DegQ and SwrA differentially modulate protease production and flagellar motility in Bacillus subtilis

Abstract

In Bacillus subtilis, the DegS-DegU two-component system modulates a range of adaptive responses, including extracellular protease production and the flagellar-surface motility called swarming. Two proteins, DegQ and SwrA, are thought to increase the amount of phosphorylated DegU and alter DegU DNA-binding abilities of DegU, respectively. Here we compare the relative contributions of DegQ and SwrA to DegU-related phenotypes in the B. subtilis ancestral strain that is functional for both proteins. We found that DegQ enhances the production of secreted proteases in stationary phase and inhibits swarming motility during exponential growth by a mechanism dependent on both DegS and DegU. Moreover, DegQ-mediated swarming inhibition was due, in part, to the attenuation of transcription levels within the long fla/che operon. SwrA by contrast, activates the promoter of the fla/che operon by a mechanism that depends on DegU but not DegS. Moreover, SwrA inhibits extracellular protease production by inhibiting the expression of DegQ. We conclude that DegQ and SwrA have opposing effects on DegU-dependent phenotypes and do so by different mechanisms. Thus, B. subtilis evolved third-party modulators to expand the conventional two-component system and achieve differential gene expression.

IMPORTANCE

The response regulator DegU regulates the expression of a wide array of adaptive responses including extracellular protease production and swarming motility in B. subtilis. Here we show that DegQ and SwrA modulate DegU activity to inversely regulate expression of the AprE protease and the fla/che operon for flagellar biosynthesis. This work highlights the roles of accessory factors in two-component system regulation, and clarifies the relationship between DegQ and SwrA, two proteins that were disabled during the domestication of laboratory strains.

INTRODUCTION

B. subtilis secretes at least eight extracellular proteases, which while dispensable for growth in the laboratory, are likely necessary for nutrient acquisition in the environment (1, 2). Of the eight proteases, serine protease subtilisin E or AprE seems to be predominant and contributes to over 70% of the extracellular proteolytic activity in culture (3). Moreover, AprE has industrial applications with roles in leather and textile processing, food production, waste management, and pharmaceutical synthesis (4). Due to the biological consequences and extensive commercial applications, expression of aprE, the gene that encodes AprE, has been heavily studied and optimized (5–7). The primary regulator of aprE gene expression is thought to be the DegS-DegU two-component system (8).

DegU is a response regulator DNA-binding transcription factor that is phosphorylated by the soluble DegS kinase (9–12). Together, the DegS-DegU system is highly pleiotropic and regulates sporulation, biofilm formation (13, 14), genetic competence (15, 16), motility (17–19), poly-glutamate production (20, 21) and extracellular protease production (22–24) but discriminating direct from indirect effects is difficult due to the lack of a DegU DNA binding site consensus sequence (25, 26), and the input for the DegS kinase is unknown. Complicating matters further, DegU has been reported to activate and repress transcriptional activity in both its phosphorylated (DegU~P) and non-phosphorylated states, and DegS-independent DegU activity and phosphorylation have been reported. Finally, DegU dependent gene expression is modulated by two small proteins, SwrA and DegQ, and both are defective in commonly studied laboratory strains of B. subtilis (13, 20, 27, 28).

The swrA gene that encodes SwrA contains a loss of function frameshift mutation in laboratory strains, and the first indication of SwrA function was the isolation of a mutant allele called ifm, that restored the reading frame to functionality and enhanced motility (27–29). Subsequent research indicated that SwrA increased the frequency of motile cells in the population (30) and increased the number of flagella per cell on motile individuals (30–32). ChIP-seq analysis indicated that SwrA enriched DNA in a DegU-dependent manner and did so at a subset of DegU regulated promoters (33). Moreover, SwrA induced DegU oligomerization at an activation sequence upstream of the fla/che operon dedicated to flagellar biosynthesis, allowing cells to upregulate transcription and reach the hyperflagellated state required for swarming (19, 33, 34). Despite SwrA enrichment at multiple DegU regulated promoters, what governs specificity is unknown, and SwrA does not currently have a clear role outside the regulation of motility.

The degQ gene encoding DegQ contains a mutation in the promoter that all but abolishes expression in laboratory strains, and degQ was originally discovered in a screen in which a reversion of this mutation enhanced extracellular protease production (20, 22, 35). Moreover, DegQ was shown to activate the promoter of aprE (P_aprE) and does so in a DegU-dependent manner (6, 13, 36). DegQ is thought to modulate DegU phosphorylation, and either promotes the phospho-transfer from DegS, inhibits DegS phosphatase activity, or stabilizes the phosphorylated form of DegU~P (13, 37). The precise mechanism by which DegQ stimulates DegU-dependent expression is unknown, and its role in regulating DegU-related phenotypes outside of extracellular protease production is poorly understood.

Here we explore the effect of DegQ and SwrA and their relative contribution to DegU-mediated gene regulation in the ancestral 3610 strain of B. subtilis. Whereas cells with high levels of DegQ increase production of extracellular protease activity but inhibit swarming motility. Conversely, SwrA is required for swarming but inhibits extracellular proteases. The phenotypic consequences of both proteins are reflected in their opposing effects on DegU-dependent expression of the P_aprE and P_fla/che. In the course of the study, we find that extracellular proteases are not expressed until the transition to stationary phase and that while the DegQ effect requires the DegS kinase, the SwrA effect does not. Moreover, we show that SwrA appears to act as an inhibitor of aprE gene expression by inhibiting the expression of DegQ. Thus, the two small protein modulators appear to act in opposition at the transition between exponential growth and stationary phase. How such different regulatory outcomes are achieved through the same DNA binding transcription factor remains unknown.

RESULTS

The DegS-DegU two-component system activates expression of aprE.

To determine whether DegU was required for extracellular protease activity in the ancestral strain of B. subtilis, supernatants were collected over a time course and protease activity was measured. At low OD₆₀₀ during exponential growth (Fig 1A), little protease activity was detected but increased at high optical density during the transition to stationary phase (Fig 1B). Moreover, disruption of DegU had no effect on growth (Fig 1A) but reduced extracellular protease activity relative to wild type at all optical densities tested (Fig 1C). We conclude that extracellular protease activity is a stationary phase phenomenon under the conditions tested and is enhanced by DegU in the ancestral strain of B. subtilis.

Figure 1. Extracellular protease production is activated by DegQ in stationary phase.

A) Spectrophotometric growth curve of wild type and various strains measured at OD₆₀₀ and grown in LB at 37 °C. Horizontal colored bars indicate regions of cell density where orange is less than 1 OD₆₀₀, yellow is between 1 and 3 OD₆₀₀, and green is above 3 OD₆₀₀. The color coding is maintained throughout the remainder of the figure. The following strains were used to generate this panel: wild type (gray, DB3692), degU (blue, DB1815), and cells overexpressing degQ⁺⁺ (pink, DB3695). B-D) Double Y bar graphs where white bars measure extracellular protease activity and blue bars measure the expression of a P_aprE-lacZ transcriptional reporter. Each datapoint is the average of three replicates where the bar is the average, the whiskers represent the standard deviation, and the individual data points are shown as dots. The following strains were used to generate these panels and are the same as panel A: B) wild type (DB3692), C) degU (DB1815), and D) degQ⁺⁺ (DB3695). To generate this figure, cells were grown to an OD600 of approximately 2, samples were collected, extracellular protease activity and β-galactosidase expression were measured in parallel.

The alkaline protease AprE is encoded by the aprE gene, the expression of which is considered to be a primary indicator for proteolysis (3, 5), and disruption of AprE reduced the total amount of extracellular proteolytic activity (Fig 2). To measure aprE expression, a reporter construct in which the lacZ gene encoding β-galactosidase was fused to the P_aprE promoter, inserted at an ectopic site, and measured at various optical densities. Consistent with the protease activity profile, expression of P_aprE was low at low OD and increased during the transition to stationary phase (Fig 1B). Also consistent with previous observations, deletion of degU (Fig S1A) reduced expression from P_aprE at all densities tested (Fig 1C). Based on these results, we chose an optical density of 2 for our standard conditions when measuring P_aprE expression, as this was a condition with substantial activity that was reduced by mutation of degU. Moreover, the degU mutant phenotype could be complemented when degU was from its native promoters (10, 38), and inserted at an ectopic site in the chromosome (Fig S1A, B). Finally, deletion of degU resulted in a significant decrease in extracellular protease activity under the standard conditions (Fig 2).

Figure 2. Extracellular protease activity is enhanced by DegQ and inhibited by SwrA.

Extracellular protease activity as measured by the degradation of fluorescent azocasein in supernatants of cells grown to OD₆₀₀ of ~2 in LB (+1mM IPTG when appropriate) from strains of the indicated genetic backgrounds. Each bar is the average activity taken from at least three independent measurements, error bars represent the standard deviation. The dashed line indicates the amount of activity taken from wild type. The following strains were used to generate this panel: wild type (DK1042) aprE (DS5648), degU (DS3649), degS (DK4236), degQ (DB3074), degQ⁺⁺ (DK6249), degU degQ⁺⁺ (DB3745), degS degQ⁺⁺ (DB2863), swrA degQ⁺⁺ (DB3185), swrA (DS2415), swrA⁺⁺ (DB3835), swrA⁺⁺ degQ⁺⁺ (DB3729).

The DegU response regulator is phosphorylated, at least in part, by the cognate kinase DegS (11, 12, 39). To test whether the DegS kinase was required for the expression of aprE, the reporter construct was inserted into a strain deleted for the degS gene in a manner that preserved previously reported internal promoters that drive degU gene expression (38) (Fig S1A). The degS deletion significantly decreased expression of the aprE promoter and decreased extracellular protease activity (Fig 2, Fig 3A). Moreover, the mutation could be partially complemented by introduction of the degS gene expressed from its native promoter and inserted at an ectopic site in the chromosome (Fig S1A, B). We conclude that expression of aprE is dependent on both DegS and DegU and we infer that the phosphorylation of DegU is required for aprE expression.

Figure 3. DegQ activates P_aprE and SwrA inhibits DegQ expression.

β-galactosidase activities measured in Miller units (MU) for each of the indicated transcriptional reporter fusions and mutants grown to an OD of 2 in LB media. A) Expression levels from the P_aprE reporter. The following strains were used to generate this panel: wild type (DB3692), degU (DB1815), degS (DB3741), degQ (DB3751) degU degQ (DB3767), degS degQ (DB2877), degQ⁺⁺ (DB3695), degU degQ⁺⁺ (DB3769), degS degQ⁺⁺ (DB3693), swrA degQ⁺⁺ (DB3748), swrA (DB3698), degU swrA (3750), degS swrA (DB3765), degQ swrA (DB3746), swrA⁺⁺ (DB3865). B) Expression levels from the P_fla/che reporter. The following strains were used to generate this panel: wild type (DK5183), degU (DB3697), degS (DB3742), degQ (DB3752) degU degQ (DB3768), degS degQ (DB3131), degQ⁺⁺ (DB3696), degU degQ⁺⁺ (DB3770), degS degQ⁺⁺ (DB3764), swrA degQ⁺⁺ (DB3749), swrA (DK6345), degU swrA (DB3753), degS swrA (DB3766), degQ swrA (DB3747), swrA⁺⁺ (DB3701). Each bar is the average of three biological replicates, and the error bars represent the standard deviation.

To determine whether the DegS-DegU two-component system activates transcription of the other extracellular proteases, RNA was purified from the wild type and cells deleted for either degS or degU, grown to an optical density of 2, sequenced, and compared in transcriptomic analysis. Consistent with β-galactosidase reporter assays, deletion of either degS or degU resulted in a decrease in the amount of transcript from the aprE gene and also reduced expression from gene bpr encoding the Bpr protease (Fig 4A) (26, 34). Transcripts from the 6 remaining genes that encode extracellular proteases, epr, mpr, nprB, nprE, vpr, and wprA were unchanged from wild type (Fig 4A) (25, 26). We conclude that DegS and DegU increase transcript abundance of aprE and bpr. While it does not appear that DegS and DegU regulate the remaining 6 extracellular protease encoding genes, we note that additional regulation may apply at those loci and potentially obscure the effect under our standard conditions.

Figure 4. DegQ activates a subset of extracellular protease transcripts and inhibits fla/che operon transcript downstream of the promoter.

Transcript abundance expressed as a log₂ fold change relative to wild type for the indicated genes in the indicated strains: degU deletion mutant (blue, DS3649), degS deletion mutant (orange, DK4236), and a strain overexpressing degQ from a P_hyspank promoter induced with 1 mM IPTG (degQ⁺⁺, magenta, DK6294). Cells were collected after growth in LB to an OD₆₀₀ of 2. Genes and operons not drawn to scale. A) Change in transcript abundance for the 8 genes known to encode extracellular proteases. B) Change in transcript abundance for each of the genes in the 32 gene fla/che operon. C) Change in transcript abundance for genes under SigD control.

DegQ is an activator of extracellular proteases and an inhibitor of swarming motility.

DegQ is thought to control the phosphorylation state of DegU by modulating either the kinase and/or phosphatase activity of DegS (13, 37). Consistent with a positive role on DegU-dependent phenotypes, mutation of the degQ gene reduced extracellular protease activity (Fig 2), and reduced transcript levels of both the aprE and bpr genes in transcriptomic analysis (Fig 4A). We note, however, that the mutation of degQ did not result in reduced expression of the P_aprE reporter (Fig 3A). We conclude that while DegQ appears necessary for activating DegU-dependent expression of extracellular protease genes, it does not appear to do so at the level of promoter initiation. Rather, we infer that DegQ may act downstream of the P_aprE promoter to increase the level of the aprE gene transcript.

To determine whether DegQ was sufficient to activate aprE expression, the degQ gene was placed under an IPTG-inducible promoter, and cells were grown in the presence of inducer (degQ⁺⁺). Induction of DegQ did not cause premature expression from the P_aprE reporter or extracellular protease activity, but upon entry into stationary phase, both measurements transiently increased by approximately 10-fold over wild type (Fig 1D, Fig 2). Moreover, elevated P_aprE reporter expression was dramatically reduced when DegQ was induced in either a degU or degS mutant background (Fig 3A). Finally, RNAseq analysis of cells overexpressing DegQ indicated an increase in the transcript level of aprE, bpr, nprB, nprE, and vpr (Fig 4A). We conclude that DegQ is not essential for the expression of aprE but the expression of aprE is elevated in a DegS-DegU dependent manner when DegQ is overproduced.

To determine whether DegQ was a generalized enhancer for the DegS-DegU two component system, the expression of P_fla/che, another promoter under DegU control, was measured under the same early-stationary phase conditions that was used for measuring protease gene expression. Consistent with previous reports, deletion of degU reduced expression from P_fla/che (18, 19, 33), but here we find that deletion of either degS or degQ did not (Fig 3B). Moreover, artificial overexpression of DegQ did not reduce P_fla/che expression in wild type, and deletion of degU was epistatic to degQ overexpression (Fig 3B). We conclude that DegQ is not a generalized enhancer for DegU-dependent gene expression and that transcriptional activation by DegQ is promoter-dependent. We further conclude that unlike that observed with the P_aprE promoter, DegS was not required for the DegU-dependent activation of the P_fla/che promoter.

The level of P_fla/che promoter expression is a critical determinant of swarming motility (30, 32). Consistent with the reduced reporter activity and previous reports, deletion of degU (Fig 5A) abolished swarming (13, 14), but mutation of degS did not (Fig 5B, Fig S1B). Despite having little effect on the P_fla/che reporter gene expression, deletion of degQ resulted in a reduced swarm lag period (Fig 5C), and overexpression of DegQ severely impaired swarming (Fig 5D). Simultaneous mutation of both degS and degQ restored a wild type swarming lag period suggesting that the degS mutation was epistatic (Fig 5E). Moreover, mutation of degS also abolished the inhibition of swarming due to DegQ overexpression, further supporting degS epistasis (Fig 5F). We conclude that DegQ inhibits swarming motility in a DegS-dependent manner. We infer that whatever the mechanism by which DegQ inhibits swarming, its repressive effect likely operates downstream of the P_fla/che promoter.

Figure 5. DegQ inhibits swarming motility.

Swarm expansion assays for each indicated strain grown on 0.7% LB at 37 °C. In each graph, swarming of the wild type (DK1042) is represented by open symbols and swarming of the indicated mutant strain is in closed symbols. The following strains were used to generate this panel: A) degU (DS3649) B) degS (DK4236) C) degQ (DB3074) D) degU degQ (DB3744) E) degS degQ (DB2851) F) degQ⁺⁺ (DK6294) G) degU degQ⁺⁺ (DB3745) H) and degS degQ⁺⁺ (DB2863) I) swrA (DS2415) J) swrA degQ (DB3726) K) swrA degQ++ (DB3185) L) swrA⁺⁺ (DB3835) M) swrA⁺⁺ degQ⁺⁺ (DB3729) N) slrR (DB3899) O) slrR degQ⁺⁺ (DB3910).

To determine where DegQ inhibited motility within the flagellar regulon, RNAseq transcriptomics were performed on a strain that overexpressed DegQ. Transcript levels of genes early in the fla/che operon were initially higher than wild type when DegQ was overexpressed, but the transcript levels gradually decreased in a distance dependent manner until the transcript levels towards the back of the operon resembled cells lacking DegU. We noted that transcript decline appeared to occur in discrete steps, reminiscent of roadblock attenuation within the fla/che operon recently reported to be mediated by a heteromeric complex of DNA binding proteins SinR and SlrR (40). Disruption of SlrR in a strain is capable of swarming motility (Fig 5G) that overexpressed DegQ improved swarming motility but did not restore behavior to that of the wild type (Fig 5H). We conclude that DegQ inhibits the fla/che operon transcript levels downstream of initiation at the P_fla/che promoter and attenuation within the operon by the SinR•SlrR heteromer contributes to the swarming deficiency.

DegQ and SwrA are antagonistic on DegU activity.

DegQ appeared to have opposing effects on various DegU-activated phenotypes. Specifically, it enhanced extracellular protease activity but inhibited swarming over surfaces. We wondered whether the differential effects were due to the presence of another DegU modulator protein, SwrA (33, 34, 41). Indeed, mutation of swrA resulted in a dramatic increase in the expression of P_aprE and extracellular proteolytic activity and overexpression of swrA, reduced P_aprE expression and proteolysis to levels resembling the degU mutant (Fig 2A). Moreover, the repressive effects of SwrA on both P_aprE expression and extracellular protease activity were DegQ dependent, as elevated aprE expression was restored to wild type when both swrA degQ were simultaneously deleted (Fig 2A, Fig 3). Whereas SwrA appears to repress aprE expression indirectly through DegQ. SwrA has been shown to directly activate the P_flache promoter directly, and consistent with direct activation, mutation of swrA was epistatic to mutation of degQ for P_fla/che reporter activity (Fig 2B). We conclude that SwrA activates some members of the DegU regulon and inhibits others, and at least in the case of aprE, SwrA inhibition was DegQ-dependent.

One way in which SwrA might inhibit P_aprE expression in a DegQ-dependent manner is if SwrA inhibits the expression of DegQ itself. To determine whether the degQ gene is regulated by swrA, a transcriptional reporter was generated in which the promoter region of degQ (P_degQ) was fused to the lacZ gene and inserted at an ectopic site in the chromosome. Mutation of swrA resulted in a dramatic increase in the expression of degQ relative to wild type, consistent with SwrA functioning as a degQ repressor (Fig 6). Seemingly inconsistent with SwrA acting as a degQ repressor however, overexpression of SwrA did not decrease degQ expression, suggesting that the absence of SwrA and its overexpression, had two different effects depending on the promoter (Fig 6). We conclude that SwrA inhibits the expression of AprE indirectly, at least in part, due to the inhibition of the DegQ activator. We infer, due to the inconsistent results between the swrA mutant and swrA overexpression state, that SwrA likely inhibits degQ gene expression indirectly.

Figure 6. SwrA inhibits DegQ and DegQ autoactivates P_degQ expression.

β-galactosidase activities measured in Miller units (MU) for each of the indicated transcriptional reporter fusions and mutants grown to an OD of 2 in LB media. Expression levels from the P_degQ and P_swrA reporters. The following strains containing the reporter were used to generate this panel (P_degQ/P_swrA): wild type (DB1051/DK6613), degU (DB2508/DK6626), degS (DB3827/DB3826), degQ (DB2690/DB3776), swrA (DB3774/DB3780), swrA degQ (DB3959/DB3690), degQ⁺⁺ (DB3777/DB3778), swrA degQ⁺⁺ (DB3961/DB362), swrA⁺⁺ (DB3779/DK6624). Each bar is the average of three biological replicates, and the error bars represent the standard deviation.

Transcription of the degQ gene has been shown to be activated by DegS and DegU (42), and mutation of either degS or degU reduced expression from the P_degQ promoter. Moreover, mutation of degU was epistatic to mutation of swrA for P_degQ reporter expression (Fig 6). We conclude that SwrA inhibits degQ indirectly through DegU. and was epistatic to mutation of swrA, indicating that inhibition by SwrA was DegU-dependent. As DegQ is also a modulator of DegU-dependent gene expression we wondered whether DegQ was autoregulatory. Indeed, mutation and overexpression of degQ, decreased and increased degQ gene expression, respectively. While SwrA caused DegU-dependent repression of the P_degQ promoter, the reverse was not true, that is, mutation and overexpression seemed to have little effect on the DegU-dependent expression of the P_swrA promoter. We conclude that two different modulators of DegU differential effects on the DegU regulon, and that SwrA antagonizes the production of DegQ.

DISCUSSION

DegU is a highly-pleiotropic and poorly-understood regulator in Bacillus subtilis expressed with, and phosphorylated by, the cognate soluble histidine kinase DegS, but the input to DegS activity and the output DegU DNA binding site is largely unknown. Perhaps one reason for the difficulty in resolving the function of DegU is the extensive use of domesticated laboratory strains where null mutants have little phenotypic effect and data is complicated by the reliance on hyperactive DegU mutant alleles (e.g. DegUhy32^H12L) (9, 24, 43). Phenotypes have been reported for degU null mutants in ancestral strains however, likely because ancestral strains are functional for two small proteins, DegQ and SwrA, that enhance DegU activity (13, 14, 41, 43). Here we study the relative roles of DegQ and SwrA in the ancestral strain background of B. subtilis and find that the two proteins act in opposition to inversely activate extracellular protease production and flagellar motility, respectively.

DegQ is a 46 amino acid protein, the expression of which was abolished in domesticated strains by mutations in the promoter (35, 44, 45). Consistent with early genetics, we find that DegQ increases both extracellular protease activity and the expression of aprE (35, 46). How DegQ activates gene expression is unknown, save that it requires both DegS and DegU, and enhances DegS-dependent DegU phosphorylation in vitro (13, 42, 46). AlphaFold3 modeling suggests that DegQ may interact with the N-terminal input domain of DegS (37), and DegQ likely either potentiates DegS signaling by altering its conformation, or perhaps DegQ is the signal itself. Whatever the case, overexpression of DegQ is insufficient to activate aprE gene expression until the transition to stationary phase and this may explain the absence of DegU binding to the aprE promoter region in ChIP-seq analysis of cells during exponential growth (33). We infer that extracellular protease expression is somehow restricted, and aprE is known to be inhibited until the onset of stationary phase by the multiple repressors including ScoC and CodY (47–50).

DegQ also functions as an inhibitor of swarming motility as here we show that mutation of degQ decreases the swarming lag period while the overexpression of degQ impairs swarm expansion. Swarming has been shown to depend on a critical density of flagella on the cell surface (32) and flagellar number has been shown to be proportional to the level of expression from the P_flache promoter (30). DegQ does not seem to inhibit P_fla/che promoter activity however and instead causes a gradual decrease in transcript abundance from the front of the fla/che operon to the back. The fla/che operon was recently reported to experience transcript attenuation due to internal binding of the SinR•SlrR a heteromer that seems to cause premature termination (40), and mutation of SlrR improved motility when DegQ was overexpressed. Thus, it seems that DegQ inhibits motility, at least in part, by enhancing the amount and/or activity of the SinR•SlrR heteromeric repressor and we suspect that the DegUhy32 allele likely inhibits motility by the same mechanism (17, 40). How DegQ activates heteromer activity is unknown.

SwrA is a 117 amino acid protein, the expression of which was abolished in domesticated strains by an early loss-of-function frameshift mutation in the coding region (27, 28, 44). SwrA increases the frequency of motile cells in the population, increases the number of flagella per cell, and promotes swarming motility (27, 30). Previous work has shown that SwrA interacts with the DegU N-terminal domain and causes DegU oligomerization to activate expression of the P_fla/che promoter (33, 34). Here we find that SwrA inhibits expression of the P_aprE promoter indirectly by repressing expression of the degQ gene. How SwrA acts as a repressor is unknown but, in its absence, DegQ autoactivates its DegU-dependent expression, likely driving an increase in DegQ abundance. As the frameshift mutation in swrA is one of the earliest events in B. subtilis domestication, we wonder whether the promoter down-mutation in degQ was selected as a consequence of the absence of SwrA that would otherwise prevent DegQ positive feedback (44, 51). Thus, SwrA promotes flagellar gene expression during exponential growth, and we infer that during the transition to stationary phase, SwrA is inactivated thereby derepressing DegQ and extracellular protease gene expression.

In sum, we find that two small proteins SwrA and DegQ fundamentally alter the activity of the DegU response regulator. SwrA promotes activation of the P_fla/che promoter by mechanism that depends on DegU but not DegS. By contrast, DegQ activates the expression of aprE in a manner that depends on both components of the DegS/DegU two-component system. Moreover, while SwrA directly inhibits DegQ production, DegQ overrides SwrA function downstream of its target site and attenuates the production of new flagella during stationary phase. From the combined action of the two proteins, we infer that de novo flagellar synthesis and the secretion of extracellular proteases do not occur simultaneously, at least at the level of an individual cell. It is unclear why B. subtilis would evolve multiple modulators of a single two-component system rather than evolve an alternative regulator, but the fact that it did may explain why the degU regulon seems both vast and mysterious. We wonder whether the various modulators not only govern the timing of gene expression within the regulon, but might also fundamentally alter the DegU binding site, thereby obscuring consensus.

MATERIALS AND METHODS

Growth conditions.

B. subtilis stains were grown in Luria-Bertani (LB) broth (10 g tryptone, 5 g yeast extract, 5 g NaCl per L), or LB plates fortified with 1.5% Bacto agar at 37 °C. When appropriate, antibiotics were included in the following concentrations: 5 μg/mL chloramphenicol (cat), 5 μg/mL kanamycin (kan), 1 μg/mL erythromycin plus 25 μg/mL lincomycin (mls), 100 μg/mL spectinomycin (spec), 10 μg/mL tetracycline (tet). Isopropyl β-D-thiogalactopyranoside (IPTG, Sigma) was added to the medium at 1 mM when appropriate.

SPP1 phage transduction.

Serial dilutions of SPP1 phage were added to 200 μL of dense culture grown in TY broth (LB broth supplemented after autoclaving with 10 mM MgSO₄ and 100 μM MnSO₄). To each mixture, 3 mL of molten TY top agar (TY broth supplemented with 0.5% agar) was added. The mixture poured atop fresh TY plates (TY broth fortified with 1.5% agar), and the plates were incubated at 30 °C overnight. Top agar from a plate containing near-confluent plaques was harvested by scraping into a 15 mL conical tube with 5 mL of TY broth and centrifuged at 6,000 × g for 5 min. The supernatant was passed through a 0.22 μm syringe filter and stored at 4 °C. Recipient cells for transduction were grown to stationary phase in 3 mL TY broth at 37 °C. One milliliter of dense culture was mixed with 25 μL of SPP1 donor phage stock made above. Nine milliliters of TY broth were added and mixed by rocking at room temperature for 30 min. The transduction mixture was centrifuged at 6,000 × g for 5 min, the supernatant was discarded, and the pellet was suspended in the remaining volume. 100 μL of cell suspension was then plated on LB plates fortified with the appropriate antibiotic for selection and 10 mM sodium citrate if the antibiotic was cat, kan, or spec.

Strain construction.

All constructs were either first introduced into the domesticated B. subtilis strain PY79 by natural competence and then transferred to the undomesticated B. subtilis 3610 background using SPP1-mediated generalized phage transduction (52) or transformed directly into DK1042 (53). All plasmids used in this study are listed in Table S1. All primers used in this study are listed in Table S2. All strains used in this study are listed in Table 1.

Table 1:

Strains

Strain	Genotype
DB3692	comIQ12L amyE::PaprE-lacZ cat
DB1051	comIQ12L amyE::PdegQ-lacZ cat
DB1815	ΔdegU amyE::PaprE-lacZ cat
DB2508	ΔdegU amyE::PdegQ-lacZ cat
DB2690	comIQ12L degQ::tet amyE::PdegQ-lacZ cat
DB2851	comIQ12L ΔdegS degQ::tet
DB2863	comIQ12L ΔdegS thrC::Physpank-degQ mls
DB2877	comIQ12L ΔdegS degQ::tet amyE::PaprE-lacZ cat
DB3074	comIQ12L degQ::tet
DB3131	comIQ12L ΔdegS degQ::tet amyE::Pfla/che-lacZ cat
DB3185	ΔswrA thrC::Physpank-degQ mls
DB3693	comIQ12L ΔdegS thrC::Physpank-degQ mls amyE::PaprE-lacZ cat
DB3695	comIQ12L thrC::Physpank-degQ mls amyE::PaprE-lacZ cat
DB3695	comIQ12L thrC::Physpank-degQ mls amyE::PaprE-lacZ cat
DB3696	comIQ12L thrC::Physpank-degQ mls amyE::Pfla/che-lacZ cat
DB3697	ΔdegU amyE::Pfla/che-lacZ cat
DB3698	ΔswrA amyE::PaprE-lacZ cat
DB3701	thrC::Physpank-swrA mls amyE::Pfla/che-lacZ cat
DB3726	ΔswrA degQ::tet
DB3729	amyE::Physpank-swrA spec thrC::Physpank-degQ mls
DB3741	comIQ12L ΔdegS amyE::PaprE-lacZ cat
DB3742	comIQ12L ΔdegS amyE::Pfla/che-lacZ cat
DB3744	ΔdegU degQ::tet
DB3745	ΔdegU thrC::Physpank-degQ mls
DB3746	ΔswrA degQ::tet amyE::PaprE-lacZ cat
DB3747	ΔswrA degQ::tet amyE::Pfla/che-lacZ cat
DB3748	ΔswrA thrC::Physpank-degQ mls amyE::PaprE-lacZ cat
DB3749	ΔswrA thrC::Physpank-degQ mls amyE::Pfla/che-lacZ cat
DB3750	ΔdegU swrA::kan amyE::PaprE-lacZ cat
DB3751	comIQ12L degQ::tet amyE::PaprE-lacZ cat
DB3752	comIQ12L degQ::tet amyE::Pfla/che-lacZ cat
DB3753	ΔdegU swrA::kan amyE::Pfla/che-lacZ cat
DB3764	comIQ12L ΔdegS thrC::Physpank-degQ mls amyE::Pfla/che-lacZ cat
DB3765	comIQ12L ΔdegS swrA::spec amyE::PaprE-lacZ cat
DB3766	comIQ12L ΔdegS swrA::spec amyE::Pfla/che-lacZ cat
DB3767	ΔdegU degQ::tet amyE::PaprE-lacZ cat
DB3768	ΔdegU degQ::tet amyE::Pfla/che-lacZ cat
DB3769	ΔdegU thrC::Physpank-degQ mls amyE::PaprE-lacZ cat
DB3770	ΔdegU thrC::Physpank-degQ mls amyE::Pfla/che-lacZ cat
DB3774	ΔswrA amyE::PdegQ -lacZ cat
DB3776	comIQ12L degQ::tet amyE::PswrA-lacZ cat
DB3777	thrC::Physpank-degQ mls amyE::PdegQ-lacZ cat
DB3778	thrC::Physpank-degQ mls amyE::PswrA -lacZ cat
DB3779	thrC::Physpank-swrA mls amyE::PdegQ -lacZ cat
DB3780	ΔswrA amyE::PswrA-lacZ cat
DB3826	comIQ12L ΔdegS amyE::PswrA-lacZ cat
DB3827	comIQ12L ΔdegS amyE::PdegQ-lacZ cat
DB3835	comIQ12L thrC::Physpank-swrA mls
DB3865	comIQ12L thrC::Physpank-swrA mls amyE::PaprE-lacZ cat
DB3899	comIQ12L slrR::tet
DB3910	comIQ12L slrR::tet amyE::Physpank-degQ spec
DB3959	ΔswrA degQ::tet amyE::PdegQ -lacZ cat
DB3960	ΔswrA degQ::tet amyE::PswrA-lacZ cat
DB3961	ΔswrA thrC::Physpank-degQ mls amyE::PdegQ-lacZ cat
DB3962	ΔswrA thrC::Physpank-degQ mls amyE::PswrA -lacZ cat
DK1042	comI Q12L	(53)
DK4236	comIQ12L ΔdegS
DK5183	comIQ12L amyE::Pfla/che-lacZ cat
DK6294	comIQ12L thrC::Physpank-degQ mls
DK6345	ΔswrA amyE::Pfla/che-lacZ cat
DK6613	comIQ12L amyE::PswrA-lacZ cat	(62)
DK6624	thrC::Physpank-swrA mls amyE::PswrA -lacZ cat
DK6626	ΔdegU amyE::PswrA-lacZ cat
DS2415	ΔswrA	(63)
DS3649	ΔdegU	(63)
DS5648	ΔaprE	(64)

slrR::tet.

The slrR::tet allele was a generous gift from Frances Chu (Harvard University).

degQ::tet.

To generate the degQ::tet insertion deletion construct, DK1042 chromosomal DNA was used as a template to PCR amplify the region upstream of the degQ gene with primer pair 8506/8507 and the region downstream of the degQ gene with primer pair 8508/8509. Next, the tetracycline resistance cassette was amplified from plasmid pDG1515 (54) using primer pair 3250/3251. The three fragments were ligated together using isothermal assembly (55) and amplified with primer pair 8506/8509 prior to transformation into DK1042.

ΔdegS.

To generate the ΔdegS in frame marker-less deletion construct, the region upstream of degS was PCR amplified using the primer pair 5172/5173, the region downstream of degS was PCR amplified using the primer pair 5170/5171, and the two fragments were combined and ligated by isothermal assembly. Next, the isothermal assembly product was used as a PCR template and amplified using primers 5172/5171, the product was digested with BamHI and KpnI and ligated into the BamHI and KpnI sites of the plasmid pMiniMAD (56) which carries a temperature sensitive origin of replication and an erythromycin resistance cassette to generate pAC001. The plasmid pAC001 was purified from the recA+ E. coli strain TG1, transformed into DK1042, plated on LB plates containing mls antibiotic, and incubated at the permissive temperature of 30 °C. An mls resistant colony was struck on a LB mls plate and incubated overnight at the non-permissive temperature of 37 °C to force single crossover integration into the chromosome. Next a colony was incubated in 3 mL LB broth overnight at room temperature and a serially dilution was plated on LB agar at 37 °C. Individual colonies were patched on LB plates and LB plates containing mls to identify mls sensitive colonies that had evicted the plasmid. Chromosomal DNA from colonies that had excised the plasmid was purified and screened by PCR using primers 5172/5171 to determine which isolate had retained the ΔdegS allele.

Complementation constructs.

To generate the degS complementation construct pDP486, the degS gene was amplified from DK1042 chromosomal DNA using primer pair 5299/5301, the amplicon was digested with BamHI/SalI and cloned into the BamHI/SalI sites of plasmid pAH25 containing a spectinomycin resistance cassette and polylinker between two arms of the amyE gene (generous gift of Amy Camp, Mount Holyoke College). To generate the degU complementation construct pDP487, the degU gene was amplified from DK4326 ΔdegS chromosomal DNA using primer pair 5299/5301, the amplicon was digested with BamHI/SalI and cloned into the BamHI/SalI sites of plasmid pAH25.

Inducible constructs.

To generate the inducible degQ construct pEC20, the degQ gene was amplified from DK1042 chromosomal DNA with primer pair 2957/2958, the amplicon was digested with HindIII/NheI and cloned into the HinDIII/NheI sites of pDR111 containing a spectinomycin resistance cassette, the lacI gene, and a polylinker downstream of the P_hyspank promoter between the arms of the amyE gene (generous gift of David Rudner, Harvard Medical School).

To generate the inducible degQ construct pAC016, the degQ gene was amplified from DK1042 chromosomal DNA with primer pair 5310/5311, the amplicon was digested with NheI/SphI, and cloned into the NheI/SphI sites of pDP150 containing an mls resistance cassette, the lacI gene, and a polylinker downstream of the P_hyspank promoter between the arms of the thrC gene (57).

Transcriptional reporter constructs.

To generate the β-galactosidase (lacZ) reporter constructs pDP369 and pAEJ15, PCR products containing the following promoters were amplified for B. subtilis DK1042 chromosomal DNA using the primers indicated in parentheses: P_aprE (8442/8443), and P_degQ (8119/8120). Each PCR product was digested with EcoRI and either BamHI or HindIII and cloned independently into the EcoRI and either BamHI or HindIII sites of plasmid pDG268, which carries a chloramphenicol resistance marker and a polylinker upstream of the lacZ gene between two arms of the amyE gene (58).

β-galactosidase assays.

Three biological replicates of each strain were grown in LB broth at 37 °C to an OD₆₀₀ of 1.7–2.2. Two hundred microliters of culture were pelleted and resuspended in 200 μL of Z-buffer (16.1 g Na₂HPO₄•7H₂O + 5.5 g NaH₂PO₄• H₂O + 0.75 g KCl + 0.246 g MgSO₄•7H₂O in 1 L H2O, pH = 7) with 0.27% β-mercaptoenthanol. Four microliters of 10 mg/mL lysozyme (Dot Scientific) were added, and samples were incubated at 30 °C for 15 min to lyse. In a 96-well plate, lysates were diluted in Z-buffer to 200 μL depending on the strength of the promoter fused to lacZ (P_aprE = 1:1 [undiluted], P_fla/che = 1:10, P_degQ = 1:1, P_swrA = 1:2). Three negative control wells contained only 200 μL of Z-buffer. Forty microliters of 4 mg/mL ortho-nitrophenyl-β-galactopyranoside in Z-buffer was added to each well. The plate was incubated at 30 °C shaking in a double orbital pattern, and the OD420 and OD550 were collected every 2 min. The slope of each sample’s ODs over time was derived. The average slope of the three negative control wells was subtracted from the slope of each experimental trial to control for background. The Miller Units were calculated using the following formula:

Protease assays.

In biological triplicates, cells were inoculated in LB, supplemented with IPTG when appropriate, and incubated rolling at 37 °C until OD₆₀₀ = 1.7–2.2. One milliliter of culture was centrifuged at 6000 × g for 10 min, and 900 μL of supernatant was collected in a fresh tube and incubated with 400 μL of 1% azocasein pH = 8.0 for 30 min at 37 °C. Each reaction was quenched by the addition of 600 μL of 10% trichloroacetic acid and incubated on ice for 30 min. Cells were centrifuged at 16,000 rpm for 5 min, and 800 μL of supernatant was added to 200 μL of 1.8 N NaOH in a clean cuvette. Absorbance was measured at OD₄₄₀, and protease activity was calculated by dividing OD₄₄₀/OD₆₀₀.

RNA extraction and sequencing.

RNA was extracted from B. subtilis cultures grown in LB to an OD600 of ~2.0. Five mL of culture was flash frozen by pre-chilled methanol in an equal volume and centrifuged at 5000 g for 10 min at 4 °C. Supernatant was removed and the pellet was resuspended in 800 μL of LETS buffer (10 mM Tris-HCl pH 7.4, 50 mM LiCl, 20 mM EDTA pH 8.0, 1% SDS) heated to 75 °C. The mixture was added to 650 mg acid-washed glass beads and 600 μL of heated acid-saturated phenol pH 4, and vortexed for 3 min. 600 μL of chloroform was added and the mixture was vortexed for 30 sec before being centrifuged at 3200 g for 10 min at 4 °C. Equal parts of aqueous phase and isopropanol were combined, inverted, and incubated at room temperature for 10 min before centrifugation at 6000 × g for 25 min at 4 °C. Supernatant was discarded, and the pellet was resuspended in a mL of cold 75% ethanol and centrifuged for 5 min at 4 °C. The resuspension was centrifuged at max 6000 × g for 5 min at 4 °C, supernatant removed, pellet air-dried and rehydrated in 20 μL nuclease-free water. Samples were heated for 5 min at 55 °C to encourage resuspension before 300 μL of TRIzol was added, and vortexed for 15 sec. Samples incubated at room temperature for 5 min and combined with 60 μL chloroform. The mixture was inverted and incubated at room temperature for 2 min before centrifugation at max speed for 15 min at 4 °C. The steps starting with the collection of the aqueous layer through the 5 min incubation at 55 °C were repeated. The RNA was treated with RNase-free DNase I at 37 °C for 30 min (Invitrogen) according to the manufacturer’s instructions and RNA was re-extracted via the process above. rRNA depletion, library preparation and analysis were performed by SeqCoast Genomics (Portsmouth, New Hampshire, USA). Illumina paired-end sequencing of the libraries was performed, and at least 12 million paired-end reads were obtained for each sample. Transcript per kilobase million were quantified using Salmon (59) and differential gene expression was calculated with DESeq2 (60) using the NCBI 3610 genome (61). Differentially expressed genes were considered significant with a P-value < 0.05 and a [log₂(fold-change)] > 2.0.

Quantitative swarm expansion assays.

Cells were grown overnight at room temperature in LB broth, back diluted and grown to mid-log phase at 37 °C in LB broth supplemented with 1 mM IPTG when applicable and resuspended to an OD600 of 10 in MQ H₂O containing 0.5% India ink. Freshly prepared LB plates containing 0.7% Bacto agar (25 mL/plate) and IPTG (if applicable) were dried by laminar flow hood for 10 min, centrally inoculated with 10 μL of the cell suspension in ink, dried for another 10 min, and incubated at 37 °C for 6 h. Each strain was surveyed in technical triplicate, and measurements of swarm expansion in millimeters were collected every 30 min. The India ink demarcates the origin of inoculation and for consistency an axis was drawn on the back of the plate and measurements were taken along this axis.

Figure 7. DegQ and SwrA have opposite effects on extracellular protease production and swarming motility.

Schematic depicting the regulatory effects of the DegS-DegU two-component system, DegQ, and SwrA on extracellular protease production and swarming motility. Arrows indicate activation, T-bars indicate repression. Dashed T-bar indicates that DegQ inhibits swarming motility downstream of initiation at the P_fla/che promoter.