The alternative sigma factor SigN of Bacillus subtilis is intrinsically toxic

ABSTRACT Sigma factors bind and direct the RNA polymerase core to specific promoter sequences, and alternative sigma factors direct transcription of different regulons of genes. Here, we study the pBS32 plasmid-encoded sigma factor SigN of Bacillus subtilis to determine how it contributes to DNA damage-induced cell death. We find that SigN causes cell death when expressed at high levels and does so in the absence of its regulon suggesting it is intrinsically toxic. One way toxicity was relieved was by curing the pBS32 plasmid, which eliminated a positive feedback loop that led to SigN hyper-accumulation. Another way toxicity was relieved was through mutating the chromosomally encoded transcriptional repressor protein AbrB, thereby derepressing a potent antisense transcript that antagonized SigN expression. SigN efficiently competed with the vegetative sigma factor SigA in vitro, and SigN accumulation in the absence of positive feedback reduced SigA-dependent transcription suggesting that toxicity may be due to competitive inhibition of one or more essential transcripts. Why B. subtilis encodes a toxic sigma factor is unclear but SigN may function in host-inhibition during lytic conversion, as phage lysogen genes are also encoded on pBS32. IMPORTANCE Alternative sigma factors activate entire regulons of genes to improve viability in response to environmental stimuli. The pBS32 plasmid-encoded alternative sigma factor SigN of Bacillus subtilis however, is activated by the DNA damage response and leads to cellular demise. Here we find that SigN impairs viability by hyper-accumulating and outcompeting the vegetative sigma factor for the RNA polymerase core. Why B. subtilis retains a plasmid with a deleterious alternative sigma factor is unknown.

easy culturing and natural competence which provides facile genetic manipulation (1)(2)(3).The ancestral strain of Bacillus subtilis NCIB3610 is less tractable and carries a large 84 kb plasmid called pBS32 that was lost during the domestication of commonly used laboratory strain derivatives (4)(5)(6).The function of many pBS32-encoded products is unknown, but some have been shown to alter the physiology of the host, including an inhibitor of natural transformation (ComI) (7), an inhibitor of biofilm formation (RapP) (8), and an RNase (RnhP) that contributes to chromosome stability (9).Moreover, nearly half of the episome encodes a large contiguous set of genes resembling a prophage (7), and treatment of cells with the DNA damaging agent mitomycin C (MMC) causes rapid pBS32-dependent cell death (10,11).To date, however, no pBS32-derived phage-like particles have been observed after MMC treatment and large deletions of the putative prophage genome do not abolish MMC-dependent death (10).Precisely how and why the plasmid kills the host is unknown.
MMC-induced pBS32-dependent cell death requires the alternative sigma factor SigN encoded on pBS32 (10,11).Expression of SigN is complex and at least three different promoters drive expression of the sigN gene.The first promoter (P sigN1 ) is repressed by LexA and derepressed by the DNA damage response (11)(12)(13).The second promoter (P sigN2 ) is SigA-dependent and constitutive, and the third promoter (P sigN3 ) is SigNdependent.In the presence of MMC, SigN activates itself and other members of the SigN regulon encoded exclusively on pBS32 (11).Not only is SigN necessary for MMC-induced pBS32-mediated cell death, artificial IPTG-induction of SigN alone is sufficient to cause death when pBS32 is present (10).One way SigN might cause death is if it induced lytic conversion of the putative prophage, but the prophage was not necessary for MMC-induced death and SigN did not appear to activate the prophage structural genes (10,11).Instead, it was presumed that one or more of the genes expressed from the over 20 SigN-dependent promoters on the plasmid was responsible for toxicity.The identity of the toxic gene or genes under SigN control and the mechanism of SigN-dependent cell death were unknown.
Here we explore the mechanism of SigN-mediated cell death by selecting suppressor mutations that restored growth in the presence of artificial SigN induction.Through the analysis of the suppressor mutants in SigN, we found that artificial induction of wild-type SigN became toxic even in the absence of pBS32 when expressed at higher levels.Other suppressors spontaneously cured pBS32 and still others mutated AbrB thereby derepressing an anti-sense sigN transcript that antagonized SigN accumulation.In sum, we show that at least one toxic product under SigN control is SigN itself.Thus, when the sigN transcript exceeds a threshold determined by the anti-sense transcript abundance, the SigN protein is made and initiates a positive feedback loop.SigN toxicity appears to be due to its accumulation combined with its ability to outcompete SigA for RNA polymerase and impair vegetative gene expression.SigN is the founding member of a new subfamily of the sigma 70 family of sigma factors, but the selective advantage, if any, of SigN activity is unclear.

Spontaneous suppressors alleviated SigN-mediated cell death
An IPTG-inducible sigN construct using the native sigN ribosome binding sequence (nRBS) was toxic when introduced to a pBS32-proficient strain such that transformant colonies appeared to be sick even in the absence of IPTG (10).An IPTG-inducible sigN construct using a weakened ribosome binding sequence (wkRBS), however, caused a decrease in optical density (OD 600 ) only in the presence of IPTG (10,11), but the OD loss was transient, and growth typically recovered 4-5 h post induction (Fig. 1A).We inferred that growth recovery was due to genetic suppression.Cultures in which the OD had recovered were dilution plated for single colonies and three distinct classes of suppressor of sigN-mediated death (ssd) mutants were separated by their colony morphology.Class I mutants had a "ridged" colony morphology and were numerically dominant, whereas class II and III mutants were rare and had either a "wild-type" colony morphology or "hyper-rough/hyper-white" colony morphology, respectively (Fig. 1B; Table 1).When representatives of each class were isolated and re-challenged with IPTG, cells grew like the wild type and no indication of OD loss was observed.We conclude that SigN-medi ated cell death may be genetically suppressed, and we infer that suppressor mutations likely arose in at least three different genetic loci.

Class I ssd mutants were cured of pBS32
SigN-mediated cell death was previously shown to require the presence of the pBS32 plasmid (10).Thus, one way in which ssd mutations might abolish SigN-mediated cell death is by spontaneous curing of pBS32.To determine whether pBS32 was present in any of the ssd suppressor strains, polymerase chain reaction (PCR) was used to amplify three different loci on the plasmid, zpaB, zpcJ, and sigN (using primers that hybridize to the native site and not the ectopic integration construct), and one locus on the chromosome as a control (fliG).Whereas the class II and class III ssd mutants amplified all plasmid-encoded loci, the class I ssd mutants did not (Fig. 2A).We conclude that class I ssd mutants abolished SigN-mediated cell death by eliminating pBS32 from the genome (Table 1).We note that each class I ssd allele conferred a ridged colony morphology that phenocopied a strain that was force-cured for pBS32 (7, 8) (Fig. 1B).
Because the class I mutants were cured for the pBS32 plasmid and were the most frequent mutant class, we set out to determine the frequency of plasmid curing in the presence and absence of sigN induction.To do so, we constructed a chromosomally integrated reporter construct in which the lacZ gene, encoding the β-galactosidase LacZ, was expressed from the P eps promoter, the activity of which is repressed by the biofilm repressor SinR (14,15).Next, we deleted the chromosomal copies of the genes encoding SinR and its antagonist SinI (16), and reintroduced a copy of the sinR gene, expressed from its native promoter (17), on the pBS32 plasmid.Thus, when pBS32 expressing SinR was present, P eps -lacZ activity would be inhibited and colonies would appear white on plates containing the chromogenic substrate, X-gal (Fig. 2B).If pBS32 were lost, however, the gene encoding SinR would also be lost, P eps -lacZ expression would be de-repressed, and the colony would appear blue.
In the absence of SigN induction, approximately 1 in 3,000 colonies of a late exponential phase culture (~1 OD 600 ) was blue when plated on media containing X-Gal indicating that the frequency of spontaneous plasmid curing was roughly 0.03%, a frequency lower than what is noted for the related pLS32 miniplasmid derivative pBET131 (0.5%) (18).To test the effect of SigN, IPTG was added to the culture, and cells were dilution plated on media containing X-gal that lacked inducer at various timepoints.After induction of SigN, the frequency of blue colonies increased exponen tially relative to the white colonies (Fig. 2C).We conclude that most cells died after SigN induction but the subpopulation that had spontaneously cured the pBS32 plasmid prior to induction survived and proliferated.Consistent with the existence of class II and class III suppressors, rare white colonies were observed even after prolonged SigN-induction (Fig. 2C), suggesting that pBS32 remained in a subpopulation and that there were other mechanisms of SigN-resistance besides plasmid curing.

Class II ssd mutants have mutations in the IPTG-inducible sigN construct
Another way in which IPTG-induced SigN-mediated cell death might be abolished is by a loss-of-function mutation in the IPTG-inducible sigN construct.To test for sigN function ality, the IPTG-inducible sigN constructs from the remaining mutants were backcrossed into a wild-type background; the resulting strains were induced with IPTG and monitored for growth.Backcrossed IPTG-induced sigN alleles of the class II, but not class III, ssd mutants failed to decrease in OD after induction, indicating that the inducible sigN alleles of class II mutants were non-functional.Sequencing of the IPTG-inducible sigN gene indicated that one ssd allele mutated the −10 element of the IPTG-inducible promoter away from consensus, likely impairing expression (Table 1).The remaining five alleles were mutated in the sigN open-reading frame including two likely loss-of-function nonsense mutations (SigN W59stop and SigN Q35stop ) and three missense mutations (SigN E55G , SigN K83R , and SigN I198T ) (Table 1).We conclude that the class II ssd mutants eliminated SigN-mediated cell death by either reducing or abolishing SigN expression or activity.
To determine whether the missense mutations in sigN impaired or abolished SigN accumulation, we set out to detect SigN protein in lysates.After 2 h of IPTG induction, levels of SigN protein were high in the wild type but reduced in the sigN mutants (Fig. 3A).One reason that the mutant substitutions might experience reduced SigN protein levels is if each caused severe structural defects and resulted in rapid proteolysis after synthesis.Alternatively, the mutants might make fully stable proteins but be impaired for activity and exhibit reduced protein level due to the nature of the experimental set-up.An activity defect could impair accumulation because the IPTG-inducible SigN construct, already restricted for accumulation by a weakened RBS (wkRBS), may require auto-activation of its own expression at the native sigN gene on the pBS32 plasmid to produce detectable levels of protein.Thus, high levels of SigN protein may require SigN functionality as non-functional alleles might fail to stimulate positive feedback at the native site.We conclude that the IPTG-inducible SigN mutant alleles produced low levels of protein, but precisely why protein levels were low remained uncertain.
To determine whether SigN protein could be produced at detectable levels in the absence of positive feedback, the wild-type allele was expressed from an IPTG-inducible a Class II ssd mutants are mutants that relieve toxicity by mutation of the sigN gene and were later subdivided into subclasses a and b.Class IIa ssd alleles are missense mutations isolated from a strain containing pBS32 induced for wkRBS sigN expression.Class IIb ssd alleles are missense mutations isolated from a strain lacking pBS32 induced for nRBS sigN expression.
b" Hyper-rough" colony morphology was distinguishable because it appeared more rugose than the "rough" architecture, was bright white in appearance, and the entire colony had the tendency to leave the agar when prodded with a toothpick.
promoter using either the native SigN RBS (nRBS) or wkRBS sequence and inserted at an ectopic site in a laboratory strain lacking pBS32.Unlike in the presence of pBS32, integration of the IPTG-inducible wild-type sigN allele with nRBS did not appear to impair transformation or reduce growth rate in the absence of inducer.Moreover, the SigN protein was observed after induction but only when expressed using the nRBS sequence, supporting the notion that the wkRBS construct required positive feedback on pBS32 to produce detectable levels (Fig. 3B).We noted however that expression of SigN with the nRBS inhibited cell growth even in the strain lacking pBS32 whereas SigN with the wkRBS did not (Fig. 4A).We conclude that expression of SigN from its nRBS is sufficient to produce detectable levels of protein in the absence of positive feedback from the pBS32 plasmid.We further conclude that elevated levels of SigN protein were able to inhibit growth by a mechanism that was pBS32-independent.Next, each of the class II sigN alleles was artificially expressed with the nRBS in a strain lacking pBS32.Induction of each of the class II missense alleles produced levels of SigN protein comparable to the wild type suggesting that they were each likely defective in SigN function, rather than SigN protein stability (Fig. 3C).The alleles were not completely defective in SigN activity however as induction of each of mutant allele at least partially inhibited growth in the absence of pBS32 (Fig. 4B).We conclude the class II mutant alleles expressed from the wkRBS abolished cell death in the presence of pBS32 because they lacked an activity of SigN necessary to initiate positive feedback.We further conclude that the alleles retained an activity that inhibited growth by a pBS32-independ ent mechanism when their levels were elevated by induction and the native RBS.
To isolate mutant variants of SigN which abolished pBS32-independent growth inhibition (as was observed in a single cell line of Fig. 4A, left), the IPTG-inducible nRBS sigN construct was grown to high density in the absence of inducer and then plated on media containing IPTG.Rare colonies survived, five were clonally isolated, and sequencing revealed that each strain contained a mutation in the sigN open reading frame.Each of the new alleles that abolished pBS32-independent cell death also abolished death when crossed into a background containing pBS32, and were therefore classified as a subset ("b") of ssd class II alleles (Fig. 4C; Table 1).Finally, the level of SigN protein from each ssd curing-reporter strain that also encodes a chromosomally encoded IPTG-inducible sigN gene with weakened RBS (DK7867).Culture was induced with 1 mM IPTG at time 0, aliquots were taken over time, dilution plated on media containing X-gal, incubated overnight, and counted for the percentage of blue colonies (blue dots) and white colonies (white dots).
class IIb allele was comparable to the wild type in western blot analysis (Fig. 3C).We conclude that ssd class IIb alleles were generally defective for SigN activity regardless of whether or not pBS32 was present.We further conclude that SigN likely inhibits growth by a single mechanism and that the pBS32-dependency was largely due to the positive feedback on SigN expression necessary to elevate SigN to toxic levels.
As the SigN regulon is restricted to pBS32 and does not appear to stimulate transcrip tion from the chromosome (11), the mechanism of pBS32-independent cell death is unlikely due to a secondary gene product and likely due to intrinsic inhibition by the SigN protein itself.One way in which SigN might be intrinsically toxic is if it were able to outcompete SigA for the RNA polymerase core.To observe sigma competition, SigN was added to an in vitro transcription assay containing the RNA polymerase core, the vegetative sigma factor SigA, and a DNA fragment containing the SigA-dependent P sigN2 and the SigN-dependent P sigN3 promoters (19).In the absence of SigN, SigA-containing RNA polymerase synthesized the SigA-dependent transcript (Fig. 5A).As SigN levels rose in the reaction, SigN-dependent transcripts increased, while SigA-dependent transcripts decreased such that a two-fold excess of SigN was sufficient to inhibit SigA transcription by 50%.Increasing the amount of the alternative stress response sigma factor SigB did not reduce the SigA-dependent transcript from the same template, suggesting that SigN was a stronger inhibitor and better competitor for the RNAP core (Fig. 5B; Fig S1) (20,21).We conclude that SigN is a potent competitive inhibitor of the vegetative sigma factor and we infer that SigN-mediated growth inhibition may be due to a reduction of essential vegetative transcripts.

Class III ssd alleles have mutations in the transcriptional repressor, AbrB
The class III ssd mutants were not mutated for the inducible sigN construct nor were they cured of pBS32.We inferred that the class III mutants contained loss-of-function mutations like the class II mutants, because the two classes occurred at approximately (C) Each of the indicated SigN amino acid substitution mutants selected to abolish pBS32-dependent cell death was expressed from an IPTG-inducible promoter and native SigN RBS ( nRBS SigN) in a strain lacking pBS32.The following strains were used to generate this panel: DK9210 (SigN E55G ), DK9211 (SigN K83R ), and DK9209 (SigN I198T ).(D) Each of the indicated SigN amino acid substitution mutants selected to abolish pBS32-independent cell death was expressed from an IPTG-inducible promoter and native SigN RBS ( nRBS SigN) in a strain lacking pBS32.The following strains were used to generate this panel: DK9282 (SigN R22G ), DK9283 (SigN L136P ), DK9345 (SigN D156G ), DK9342 (SigN N160S ), and DK9325 (SigN Y165V ).For each lane, cells were induced at 0.5 OD 600 or less with 1 mM IPTG for 2 h prior to harvesting cell pellets for lysis.Panels B, C, and D are from the same gel and are separated by dashed lines to facilitate citation in text.
the same rare frequency in the spontaneously suppressed population.To find the location of the class III mutations, a strain containing the IPTG-inducible P hyspank -wkRBS sigN construct was mutagenized with a mariner transposon carrying a chloramphenicol resistance cassette, and the mutagenized pool was plated on media supplemented with both chloramphenicol and IPTG.Some of the rare colonies that grew had a hyper-rough/ hyper-white colony morphology (like the spontaneous class III mutants), two of which were isolated and backcrossed to ensure linkage between the chloramphenicol resist ance, colony morphology phenotype, and survival in the presence of inducer.Finally, the location of the transposon was determined and for both strains, the transposon was inserted within the chromosomally encoded gene, abrB (Fig. 1B and Fig. 6A).Sequencing of abrB in each of the class III ssd spontaneous mutants indicated the presence of mutations within the gene that altered either the protein sequence or likely reduced translation of the gene product (Table 1).
The abrB gene encodes AbrB, a DNA binding transcription factor (22,23) and a known transcriptional repressor of biofilm development and antimicrobial production (24)(25)(26).One way in which AbrB might alleviate SigN-induced cell death is if mutation of abrB abolished or reduced SigN expression.Consistent with a SigN expression defect, no SigN was detected in an abrB mutant after wkRBS SigN was induced for 2 h in the presence of pBS32, but SigN protein was detected when abrB was complemented under the control of its native promoter and integrated at an ectopic site in the chromosome (Fig. 6B).Moreover, SigN protein accumulated in the wild type 2 h after induction from the native site by treatment with the DNA damaging agent mitomycin C (MMC) (11), and MMC-dependent accumulation was abolished in the abrB mutant (Fig. 6C).We conclude that mutation of abrB prevents SigN-mediated death by preventing SigN accumulation, and does so whether SigN is induced artificially by IPTG or by the DNA damage response at the native site in the chromosome.
One way that mutation of abrB could impair accumulation of SigN is if it altered sigN transcription.To measure sigN transcription, a reporter in which the P sigN promoter region was fused to the lacZ gene encoding β-galactosidase was introduced to a strain with an IPTG-inducible wkRBS sigN construct and lacking pBS32.Consistent with previous reports, IPTG-induction of sigN induced expression of the P sigN -lacZ reporter over 50-fold (11), and mutation of abrB did not diminish P sigN induction (Fig. 6D).We conclude that AbrB does not directly act upon the P sigN promoter nor does it inhibit SigN activity.To determine whether AbrB might inhibit SigN indirectly through a pBS32-encoded product, the reporter expression experiments were conducted in a pBS32-containing strain.Consistent with an indirect effect, induction of the P sigN promoter was reduced when pBS32 was present, and mutation of AbrB abolished induction entirely (Fig. 6D).We conclude that AbrB indirectly activates SigN and does so in a manner that depends on a pBS32-derived product.
One possible candidate for a pBS32-derived regulator of SigN is a putative antisense RNA initiated downstream of, and transcribing backwards through, the sigN gene (27).To determine whether a promoter resides downstream of, and oriented toward, the sigN gene, the putative anti-sigN promoter region (P asiN ) was cloned upstream of the lacZ gene and inserted at an ectopic site in the chromosome.Consistent with the presence of one or more functional promoters, β-galactosidase activity was detected from the P asiN region and expression from P asiN increased when AbrB was mutated (Fig. 6E).REND-seq predicted three different possible start sites that could correspond to different promoter elements (27) and the P asiN region was divided into three separate fragments, P asiN1 , P asiN2 , and P asiN3 , each cloned separately upstream of the lacZ gene.The P asiN1 promoter expressed at a high level and appeared constitutive, while the P asiN2 putative promoter gave no activity either in wild type or in an abrB mutant (Fig. 6E).Consistent with repression however, the P asiN3 promoter produced very low transcript in wild type but expression increased 100-fold when abrB was mutated (Fig. 6E).We conclude that at least two promoters drive anti-sense sigN transcription, one of which is constitutive and one of which is tightly repressed by AbrB.
Each of the ectopic, inducible sigN constructs used thus far was built excluding the region downstream of sigN that contains the anti-sense promoters.To determine the biological relevance of the antisense transcript, the sigN gene was cloned with a wkRBS and the extended downstream region under the control of an IPTG inducible promoter (P hyspank -wkRBS sigN EXT ).Next, the expression construct was inserted at an ectopic site in strains containing the P sigN -lacZ reporter but lacking pBS32 (to eliminate the contribution of natively produced antisense transcript).Expression from P sigN decreased slightly in the abrB mutant when SigN was induced with 1 mM IPTG, but expression decreased dramatically when the amount of inducer was reduced 10-fold (Fig. 6F).We conclude that the AbrB effect was at least partially linked to the region downstream of sigN.We further conclude that de-repression of the P asiN3 promoter and production of the antisense sigN transcript was able to restrict SigN to levels insufficient for positive feedback.Repression by AbrB is relieved during the transition to stationary phase.To determine whether the ability of SigN to achieve lethal positive feedback was dependent on growth phase, a strain containing pBS32 and inducible wkRBS sigN was induced at various cell densities.When 1 mM of IPTG was added to cultures of 0.25 OD or less, OD increased for 30 min before steadily decreasing consistent with SigN toxicity (Fig. 7A).The toxicity of SigN decreased as the OD at induction increased above 0.25, however, and any effect on growth became undetectable when induced at OD of 0.60 and above (Fig. 7A).Moreover, the degree of SigN toxicity was correlated with SigN levels where SigN levels after 2 h of induction were the highest when induced at low cell density (Fig. 7B).We conclude that the ability of SigN to achieve toxic positive feedback is growth phase dependent and restricted to exponential phase.We infer that the growth phase dependency is due to AbrB repression of anti-sigN antisense transcript such that as AbrB becomes antagonized during the transition to stationary phase, the level of antisense transcript increases to levels greater than can be overcome by wkRBS sigN induction.

SigN induction inhibits SigA-dependent transcription
Our data indicated that high levels of SigN were correlated with toxicity, and toxicity was correlated with the ability of SigN to inhibit SigA-dependent transcription in vitro.To determine whether SigN induction inhibited SigA-dependent transcription in vivo, gene expression was measured in a strain lacking pBS32 and containing an nRBS sigN inducible construct such that growth inhibition could be achieved without SigN positive feedback or the SigN regulon (Fig. 4A).We first explored the expression of the SigN-dependent transcriptional reporter P zpdG -lacZ integrated at an ectopic site in the chromosome (11).In the absence of the inducible construct, β-galactosidase activity was at background levels (<2 MU), but the presence of the construct increased expression to over 600 MU even in the absence of inducer (Fig. 8A).Expression from P zpdG increased another 10-fold after 1 h of induction (Fig. 8A).Under the same conditions, adding inducer decreased expression from three different SigA-dependent reporters 3-fold (Fig. 8B, C, and D) (28,29).Reverse transcriptase quantitative PCR indicated that transcript levels changed in proportion to the level of LacZ-activity increase for the P zpdG reporter (Fig. 8A) and the level of LacZ-activity decrease for the rpsA and ptsG reporters (Fig. 8B, C,  and D).We conclude that induction of SigN in the absence of positive feedback was sufficient to decrease SigA-dependent expression in vivo.We further conclude that SigN is intrinsically toxic due to its ability to outcompete the vegetative sigma factor SigA from the RNA polymerase core.

DISCUSSION
SigN appears to be the founding member of a new sub-family of sigma 70 sigma factors, and SigN activates its own expression and a regulon of genes on pBS32 (11).Here we set out to determine which gene or genes under SigN control caused growth inhibition, and analysis of viable suppressor mutations supported the conclusion that at least one of the toxic genes was SigN itself.Growth inhibition by SigN was prevented by abrogating sigN positive feedback, (either by curing the pBS32 plasmid, or by increasing the levels of a sigN antisense transcript), or by mutating the inducible copy of SigN.Moreover, SigN was intrinsically toxic in the absence of the rest of its regulon and toxicity was correlated with the ability of SigN to outcompete vegetative SigA for the RNA polymerase core in vitro and in vivo.In the absence of positive feedback, SigN accumulation reduced SigA-dependent gene expression 3-fold, which when applied to a low abundance essential gene is likely sufficient to inhibit growth (Fig. 9A).In the presence of pBS32 however, positive feedback causes SigN hyperaccumulation and re-analysis of previously obtained REND-seq data appears to indicate even greater inhibition of SigA-dependent transcripts (Fig. S2).We conclude that SigN is an intrinsically toxic sigma factor and likely inhibits growth by indirectly outcompeting SigA for the RNA polymerase core and reducing expression of one or more essential genes to sub-viable levels.Sigma factors initiate transcription by binding to the RNA polymerase core and recruiting the complex to a specific DNA sequence called the promoter.Two different classes of rescue-of-viability mutations were found in SigN: one class was conditionally dependent on protein level (ssd class IIa) and one class conferred a general loss-of-func tion (ssd class IIb) (Table 1).The class IIa alleles were isolated from an inducible sigN construct with a weakened RBS such that toxicity depended on positive feedback but were still toxic when protein levels were elevated by other means.One way the class IIa alleles could have been pBS32-dependent is if they altered residues required for promoter recognition, thus abolishing positive feedback, but remained toxic in the ability to sequester the RNA polymerase core.When the residues altered by the class IIa alleles were mapped onto an Alphafold-predicted structure of SigN in complex with RNA polymerase however, all of the altered residues were clustered in a single location that did not seem to be positioned for DNA interaction (Fig. 9B).We infer either that the Alphafold prediction is partially inaccurate, that SigN requires DNA containing the promoter to assume a proper conformation, or that the sructure points to additional complexity in the mechanism of SigN.The class IIb alleles mapped to a variety of locations on the SigN structure and we infer that at least some of the alterations likely abolish interaction with the RNA polymerase core to render them fully non-toxic (Fig. 9B).
The level of toxicity of SigN is controlled, at least in part, by the level of its accumula tion.One way to control SigN levels is transcriptional by its complex native promoter region including a basal constitutive promoter (P sigN2 ), a promoter de-repressed by the DNA-damage response (P sigN1 ), and a promoter that is SigN-dependent (P sigN3 ) (Fig. 9A).Normally, the constitutive promoter is insufficient to initiate positive feedback and requires the DNA-damage response to push SigN expression above the feedback threshold (11).The threshold however could also be exceeded artificially, and the original report of SigN suggested that even expression from an uninduced IPTG-indu cible promoter was sufficient to cause morbidity (10).Accordingly, the SigN RBS was

Full-Length Text
Journal of Bacteriology weakened to make toxicity dependent on both IPTG and pBS32, and here we show that perhaps the only contribution of pBS32 was to provide the source of positive feedback.Using a newly developed assay, we find that B. subtilis spontaneously cures pBS32 at a rate of approximately 1 in 3,000 cells (18,30,31), and induction of SigN selected for those cells that had spontaneously lost the plasmid.We note that when selecting for suppressors in the presence of pBS32, mutations either cured pBS32 entirely or were found in the chromosomally integrated inducible copy of sigN.We find it curious that no mutations were recovered in either the native sigN allele or in the promoter for positive feedback but their absence may point to some as-yet-unknown aspect of pBS32 molecular biology or gene expression.SigN levels are also regulated by cis-encoded antisense transcripts.REND-seq analysis predicted that one or more antisense transcripts overlapped sigN (27) and here we identify at least two different promoters downstream of, and oriented toward, the sigN gene (Fig. 9A).One of the antisense promoters (P asiN1 ) appears to overbalance the constitutive sense transcript from P sigN2 and silence sigN expression.Thus, production of sigN sense transcript must exceed the abundance of the antisense transcript to produce protein.The transition state regulator AbrB represses, either directly or indirectly, another equally strong antisense promoter (P asiN3 ).AbrB is associated with the repression of a wide variety of genes that become active in post-exponential phase including those for antimicrobial production and biofilm formation (24)(25)(26).Here AbrB seems to indirectly promote SigN expression.Why a host cell factor would regulate sigN so as to lower the expression threshold and make it easier to initiate a lethal positive feedback loop is unclear.Gene repression by AbrB is relieved during the transition to stationary phase (32)(33)(34) and we show that SigN positive feedback, hyperaccumulation, and toxicity gradually fail at later stages of the growth curve.Whatever the purpose of SigN, it seems to be selectively antagonized outside of exponential growth.Why and how a plasmid with at least one toxic product on it has been stabilized in the population is unclear, and the selective advantage, if any, of either SigN or its regulon is unknown.
Many alternative sigma factors, particularly the Sigma 70 ECF subfamily, experience positive feedback and only a few sigmas, such as B. subtilis SigM and SigG, have been shown to exhibit deleterious effects on growth when dysregulated (35,36).For deleterious sigmas, it is often difficult to rule out whether one or more targets in the regulon cause toxicity indirectly.Moreover, chromosomally encoded alternative sigma factors tend to have lower affinity for RNA polymerase core than the vegetative sigma (37)(38)(39) and are often co-expressed with anti-sigma factors such that they may be antagonized before, during, and/or after expression (40)(41)(42).B. subtilis SigN is different in that it is plasmid-encoded, toxic in the absence of its regulon and as yet there is no known anti-sigma factor to hold it in check.Instead, antisense transcripts raise the expression threshold for SigN accumulation, and while perhaps a first for sigma factor control, antisense transcripts are known regulators in horizontally transferred elements such as plasmids, and bacteriophages (43)(44)(45)(46)(47).An autoactivating sigma factor with affinity for the RNA core polymerase on par with SigA seems inherently detrimental to the host and it has been speculated that pBS32 is in fact a plasmid prophage like P1 in E. coli (6,48).If pBS32 is in fact a prophage in its entirety, override of essential transcription from the host chromosome may be a strategy for phage hyper-proliferation and thus phage survival.

Strains and growth conditions
B. subtilis strains were grown in lysogeny broth (LB) (10 g tryptone, 5 g yeast extract, 5 g NaCl per L) or on LB plates fortified with 1.5% Bacto agar at 37°C.When appropriate, antibiotics were used at the following concentrations: 5 µg/mL kanamy cin, 100 µg/mL spectinomycin, 5 µg/mL chloramphenicol, 10 µg/mL tetracycline, and 1 µg/mL erythromycin with 25 µg/mL lincomycin (mls).Mitomycin C (MMC, DOT Scientific) was added to the medium at a final concentration of 0.3 ug/ml (note that the u is the micro symbol) when appropriate.Isopropyl β-D-thiogalactopyranoside (IPTG, Sigma) was added to the medium as needed at the indicated concentration.

Strain construction
All constructs were first introduced into strains by transformation and then mobilized using SPP1-mediated generalized phage transduction (7,49).All strains used in this study are listed in Table 2.All primers used in this study are listed in Table S1.All plasmids used are listed in Table S2.The abrB::kan allele was obtained from the Bacillus subtilis Genetic Stock Center (BGSC, The Ohio State University) (50).

LacZ reporter constructs
To generate the β-galactosidase (lacZ) reporter thrC::P eps -lacZ mls amp, the promoter region of epsA was amplified with PCR using the primer set 709/3025 from B. subtilis 3610 chromosomal DNA.This DNA fragment was digested and cloned with BamHI and EcoRI into pDG1663 (54), which carries an erythromycin-resistance marker and a polylinker upstream of the lacZ gene between the two arms of the thrC gene to create pDP146.
To generate the pATB28 thrC::P asiN -lacZ mls reporter construct, the region down stream of sigN was PCR amplified from B. subtilis 3610 genomic DNA using primers 6576/6577, digested and cloned with BamHI and EcoRI into pDG1663.To generate the pATB51 thrC::P asiN1 -lacZ mls reporter construct, the region downstream of sigN was PCR amplified from B. subtilis 3610 genomic DNA using primers 6576/6909, digested and cloned with BamHI and EcoRI into pDG1663.To generate the pATB52 thrC::P asiN2 -lacZ mls reporter construct, the region downstream of sigN was PCR amplified from B. subtilis 3610 genomic DNA using primers 6910/6911, digested and cloned with BamHI and EcoRI into pDG1663.To generate the pATB53 thrC::P asiN3 -lacZ mls reporter construct, the region downstream of sigN was PCR amplified from B. subtilis 3610 genomic DNA using primers 6912/6577, digested and cloned with BamHI and EcoRI into pDG1663.

Inducible wkRBS sigN EXT construct
To generate the inducible sigN construct pATB70, the sigN gene plus downstream region were PCR amplified using primer pair 3959/7068 and 3610 genomic DNA as a template.The PCR product was purified, digested with NheI/SphI, and cloned into the NheI/SphI sites of plasmid pDR111 containing a polylinker downstream of the P hyspank promoter with a spectinomycin resistance cassette all between the arms of the amyE gene (generous gift from David Rudner, Harvard Medical School).

Plasmid curing reporter system
To generate the β-galactosidase (lacZ) blue-white plasmid curing reporter strain, an ΔepsH in-frame markerless deletion construct was generated by transforming pSG37 into DK1042 and selecting for resistance to mls at 37° (55).The resulting strain was incuba ted in 3 mL LB at the permissive temperature for plasmid replication (22°C) for 14 h and the culture was serially diluted and plated on LB agar at 37°C.Individual colonies were replica patched onto LB plates and plates containing mls to identify mls-sensitive colonies that had potentially evicted the plasmid.Colonies that were mls-sensitive were screened for smooth colony morphology indicative of a defect in extracellular polysac charide production and retention of the ΔepsH allele (DK7719) was confirmed by PCR length polymorphism using primers 2114/2117.
Next, a markerless P sinR -sinR complementation construct was introduced into the ΔepsH strain DK7719 at the ectopic site comI on the pBS32 plasmid.To achieve this, the comI 5′ region was PCR amplified with primers 7122/7123 and digested with EcoRI and NheI, and the comI 3′ region was amplified with primers 7124/7125 and digested with NheI and SalI.The two fragments were simultaneously ligated into the EcoRI and SalI sites of pMiniMAD (56) to generate pDP532.Next, the P sinR -sinR region was PCR amplified with primers 7126/7127, digested with NheI and KpnI, and ligated into the NheI and KpnI of the comI plasmid pDP532 to create pDP533.Plasmid pDP533 was transformed into DK7719 by transformation and selection for mls resistance at 30°C.One colony was restruck on LB with mls at 37°C overnight to force integration of pDP533 into pBS32 and then regrown overnight in 3 mL LB at 30°C to promote pDP533 eviction.The culture was dilution plated for single colonies on LB at 37°C overnight.Retention of P sinR -sinR at comI was confirmed by colony PCR using primers 7122/7125 to generate the markerless P sinR -sinR integrant DK7804.Finally, the native copy of sinIR was interrupted with a spectinomycin antibiotic cassette (14).Then, a thrC::P eps -lacZ reporter with an erythromycin resistance cassette and an amyE::P hyspank -wkRBS sigN construct with a chloramphenicol resistance cassette were sequentially introduced by SPP1-mediated phage transduction to generate DK7867.

SPP1 phage transduction
To a 0.2-mL dense culture grown in TY broth (LB supplemented with 10 mM MgSO 4 and 100 µM MnSO 4 after autoclaving), serial dilutions of SPP1 phage stock were added.This mixture was allowed to statically incubate at 37°C for 15 min.A 3-mL volume of TYSA (molten TY with 0.5% agar) was added to each mixture and poured on top of fresh TY plates.The plates were incubated at 37°C overnight.Plates on which plaques formed had the top agar harvested by scraping into a 50-mL conical tube.To release the phage, the tube was vortexed for 20 s and centrifuged at 5,000 × g for 10 min.The supernatant was passed through a 0.45-µm syringe filter and stored at 4°C.Recipient cells were grown in 2 mL of TY broth at 37°C until stationary phase was reached.A 5-µL volume of SPP1 donor phage stock was added to 0.9 of cells and 9 mL of TY broth was added to this mixture.The transduction mixture was allowed to stand statically at room temperature for 30 min.After incubation, the mixture was centrifuged at 5,000 × g for 10 min, the supernatant was discarded, and the pellet was resuspended in the volume left.One hundred to two hundred microliters of the cell suspension was plated on TY fortified with 1.5% agar, 10 mM sodium citrate, and the appropriate antibiotic for selection.

Western blotting
B. subtilis strains were grown in LB and treated with IPTG or MMC (final concentration 0.3 µg/mL) as reported in reference (10).Cells were harvested by centrifugation at 1 h after treatment unless specified.Cells were resuspended to 10 OD 600 in Lysis buffer [20 mM Tris-HCL (pH 7.0), 10 mM EDTA, 1 mg/mL lysozyme, 10 µg/mL DNAse I, 100 µg/mL RNAse I, 1 mM PMSF] and incubated for 1 h at 37°C.Twenty microliters of lysate was mixed with 4 µL 6 × SDS loading dye.Samples were separated by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).The proteins were electroblotted onto nitrocellulose and developed with a primary antibody used at a 1:5,000 dilution of anti-SigN, 1:80,000 dilution of anti-SigA, and a 1:10,000 dilution of secondary antibody (horseradish peroxidase-conjugated goat anti-rabbit immunoglobu lin G).Immunoblot was developed using the Immun-Star HRP developer kit (Bio-Rad).

β-galactosidase assay
Biological replicates of B. subtilis strains were grown in LB and induced with IPTG to a final concentration of 1 mM.Cells grew to an OD 600 0.6 and 1 mL was harvested.Cells that contained the plasmid, pBS32, were induced with IPTG and allowed to grow for 1 h and then harvested by centrifugation.Cells were resuspended in 1 mL of Z-buffer (40 mM NaH 2 PO 4 , 60 mM Na 2 HPO 4 , 1 mM MgSO 4 , 10 mM KCl, and 38 mM β-mercaptoethanol) with 0.2 mg/mL of lysozyme and incubated at 30°C for 15 min.Each sample was diluted accordingly with Z-buffer to 500 µL.The reaction was started with 100 µL of 4 mg/mL O-nitrophenyl β-D-galactopyranoside (in Z buffer) and stopped with 1M Na 2 CO 3 (250 µL).The OD 420 of each reaction was noted and the β-galactosidase specific activity was calculated using this equation: [OD 420 /(time × OD 600 )] × dilution factor × 1,000.
Cells containing the plasmid for overproduction of SigN (LK2531) were grown to OD 600 ~0.5 when IPTG was added to a final concentration of 0.3 mM.Cells were then allowed to grow for 3 h at room temperature, harvested, washed, and resuspended in P buffer (300 mM NaCl, 50 mM Na2HPO4, 3 mM β-mercaptoethanol, 5% glycerol).All purification steps were done in P2 buffer (the same composition as P buffer, but pH 9.5).Cells were then disrupted by sonication and the supernatant was mixed with 1 mL Ni-NTA agarose (QIAGEN) and incubated for 1 h at 4°C with gentle shaking.Ni-NTA agarose with the bound His-SUMO-SigN was loaded on a Poly-Prep Chromatography Column (Bio-Rad), washed with P2 buffer and subsequently with the P2 buffer with the 30 mM imidazole.The protein was eluted with P2 buffer containing 400 mM imidazole and fractions containing His-SUMO-SigN were pooled together and dialyzed against P2 buffer.The SUMO tag was subsequently removed by SUMO protease (Invitrogen).The cleavage reaction mixture was then mixed with 1 mL Ni-NTA agarose and allowed to bind for 1 h at 4°C and centrifuged to pellet the resin.Supernatant was removed, the resin was washed once more with P2 buffer with 3 mM β-ME.The supernatants (containing SigN) were pooled together and dialyzed against storage P2 buffer (P2 buffer and 50% glycerol).The protein was stored at −20°C.

Transcription in vitro
Multiple round transcriptions in vitro were performed with the B. subtilis RNAP core reconstituted with SigA (ratio 1:1) and either without or with increasing amounts of SigN or SigB (ratios 1:0.5, 1:1, 1:2, 1:4, 1:8; RNAP was 1) in storage buffer (50 mM Tris-HCl, pH 8.0, 0.1 M NaCl, 50% glycerol) for 30 min at 30°C (52,53).First, the two sigma factors were mixed together and then the RNAP core was added.The obtained holoenzymes reflected the competition between these sigma factors for the core.Reconstituted holoenzymes were then used in multiple round transcription reactions in 10 µL reaction volumes.The transcription buffer contained 40 mM Tris-HCl pH 8.0, 10 mM MgCl2, 1 mM dithiothreitol (DTT), 0.1 mg/mL BSA, 150 mM KCl, and all four NTPs (200 µM ATP, CTP, GTP each) plus 10 µM UTP and 2 µM radiolabeled [α-32 P] UTP.Every reaction also contained supercoiled DNA template (100 ng/reaction).The LK2672 plasmid containing P sigN2 +P sigN was used for competition experiments.As a control to verify that SigB was active, LK1231 (P trxA ) was used.The transcriptions were initiated with 30 nM RNAP holoenzyme (final concentration), allowed to proceed for 15 min at 37°C and stopped with equal volumes of formamide stop solution (95% formamide, 20 mM EDTA, pH 8.0).Samples were loaded onto 7 M urea-7% polyacrylamide gel and electrophoresis was performed.The gels were dried and scanned with Amersham Typhoon (Cytiva); visualization and analysis were made using the Quantity One software (Bio-Rad).

Promoter activity monitored by RT-qPCR
The strains DB466 (P zpdG -lacZ), DB475 (P flache -lacZ), DB541 (rpsA::TnlacJump), and DB542 (ptsG::TnlacJump) were grown to OD 600 ~1 in LB medium.Then, 1 mL of cells was diluted into 2 mL of fresh LB media and SigN production was either induced with 1 mM IPTG (final concentration) or the cells were left uninduced.The cells were allowed to grow for 2 h.Two and a half milliliters of cells were then withdrawn and treated with RNAp rotect Bacteria reagent (QIAGEN), pelleted and immediately frozen.RNA was isolated with RNeasy Mini Kit (QIAGEN), and recovery marker RNA (RM RNA) was added at the time of extraction to control for differences in degradation and pipetting errors during extraction.The RM RNA was a fragment of 16S rRNA from Mycobacterium smegmatis mc 2 155.It was prepared by in vitro transcription of DNA fragment obtained by PCR, using primers 1281 and 1282 (58).The 1281 primer contains the T7 RNAP promoter sequence.Finally, RNA was DNase treated according to manufacturers' instructions (TURBO DNA-free Kit, Ambion).Total RNA was reverse transcribed to cDNA with reverse transcriptase (SuperScript III Reverse Transcriptase, Invitrogen) using random hexamers, and this was followed by qPCR in a LightCycler 480 System (Roche Applied Science) containing LightCycler 480 SYBR Green I Master and 0.5 µM primers (each).RM cDNA was amplified with primers #2618 and #2619 and the test lacZ cDNA with primers 4915 and 4916.The final data were normalized to RM and the amount of cells (OD 600 ).Control experiments (qPCR without prior RT) yielded no signal.

FIG 1
FIG 1 Spontaneous suppressors of SigN-mediated death.(A) Optical density (OD 600 ) growth curve of a strain containing pBS32 and the ectopic SigN expression construct with a weakened RBS (DK1634) in the absence (open circles) and presence (closed circles) of IPTG.X-axis is the time of spectrophotometry after the addition of IPTG.(B) Colony images of ssd class I (DK7378), ssd class II (DK9591), ssd class III (DK6529), ∆pBS32 (DK451), WT (DK607), and abrB (DK5435) mutants.Scale bar is 2 mm.

FIG 2
FIG 2 Plasmid curing prevents death.(A) PCR analyses of representatives of each ssd class are shown.Three loci on the plasmid (sigN, zpbH, zpcJ) and one chromosomal locus (fliG) were amplified, the sizes of which are indicated by carets.The NEB 1 kb size standard included as a reference in the leftmost lane.Chromosomal DNA from the following strains was used as a template: WT (DK607), ssd class I (DK7378), ssd class II (DK9591), and ssd class III (DK6529).(B) Cartoon diagram of the plasmid curing assay.Briefly a chromosomally encoded β-galactosidase (lacZ) reporter is repressed by SinR expressed from the pBS32 plasmid causing colonies to look white when plated on media containing chromogenic substrate X-gal.If the plasmid is cured, SinR is lost and the reporter is de-repressed causing colonies to look blue when plated on media containing X-gal.Thick blue arrow is the lacZ gene, thin blue arrow is the lacZ transcript, and bent arrow is the SinR-repressed P eps promoter.Black arrows indicate activation and black T-bars indicate repression.(C) Blue/white colony frequency in a pBS32

FIG 3
FIG 3 Missense mutants that alter the SigN sequence produce levels of protein like the wild type.Western blot analysis of whole cell lysates detected with anti-SigA as a loading control (top panel) and anti-SigN (bottom panel) is shown.(A) Wild-type SigN protein (DK1634) or the indicated SigN alleles (DK9590, DK9591, DK9589) were expressed from an IPTG-inducible promoter with a weakened RBS "wk" in a strain containing pBS32.(B) Wild-type SigN protein was expressed from an IPTG-inducible promoter with either the native SigN RBS "n" (DK9067), or a weakened RBS "wk" in a strain lacking pBS32 (DK9069).

FIG 4
FIG 4 Expression of SigN from its native RBS arrests growth in the absence of pBS32.(A) Plate reader growth curves of strains containing an IPTG-inducible SigN with either native RBS (nRBS, DK9067) or weakened RBS (wkRBS, DK9069) grown in the presence (red) and absence (black) of 1 mM IPTG.Both strains lack the pBS32 plasmid.(B) Plate reader growth curves of strains containing IPTG-inducible SigN E55G (DK9210), SigN K83R (DK9211), or SigN I198T (DK9209) with native RBS grown in the presence (red) and absence (black) of 1 mM IPTG.(C) Plate reader growth curves of strains containing IPTG-inducible SigN R22G (DK9282), SigN L136P (DK9283), SigN D156G (DK9345), SigN N160S (DK9342), or SigN Y165V (DK9325) with native RBS grown in the presence (red) and absence (black) of 1 mM IPTG.Each strain was grown in triplicate and each growth curve is presented separately.

FIG 5
FIG 5 Relative affinities of B. subtilis SigA, SigN, and SigB for the RNAP core.Multiple-round transcriptions were carried out with increasing ratios of alternative sigma factors to SigA.The DNA template contained two promoters: P sigN2 (SigA-dependent; longer transcript) +P3 and P sigN3 SigN-dependent; shorter transcript).Representative primary data are shown above the graphs.Individual data points are averages from at least three biological replicates, the error bars indicate standard deviations of density scans.(A) Competition between SigA and SigN for the RNAP core.SigA (open circles) and SigN transcripts (closed circles) were detected.Transcription from P2 in the absence of SigN was set as 1 (first lane).(B) Competition between SigA and SigB for the RNAP core.Transcription from P sigN2 in the absence of SigB was set as 1 (first lane).

FIG 6
FIG 6 Mutation of AbrB suppresses SigN-mediated cell death.(A) Cartoon diagram of the abrB genetic neighborhood.Open arrow indicate genes, bent arrow indicates promoter.Black carets indicate the location of transposon insertions in abrB (DK3140 and DK3144).(B) Western blot of SigN accumulation when wkRBS SigN was induced immediately after IPTG addition (T 0 ) and 2 h afterwards (T 2 ) in an abrB mutant (DK6955) and an abrB mutant complemented with abrB ectopically expressed from its native promoter (abrB + ) (DK7132).The vegetative sigma factor SigA was also included as a control.(C) Western blot of SigN accumulation immediately after the indicated times following addition of the DNA damaging agent mitomycin C in otherwise wild type (DK607) and an abrB mutant (DK5435).The vegetative sigma factor SigA was also included as a control.(D) β-galactosidase assays of strains containing an IPTG-inducible wkRBS sigN gene and a P sigN -lacZ reporter grown in the absence (light gray bars) and presence (light gray bars) of 1 mM IPTG for 1 h.Ø indicates an otherwise wild-type background; ΔpBS32Δ indicates strains in which the pBS32 plasmid has been cured.The following strains were used to generate this panel: Ø ΔpBS32 (DB250), abrB ΔpBS32 (DB251), Ø (DB249), and abrB (DB261).Error bars are the standard deviations of three replicates.(E) β-galactosidase assays of strains containing a lacZ reporter fused to the promoter region indicated on the x-axis.Each strain contained either an abrB mutation (dark gray bars) or no additional modification (Ø, light gray bars).The following strains were used to generate this panel: P asiN -lacZ Ø (DB311) and abrB (DB307), P asiN1 -lacZ Ø (DB312) and abrB (DB308), P asiN2 -lacZ Ø (DB313) and abrB (DB309), P asiN3 -lacZ Ø (DB314) and abrB (DB310).Error bars are the standard deviations of three replicates.(F) β-galactosidase assays of ΔpBS32 otherwise wild type (Ø, light gray bars, DB259) or abrB mutant (dark gray bars, DB260) strains containing an IPTG-inducible wkRBS sigN EXT gene with extended downstream region and a P sigN -lacZ reporter grown in the presence of either 1 mM IPTG or 0.1 mM IPTG for 1 h.Error bars are the standard deviations of three replicates.

FIG 7
FIG 7 SigN fails to achieve positive feedback in late exponential phase.(A) OD 600 growth curves of DK1634 expressed as time after 1 mM IPTG addition.Lines colored according to the effect of SigN induction: black indicates toxicity, medium gray indicates partial growth inhibition, and light gray indicates no growth inhibition.(B) Western blot analysis probed with anti-SigA and anti-SigN antibodies taken after 2 h of SigN induction from the same experiment is shown in panel A. OD at induction given as 100× values to accommodate presentation and use the same grayscale coding as panel A.

FIG 8
FIG 8 SigN induction decreases expression of SigA-dependent genes.Double Y-axis column graphs for β-galactosidase assays of the indicated reporter (light gray) and relative mRNA levels of the lacZ transcript (dark gray) of the same strain.Each strain was deleted for pBS32 and contained an IPTG inducible nRBS sigN construct grown for 2 h either in the absence (−) or presence of 1 mM IPTG (+).The following lacZ reporters were tested: (A) the SigN-dependent ectopically integrated transcriptional reporter P zpdG -lacZ (DB467); (B) the SigA/SigD-dependent ectopically integrated transcriptional reporter P flache -lacZ (DB475); (C) the SigA-dependent natively integrated rpsAΩTnFLXlacZ (DB541); and (D) the SigA-dependent natively integrated ptsGΩTnFLXlacZ (DB542).For B-galactosidase assays, error bars are the standard deviation of three replicates; for RT-qPCR measurements, error bars are the standard error of three replicates.RT-qPCR values were normalized to RNA transcribed from the P flache promoter in the absence of IPTG (set as 1).

FIG 9
FIG 9 Model of SigN regulation and growth inhibition.(A) Cartoon diagram of the sigN genetic neighborhood.Open arrow indicate genes, bent arrow indicates promoter.Heavy arrows indicate activation.Heavy T-bars indicate repression.(B) Alphafold predicted structure of the RNA polymerase core from Bacillus subtilis (gray) docked with SigN (gold).Residues altered by ssd class II mutations that abolished pBS32-dependent toxicity are space filled and colored teal.Residues altered by ssd class IIb mutations that abolished both pBS32-dependent and pBS32-dependent toxicity are space filled and colored magenta.The sequences for RNA polymerase β, RNA polymerase β′, 2 RNA polymerase α, and SigN were run with the AlphaFold2 Multimer software version 2.3.1.The highest scoring model is shown above.Protein sequences for RNAP were retrieved from Accession #CP020102.RNA polymerase β (Protein ID: AQZ89020.1),RNA polymerase β′ (Protein ID: AQZ89021.1),RNA polymerase α (Protein ID: AQZ89056.1),and SigN sequence (Protein ID: AQZ93228.1)were retrieved from Accession # CP020103.The PDB file used to generate the panel is included in supplemental material.

TABLE 1
Suppressors of SigN-mediated death (ssd)

TABLE 2 Strains
(Continued on next page)