The fragment of 562 bp was inserted between the HindIII and BamHI sites of pHT304-18Z.
pHT315xyl-PspoIIE′-yfp	Plasmid described by Fazion and colleagues (12)
pHT315xylΩrap63_PspoIIE′-yfp	The PspoIIE′-yfp fragment was amplified using primer set PU-EcoRI/YFP-R with the pHT315xyl-PspoIIE′-yfp plasmid as the template, and the fragment was inserted into the EcoRI site of the pHT315xylΩrap63 plasmid. To avoid the influence of the PxylA promoter, the orientation of the inserted fragment was verified by PCR using primer set PspoIIE-F/xylRout3′.
pHT315xylΩrap63-phr63_PspoIIE′-yfp	The PspoIIE′-yfp fragment was amplified using primer set PU-EcoRI/YFP-R with the pHT315xyl-PspoIIE′-yfp plasmid as the template, and the fragment was inserted into the EcoRI site of the pHT315xylΩrap63-phr63 plasmid. To avoid the influence of the PxylA promoter, the orientation of the inserted fragment was verified by PCR using primer set PspoIIE-F/xylRout3′.
pMADΩrap63-phr63::spec	The 5′ and 3′ regions of rap63-phr63 genes were amplified using primer sets 7557Amont1-F/7557Amont1-R and 7557Aval-F/Aval7557-R, respectively, with B. thuringiensis HD73 genomic DNA as the template. The 5′ end was purified as an NcoI/KpnI fragment and the 3′ end as an XbaI/EcoRI fragment. The spectinomycin resistance gene was purified as a KpnI/XbaI fragment from pUC18Ωspec (laboratory stock) and was inserted together with the 5′ and 3′ regions of rap63-phr63 between the NcoI and EcoRI sites of the thermosensitive plasmid pMAD (66).
pMADΩphr63::spec	The 5′ and 3′ regions of phr63 genes were amplified using primer sets 7557Amont2-F/7557Amont2-R and 7557Aval-F/Aval7557-R, respectively, with B. thuringiensis HD73 genomic DNA as the template. The 5′ end was purified as an NcoI/KpnI fragment and the 3′ end as an XbaI/EcoRI fragment. The spectinomycin resistance cassette was purified as a KpnI/XbaI fragment from pUC18Ωspec and was inserted together with the 5′ and 3′ regions of phr63 between the NcoI and EcoRI sites of pMAD.
xyl_rap8	Plasmid pHT-xylR from the work of Fazion and colleagues (12)
xyl_rap8-phr8	pHT-xylRP from the work of Fazion and colleagues (12)

FIG 3.

Rap63 delays the expression of a Spo0A-regulated gene. Shown is the kinetics of spoIIE expression in HD73 wild-type strains carrying the pHT315xyl_P_spoIIE′-yfp, pHT315xylΩrap63_P_spoIIE′-yfp, or pHT315xylΩrap63-phr63_P_spoIIE′-yfp plasmid. YFP fluorescence was measured during growth in HCT medium at 30°C in the presence of 20 mM xylose added at t0 (entry into stationary phase). The results are expressed in arbitrary units per OD₆₀₀ unit. Error bars represent the standard errors of the means.

The Δphr8 Δphr63 double mutation negatively affects the commitment to sporulation.

The complete infectious cycle—pathogenesis, necrotrophism, and sporulation—of B. thuringiensis in insect larvae has been shown to be controlled by quorum-sensing systems of the RNPP family (20). Hence, we analyzed the role of Rap63-Phr63 under such naturalistic conditions, in larvae of the lepidopteran insect Galleria mellonella. For this purpose, HD73 mutant strains with rap63-phr63 or phr63 deletions were constructed, and we also used the Δrap8-phr8 and Δphr8 mutant strains constructed previously (12). A double phr mutant strain (Δphr8 Δphr63) was also constructed to investigate the cumulative effect of the two Rap proteins in the absence of their cognate signaling peptides. The sporulation efficiency of each strain was evaluated in dead larvae 96 h after intrahemocoelic injection (Fig. 4A; Table S1), a mode of infection that bypasses the intestinal barrier and therefore does not require the presence of the Cry toxins (1). The wild-type strain (HD73) and the Δrap63-phr63, Δphr63, and Δrap8-phr8 mutant strains had similar sporulation efficiencies (22.3%, 24.5%, 16.7%, and 23.6% of spores, respectively). However, sporulation efficiency was significantly reduced in the Δphr8 mutant strain (3.1%) and to a greater extent in the Δphr8 Δphr63 double mutant strain (0.23%). The results obtained in vitro (Fig. 4B; Table S1), in HCT medium, showed that the Δrap63-phr63 and Δphr63 strains sporulated like the wild-type strain, whereas sporulation was slightly but significantly affected in the Δphr8 Δphr63 strain. Under both conditions, the effect on sporulation caused by the loss of the two phr genes was greater than the sum of the loss of the single phr genes. These results suggest a synergistic effect of the Rap8-Phr8 and Rap63-Phr63 systems to regulate the sporulation process in insect larvae.

FIG 4.

Synergistic activity of the Rap63-Phr63 and Rap8-Phr8 systems on sporulation. Shown are the sporulation efficiencies of the HD73 wild-type strain and of Δrap63-phr63, Δphr63, Δrap8-phr8, Δphr8, and Δphr8 Δphr63 mutant strains. (A) In dead larvae of G. mellonella (in vivo). Viable cells and heat-resistant spores were counted in dead larvae 4 days postinfection at 30°C. (B) In HCT medium (in vitro), viable cells and heat-resistant spores were counted after 48 h of growth at 30°C. The percentages of spores were calculated as 100 multiplied by the ratio between the number of heat-resistant spores per milliliter and the number of total viable cells per milliliter. Error bars represent standard errors of the means. Experimental values are detailed in Table S1. Asterisks indicate significant differences from values for the wild type (*, P < 0.05; **, P < 0.01).

Determination of the active form of Phr63.

Rap proteins are inhibited by Phr oligopeptides, whose active form generally corresponds to the C-terminal end of the Phr peptide (27). To determine whether the mature form of Phr63 corresponds to the C-terminal end of the propeptide, we constructed a plasmid carrying the rap63 gene and a 3′-end-truncated phr63 gene (xyl_rap63-phr63_R3′) expressing a Phr peptide lacking the six C-terminal amino acids (Fig. 5A). The sporulation efficiency of the strain harboring this plasmid (32%) was similar to that of the strain expressing the rap63 gene (Fig. 5B; Table S1). This result suggests that the active form of Phr63 is included in the C-terminal part of the premature Phr.

FIG 5.

Characterization of the Phr63 active form. (A) Amino acid sequences of the pre-Phr63 peptide and the five synthetic peptides used in complementation experiments. The C-terminal end sequence truncated in xyl_rap63-phr63_R3 is shown in red, with the positively charged histidine residue in blue. (B) Sporulation efficiencies of the HD73 control strain (315xyl) and of HD73 expressing rap63 (xyl-rap63) and rap63-phr63 producing C-terminally truncated Phr63 (xyl_rap63-phr63_R3). The culture medium of the strain expressing rap63 was supplemented with the synthetic peptide Phr63-4, Phr63-5, Phr63-6, Phr63-7, or Phr63-8 added independently 1 h after the onset of the stationary phase at a 50 μM final concentration. The percentages of spores were calculated as 100 multiplied by the ratio between the number of heat-resistant spores per milliliter and the number of total viable cells per milliliter after 48 h in HCT medium at 30°C in the presence of 20 mM xylose. Error bars represent standard errors of the means. Experimental values are detailed in Table S1. Different letters above the bars correspond to significant differences in values (P < 0.001).

To define the Phr63 active form, various Phr63 peptides corresponding to the C-terminal end were synthesized: GETI (Phr63-4), HGETI (Phr63-5), AHGETI (Phr63-6), YAHGETI (Phr63-7), and QYAHGETI (Phr63-8) (Fig. 5A). To evaluate the abilities of these synthetic peptides to inhibit Rap63 activity, they were separately added to the culture medium of the strain expressing the rap63 gene (xyl_rap63). Phr63-4 was not able to inhibit Rap63 activity (30% of spores), while Phr63-5 (77%), Phr63-6 (84%), Phr63-7 (87%), and Phr63-8 (87%) efficiently counteracted the effect of Rap63 on sporulation (Fig. 5B; Table S1). These results demonstrated that the minimal active form of Phr63 is the pentapeptide HGETI.

Specificity of the Rap-Phr63 and Rap-Phr8 systems.

The C-terminal parts of the Phr8 and Phr63 peptides, which include the mature peptides, are closely related (Fig. 6A), with only 2 divergent residues among the last 8 amino acids. Due to these sequence similarities and the sporulation results of the Δphr8 Δphr63 mutant strain compared to those of the Δphr8 and Δphr63 single mutant strains, we investigated the possibility of cross talk between the Rap8-Phr8 and Rap63-Phr63 systems. We constructed the Δrap8-phr8 Δrap63-phr63 mutant strain to avoid the effect of intrinsic systems. This mutant strain was transformed with the pHT315xyl (control strain), pHT315xylΩrap8, or pHT315xylΩrap63 plasmid, and the sporulation of the resultant strains (in whose designations the Δrap8-phr8 Δrap63-phr63 mutant is indicated by ΔΔ) was assessed (Fig. 6B; Table S1). The control strain (ΔΔ 315xyl) sporulated efficiently (84% of spores). The ΔΔ xyl-rap63 mutant strain, expressing rap63, presented reduced sporulation efficiency (32%), like the wild-type strain expressing rap63 (Fig. 2). The addition of the Phr63-7 peptide restored sporulation efficiency to a level similar to that of the control strain (Fig. 6B). However, in the presence of Phr8-6, Phr8-7, or Phr8-8, sporulation was similar to that of the strain expressing rap63 alone, indicating that none of the Phr8 peptides used was able to counteract the negative effect of Rap63 on sporulation. The sporulation of the ΔΔ xyl-rap8 strain, expressing rap8, was strongly reduced, and addition of the Phr8-7 peptide inhibited Rap8 activity (Fig. 6B), as already described for the wild-type strain (12). In contrast, none of the Phr63 peptides was able to counteract the negative effect of Rap8 on sporulation (Fig. 6B). Taken together, these results showed that despite the sequence similarities between Phr8 and Phr63, there is no cross talk between the Rap8-Phr8 and Rap63-Phr63 systems.

FIG 6.

Evaluation of cross talk between the Rap8-Phr8 and Rap63-Phr63 systems. (A) Comparison of the amino acid sequences of the Phr63 and Phr8 peptides. C-terminal end similarities are highlighted in red, and the two divergent residues are shown in black boldface. Phr8-6, Phr8-7, and Phr8-8 correspond to the synthetic Phr8 peptides used by Fazion and colleagues (12). (B) Sporulation efficiency of the HD73 Δrap8-phr8 Δrap63-phr63 control strain (ΔΔ 315xyl) or of this strain expressing rap63 (ΔΔ rap63) or rap8 (ΔΔ rap8) in HCT medium at 30°C in the presence of 20 mM xylose. The culture medium of the strain expressing rap63 was supplemented with the Phr8-6, Phr8-7, or Phr8-8 synthetic peptide, and the culture medium of the strain expressing rap8 was supplemented with the Phr63-6, Phr63-7, or Phr63-8 synthetic peptide. Peptides were added independently 1 h after the onset of the stationary phase at 50 μM final concentrations. The percentages of spores were calculated as 100 multiplied by the ratio between the number of heat-resistant spores per milliliter and the number of total viable cells per milliliter. Error bars represent standard errors of the means. Experimental values are detailed in Table S1. Different letters above the bars correspond to significant differences in the values (P < 0.001).

DISCUSSION

The complete genome sequence of the serovar kurstaki HD73 Cry⁺ strain (34) includes seven plasmids (pHT7, pHT8_1, pHT8_2, pHT11, pAW63, pHT73, pHT77). Among these, three harbor a Rap-Phr system (pHT8_1, pAW63, and pHT77). The Rap8-Phr8 quorum-sensing system from pHT8_1 is involved in biofilm formation and in the regulation of sporulation (12), but in contrast, Rap-Phr from pHT77 does not affect sporulation, and its role remains unknown (31). In this study, we characterized the Rap63-Phr63 system carried by plasmid pAW63.

Our results show that in sporulation medium, the rap63-phr63 operon is transcribed from a promoter activated during the stationary phase and partially repressed by SigH. The absence of a sequence resembling a SigH promoter upstream of the rap63 gene suggests indirect repression by a SigH-dependent repressor. In contrast, the expression measured in the spo0A mutant strain is similar to that observed in the wild-type strain. In B. subtilis, SigH is produced in a spo0A mutant, but to a lesser extent than in a wild-type strain (45). Therefore, SigH could be sufficiently produced in the B. thuringiensis spo0A mutant strain to activate the expression of the repressor. The phr63 gene is also transcribed separately from a SigH-dependent promoter located in the rap63 coding sequence. This regulation is similar to the expression of B. subtilis phr genes, which are also activated by SigH (41). This additional expression of the phr genes boosts the production of the signaling peptides and allows bacteria to trigger sporulation by inhibiting the Rap proteins. Transcriptional analysis of pAW63 coding sequences was performed previously at the mid-exponential-growth phase by comparing the expression pattern of the wild-type serovar kurstaki HD73 Cry⁻ strain with that of the same strain cured of the pAW63 plasmid (46). The results showed that the rap63 gene was expressed at a moderate level, while the phr63 gene was found at a high level. However, the Phr signal was also present at a high level in the cured strain, suggesting that the probe used to detect the phr63 gene was not specific and revealed the presence of another, related phr gene in the B. thuringiensis HD73 genome, possibly phr8. Both this previous transcriptional analysis and our results point out that the expression of rap63-phr63 genes—and presumably that of other rap-phr genes—is modulated by variations in environmental conditions. For example, the rap63 and phr63 genes are expressed from the mid-exponential-growth phase in Luria-Bertani (LB) medium (46), whereas they are expressed from the beginning of the stationary-growth phase in the sporulation-specific medium HCT.

Strains belonging to the B. cereus group have, on average, six Rap-Phr systems (31, 32). Even if there are redundant regulatory operons among these systems, they might not act simultaneously, or they might be regulated by different signals (31–33). A comparison with the results obtained by Fazion and colleagues (12) revealed that the transcription of rap63-phr63 genes occurs concomitantly with that of rap8-phr8 genes from the pHT8_1 plasmid, under the same conditions. Both Rap proteins from B. thuringiensis plasmids are able to inhibit the sporulation process, but to different extents. While Rap8 strongly inhibits sporulation (12), Rap63 induces a moderate inhibition of sporulation but significantly delays the expression of the spoIIE gene controlled by Spo0A. Various reasons may explain these different levels of inhibition: the copy numbers of the plasmids carrying the rap-phr genes (see below), the levels of gene expression, and the intrinsic properties of the Rap proteins, including affinity with their substrate (i.e., Spo0F).

In agreement with the analysis of Even-Tov and colleagues, which predicted that the dominant repression feature of Rap proteins enables the accumulation of multiple quorum-sensing systems with a synergistic effect (47), the Rap63-Phr63 and Rap8-Phr8 systems act together to regulate the initiation of sporulation. In shaken HCT medium, only strains lacking both phr genes sporulated less efficiently than the wild-type HD73 strain. The synergistic effect of the two Rap-Phr systems was more prominent under in vivo conditions. Indeed, the sporulation of the Δphr63 mutant strain was not significantly reduced, but the deletion of the two phr genes in the Δphr8 Δphr63 mutant strain strongly impaired the commitment to sporulation in vivo. In insect larvae, the sporulation efficiency of the Δphr8 strain was lower than that of the HD73 strain (about 8-fold), while in the double mutant strain, sporulation efficiency dropped by a factor of 100, highlighting a synergistic effect of the two quorum-sensing systems. We suggest that this apparent synergy of the two Rap-Phr systems depends on the cumulative effect of each Rap protein on the phosphorylation of Spo0A. Thus, the combined action of Rap8 and Rap63 reduces the concentration of Spo0A-P below a threshold that has a greater impact on sporulation than that expected on the basis of the independent action of each Rap. As a result, in the absence of the two signaling peptides Phr8 and Phr63, Rap8 and Rap63 appear to act synergistically.

The results of the concomitant activity of the two Rap-Phr systems raise two main questions about the difference between Rap8 and Rap63 activities and between results obtained under in vitro and in vivo conditions. First, the copy numbers of the plasmids carrying these rap-phr genes are different and should impact their levels of expression in bacterial cells. The copy number of pHT8_1 (75 per chromosome) is certainly much higher than that of the low-copy-number plasmid pAW63 (7, 36). Thus, more Rap8 than Rap63 is likely produced, which could explain the dominant effect of Rap8 on the sporulation of the bacteria. This might be further investigated by Western blot analysis using antibodies targeting Rap8 and Rap63. Moreover, the HD73 strain harbors six other Rap-Phr systems, two of which are predicted to have effects on sporulation (31). The sporulation activities of these two systems could mask the moderate Rap63 activity, or the Rap63 activity could be inhibited by a chromosomal Phr peptide. Second, not all cells in a bacterial population sporulate, as previously reported for B. subtilis (48) and B. thuringiensis (49, 50). This is partly due to the heterogeneous activation pattern of the master sporulation regulator Spo0A (51). In B. thuringiensis populations, it has been shown that sporulation efficiency and the differentiation pathway differ according to environmental conditions (49, 50). In HCT medium, an optimized medium for B. thuringiensis sporulation, the transition phase between the exponential phase and sporulation is reduced from that in LB medium or in insects, the bacterial population is homogeneous, and sporulation is triggered synchronously (52). In sharp contrast, a high degree of heterogeneity was observed in a complex culture medium, such as LB medium, or in insect cadavers (50). Moreover, in insect larvae, the distribution of Phr peptides should be less homogeneous than that in shaken cultures, accentuating the heterogeneity of the bacterial population and preventing synchronous entry into sporulation. Taken together, these reasons could explain the difference in sporulation efficiency observed between in vitro and in vivo experiments.

Rap proteins are inhibited by their cognate Phr oligopeptides (30), and the mature signaling peptide generally corresponds to the C-terminal end of the Phr peptide sequence and contains a positively charged residue (27, 53). In accordance, we showed that the expression of the phr63 gene prevents the negative effect of Rap63 on sporulation (Fig. 2). Furthermore, we demonstrated that the active form of Phr63 is part of the C-terminal end and that its minimal form is a pentapeptide. However, the exact physiological size of the mature Phr63 was not determined. Indeed, synthetic oligopeptides Phr63-5, Phr63-6, Phr63-7, and Phr63-8 successfully inhibited the activity of Rap63 on sporulation. This suggests that Rap63 could be more versatile for binding Phr63 than Rap8, which is effectively inhibited only by the heptapeptide Phr8 (12). In B. subtilis, the mature Phr signaling peptides are penta- or hexapeptides (27, 29), while mature peptides of the RNPP regulators in B. cereus group bacteria are commonly heptapeptides, such as NprX (54), PapR (55), and Phr8 (12).

The C-terminal ends of Phr8 and Phr63 present high sequence similarity, with six of eight amino acids identical, including a histidine residue (positively charged) at position 4 of the octapeptides [QYAHG(E/K)(T/D)I]. Only the residues in positions 6 and 7 are different between Phr8 and Phr63 (Fig. 6A). Previous studies demonstrated that residues located at similar positions in PapR are critical for the interaction with PlcR and for activation of the PlcR regulon (55–58). In contrast, amino acid variations at similar positions in NprX do not affect the affinity of the peptide for NprR or the activation of the NprR regulon (44, 54, 59). Based on the relationship between NprX and the peptides of the Phr family (27), we hypothesized possible cross talk between the Rap63-Phr63 and Rap8-Phr8 plasmid systems. Our results show that Phr8 is not able to inhibit Rap63 and that Phr63 cannot prevent Rap8 activity. Even-Tov and colleagues (32) revealed high orthogonality between Rap and Phr; that is, a given Rap protein is inhibited primarily by its cognate Phr peptide. Indeed, no cross talk was observed among a large number of Rap-Phr systems from the B. subtilis group species, and the absence of cross talk among RapA-PhrA, RapC-PhrC, and RapE-PhrE has been demonstrated (30, 60). As with these quorum-sensing systems, the absence of cross talk between Rap8-Phr8 and Rap63-Phr63 highlights the specificity of the Rap and Phr interactions, supporting the hypothesis of a coevolution of the Rap-Phr system components (17). In the other two B. thuringiensis RNPP systems (PlcR-PapR and NprR-NprX), cross-activation has been reported between different pherotypes of the same quorum-sensing system, but not between different quorum-sensing systems within a given strain (54–56). However, a notable exception has been observed recently with the B. thuringiensis PlcRa regulator, a structural paralog displaying 29% sequence identity with PlcR (61). This RNPP regulator, involved in the oxidative stress response and cysteine metabolism, is activated during stationary phase by PapR, the cognate signaling peptide of PlcR, suggesting a high and unexpected versatility of PlcRa (62).

The RNPP quorum-sensing systems are largely reported to play a role in the tight control of physiological and developmental processes of B. cereus group bacteria in response to environmental changes (20, 49, 50, 52, 63, 64). Taken together, our results reinforce the role of Rap-Phr systems in this fine-tuned regulation. Concomitant activity of different Rap-Phr systems allows the bacteria to better adapt their developmental processes to various environmental conditions.

Horizontal gene transfer can increase the variability of the genetic set of a bacterial species, allowing the bacteria to better adapt to environmental changes. The B. thuringiensis plasmids had been widely studied, mainly due to the cry and cyt genes coding for the insecticidal toxins. However, B. thuringiensis strains harbor a wide range of plasmids, many of which contain Rap-Phr systems predicted to regulate sporulation (31). In the present study, the activities of two plasmid-borne Rap-Phr systems have been tested in the same strain, and we show that Rap8-Phr8 and Rap63-Phr63 work synergistically to control the commitment to sporulation in insect larvae. Since B. thuringiensis is an invertebrate pathogen, it is expected that this species will have a rich repertoire of mechanisms that facilitate its development and survival in insects. Our results reinforce the relevance of plasmid Rap-Phr quorum-sensing systems and highlight the fact that plasmids other than the cry-harboring plasmids play important roles in B. thuringiensis infectious-cycle regulation and bacterial survival.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

All strains used in this study were derived from the B. thuringiensis serovar kurstaki HD73 acrystalliferous (Cry^–; cured of the pHT73 plasmid) strain (36). Escherichia coli K-12 strain TG1 was used as the host strain for plasmid construction. E. coli strain ET12567 (Dam^– Dcm^–) was used to prepare plasmids to transform B. thuringiensis strains by electroporation (65). E. coli strains were transformed by thermal shock and were cultivated in Luria-Bertani (LB) medium at 37°C. B. thuringiensis strains were grown in LB medium or in the sporulation-specific medium HCT (5) at 30°C or 37°C. Liquid cultures were performed with shaking at 175 rpm. For bacterial selection, antibiotics were used at the following concentrations: 100 μg/ml ampicillin and 50 μg/ml spectinomycin for E. coli and 10 μg/ml erythromycin, 200 μg/ml spectinomycin, and 200 μg/ml kanamycin for B. thuringiensis. LB plates with 100 μg/ml of 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (X-Gal) were used for LacZ colorimetric screening during pMAD mutagenesis. When required, xylose was used at a concentration of 20 mM. The strains used in this study are described in Table 1.

TABLE 1.

Strains used in this study

Strain	Description	Source or reference
HD73	B. thuringiensis serovar kurstaki HD73 Cryˉ, cured of the pHT73 plasmid. This strain was designated as the wild-type strain and was used to construct all the other strains described below.	36
HD73(Prap63′-lacZ)	HD73 strain carrying the transcription fusion plasmid pHT304-18_Prap63′-lacZ	This study
HD73 Δspo0A(Prap63′-lacZ)	HD73 spo0A-deficient strain (70) carrying plasmid pHT304-18_Prap63′-lacZ	This study
HD73 ΔsigH(Prap63′-lacZ)	HD73 sigH-deficient strain (71) carrying plasmid pHT304-18_Prap63′-lacZ	This study
HD73(Pphr63′-lacZ)	HD73 strain carrying the transcription fusion plasmid pHT304-18_Pphr63′-lacZ	This study
HD73 Δspo0A(Pphr63′-lacZ)	HD73 spo0A-deficient strain (70) carrying plasmid pHT304-18_Pphr63′-lacZ	This study
HD73 ΔsigH(Pphr63′-lacZ)	HD73 sigH-deficient strain (71) carrying plasmid pHT304-18_Pphr63′-lacZ	This study
HD73 315xyl	HD73 strain harboring the empty plasmid pHT315xyl carrying the xylose-inducible promoter PxylA	12
HD73 xyl-rap63	HD73 strain harboring plasmid pHT315xylΩrap63 and expressing rap63 from PxylA	This study
HD73 xyl-phr63	HD73 strain harboring plasmid pHT315xylΩphr63 and expressing phr63 from PxylA	This study
HD73 xyl-rap63-phr63	HD73 strain harboring plasmid pHT315xylΩrap63-phr63 and expressing rap63 and phr63 from PxylA	This study
HD73 xyl-rap63-phr63_R3	HD73 strain harboring plasmid pHT315xylΩrap63-phr63–R3 and expressing rap63 and truncated phr63 from PxylA	This study
HD73 xyl_PspoIIE′-yfp	HD73 strain carrying the transcription fusion plasmid pHT315xyl_PspoIIE′-yfp	12
HD73 xylΩrap63_PspoIIE′-yfp	HD73 strain carrying the transcription fusion plasmid pHT315xylΩrap63_PspoIIE′-yfp expressing rap63	This study
HD73 xylΩrap63-phr63_PspoIIE′-yfp	HD73 strain carrying the transcription fusion plasmid pHT315xylΩrap63-phr63_PspoIIE′-yfp expressing rap63 and phr63	This study
HD73 Δrap63-phr63	HD73 strain with rap63 and phr63 deleted	This study
HD73 Δphr63	HD73 strain with phr63 deleted, constructed using plasmid pMADΩphr63::spec	This study
HD73 Δrap8-phr8	HD73 strain with rap8 and phr8 deleted	12
HD73 Δphr8	HD73 strain with phr8 deleted	12
HD73 Δphr8 Δphr63	HD73 strain with phr8 and phr63 deleted	This study
HD73 ΔΔ	HD73 strain with the rap8-phr8 and rap63-phr63 genes deleted	This study
HD73 ΔΔ 315xyl	HD73 ΔΔ strain carrying the empty plasmid pHT315xyl	This study
HD73 ΔΔ xyl-rap63	HD73 ΔΔ strain harboring plasmid pHT315xylΩrap63 expressing Rap63 from PxylA	This study
HD73 ΔΔ xyl-rap63-phr63	HD73 ΔΔ strain harboring plasmid pHT315xylΩrap63-phr63 expressing Rap63 and Phr63 from PxylA	This study
HD73 ΔΔ xyl-rap8	HD73 ΔΔ strain harboring plasmid pHT315xylΩrap8 expressing Rap8 from PxylA	This study

Plasmid and strain constructions.

The GenBank accession number of plasmid pAW63 is CP004072.1. The rap63-phr63 locus is located between nucleotides 51638 and 53496. Rap63 corresponds to NCBI Protein accession number AGE81704.1, and Phr63 corresponds to NCBI Protein accession number AGE81705.1.

Plasmid pHT315xyl, a multicopy vector with a xylose-inducible promoter (43), was used to express rap63 and/or phr63. Promoter regions of rap63 or phr63 genes were inserted into plasmid pHT304-18Z (66) in order to determine their expression kinetics. Plasmid pMAD (67) was used for gene disruption by homologous recombination.

The plasmids used in this study are described in Table 2.

DNA manipulations.

Genomic DNA from B. thuringiensis strains was extracted using the Puregene Yeast/Bact. kit (Qiagen, France), and plasmid DNA from E. coli was extracted with the QIAprep Spin Miniprep kit (Qiagen, France). Phusion High-Fidelity DNA polymerase, standard Taq DNA polymerase, restriction enzymes, and T4 DNA ligase were used according to the manufacturer’s recommendations (New England Biolabs, USA). PCRs were performed in an Applied Biosystems 2720 thermal cycler using the oligonucleotides listed in Table 3, synthesized by Eurofins Genomics (Germany). The amplified DNA fragments were purified using a QIAquick PCR purification kit (Qiagen, France), and the QIAquick gel extraction kit (Qiagen, France) was used to purify digested DNA fragments separated on 1% agarose gels. All constructs were verified by DNA sequencing (GATC Biotech, Germany).

TABLE 3.

Primers used in this study

Primer name	Sequence	Restriction sitea
Rap7557-F	CGCGGATCCGAATGAGGGGATTAAATATGAATGTG	BamHI
Rap7557-R	CCCAAGCTTTCATTATTTTAAAGCTCCTTTCTCGG	HindIII
Phr7557-F	CGCGGATCCTGATAAAAAGGCTTCCGAGAAAG	BamHI
Phr7557-R	CCCAAGCTTGGTGTTAAATAGTTTCACCATGTGC	HindIII
7557Amont1-F	CATGCCATGGCGCCTTTATTGTCAAGATACATCTACTC	NcoI
7557Amont1-R	CGGGGTACCACATTCATATTTAATCCCCTCATTC	KpnI
7557Amont2-F	CATGCCATGGTATCAATCCATCATTTCACAACATG	NcoI
7557Amont2-R	CGGGGTACCATTATTTTAAAGCTCCTTTCTCGG	KpnI
7557Aval-F	CGTCTAGACACCATAAAGTACTAAAAAGTTATGTCATTAC	XbaI
7557Aval-R	CCGGAATTCCAATTTTGACCAAAGTCAATCCAC	EcoRI
Prom7557-F (Prap F)	CCCAAGCTTCGTTACTTATAAGAAACAAACAAGAGCC	HindIII
Prom7557-R (Prap R)	CGCGGATCCACATTCATATTTAATCCCCTCATTC	BamHI
Prom7557Phr-F (Pphr F)	CCCAAGCTTGCTGCTTGTAATAACACACTAGG	HindIII
Prom7557Phr-R (Pphr R)	CGCGGATCCATTATTTTAAAGCTCCTTTCTCGG	BamHI
Phr7557R3	CCCAAGCTTAATATTGAACACAGTCTACTTTTTCTTTTG	HindIII
RT7557-2 (RT-2)	GAAGGCATCTGCTTGATCAGGTATAC
RT7557-3 (RT-3)	GCTTGTAATAACACACTAGGTCTTGC
RT7557-4 (RT-4)	CCATGTGCATATTGAACACAGTCTAC
RT7557-5 (RT-5)	GTAGACTGTGTTCAATATGCACATGG
RT7557-7 (RT-7)	CTTCAAGACATAGAAGACCAACATGTG
PU-EcoRI	CGGAATTCGCCAGGGTTTTCCCAGTCACGAC	EcoRI
YFP-R	CGGAATTCTTATTTGTATAGTTCATCCATGC	EcoRI
PspoIIE-F	AACTGCAGCTGGCTAGAGCGTACGG
xylRout3′	GGAATGTCCTCCATTGTGATTGATC

^aRestriction sites are shown in boldface in primer sequences.

RT-PCR experiments.

Growth conditions, RNA extraction, and reverse transcriptase reactions were performed as described elsewhere (12). Three different fragments were amplified by PCRs with oligonucleotides PromRapF7557 and RT7557-2 for the rap63 gene and its upstream region, with RT7557-3 and RT7557-4 for the rap63 and phr63 genes, and with RT7557-5 and RT7557-7 for the phr63 gene and its downstream region. The sequences of the oligonucleotides are given in Table 3.

β-Galactosidase assays.

Expression from the P_rap63 and P_phr63 promoter regions was analyzed by measuring β-galactosidase activity. Strains containing plasmids pHT304-18_P_rap63′-lacZ and pHT304-18_P_phr63′-lacZ were grown in HCT medium at 37°C with shaking at 175 rpm. The assays were performed as described previously (54). Specific activities are expressed in β-galactosidase units per milligram of protein. The assays were independently repeated three times.

Fluorescence analysis.

YFP fluorescence produced from the strain carrying plasmid pHT315xyl_P_spoIIE′-yfp was measured from bacterial cultures grown in HCT medium at 37°C. Cells were harvested at determined time points and were fixed as described elsewhere (68). Cultures were supplemented with 20 mM xylose at the onset of the stationary-growth phase (t0). Fixed cells were kept at 4°C until analysis. Samples were distributed into a 96-well black polystyrene microplate (Greiner) and were measured with an Infinite 200 Pro microplate reader device (Tecan, Switzerland), applying an excitation wavelength of 485 nm and an emission wavelength of 535 nm. Data were recovered by Tecan i-control software (Tecan, Switzerland), and results at each time point were expressed in arbitrary units per OD₆₀₀ (optical density at 600 nm) unit. Promoter analyses were carried out in triplicate, and mean values were calculated.

Synthetic oligopeptides.

Phr peptides, corresponding to the C-terminal end of the phr63 gene product, were synthesized, purified, and identified by mass spectrophotometry by GenScript (USA). To determine the active oligopeptide, the synthetic peptides were tested in sporulation assays (see below) at 50 μM final concentrations.

In vitro sporulation assays.

In vitro sporulation efficiency tests were carried out in the sporulation-specific medium HCT. B. thuringiensis strains were grown at 30°C for 48 h, and serial dilutions were plated before and after heat treatment (12 min at 80°C). When required, xylose (20 mM) and synthetic peptides were added to the culture at the beginning of the stationary-growth phase (t0). The sporulation percentage was calculated as 100 multiplied by the ratio between the number of heat-resistant spores per milliliter and the number of total viable cells per milliliter. Experiments were done at least in triplicate, and mean values were calculated. Results were analyzed statistically by analysis of variance (ANOVA), followed by Tukey’s test (P < 0.05).

In vivo sporulation assays.

Experiments in insect larvae were performed as described elsewhere (12, 69). Briefly, larvae (last instar) of the lepidopteran insect G. mellonella were infected by intrahemocoelic injection of 2 × 10⁴ vegetative bacteria and were kept at 30°C. Dead larvae were crushed 96 h after infection, serially diluted in 0.9% NaCl solution, and plated onto LB agar before and after heat treatment for 12 min at 80°C. The sporulation efficiencies of the harvested B. thuringiensis cells were calculated as described for the in vitro assays. At least three independent replicates were performed for each strain, and results were statistically analyzed by an unpaired t test with Welch’s correction (P < 0.05).