Global Chromosome Topology and the Two-Component Systems in Concerted Manner Regulate Transcription in Streptomyces

Abstract

ABSTRACT

Bacterial gene expression is controlled at multiple levels, with chromosome supercoiling being one of the most global regulators. Global DNA supercoiling is maintained by the orchestrated action of topoisomerases. In Streptomyces, mycelial soil bacteria with a complex life cycle, topoisomerase I depletion led to elevated chromosome supercoiling, changed expression of a significant fraction of genes, delayed growth, and blocked sporulation. To identify supercoiling-induced sporulation regulators, we searched for Streptomyces coelicolor transposon mutants that were able to restore sporulation despite high chromosome supercoiling. We established that transposon insertion in genes encoding a novel two-component system named SatKR reversed the sporulation blockage resulting from topoisomerase I depletion. Transposition in satKR abolished the transcriptional induction of the genes within the so-called supercoiling-hypersensitive cluster (SHC). Moreover, we found that activated SatR also induced the same set of SHC genes under normal supercoiling conditions. We determined that the expression of genes in this region impacted S. coelicolor growth and sporulation. Interestingly, among the associated products is another two-component system (SitKR), indicating the potential for cascading regulatory effects driven by the SatKR and SitKR two-component systems. Thus, we demonstrated the concerted activity of chromosome supercoiling and a hierarchical two-component signaling system that impacts gene activity governing Streptomyces growth and sporulation.

IMPORTANCE Streptomyces microbes, soil bacteria with complex life cycle, are the producers of a broad range of biologically active compounds (e.g., antibiotics). Streptomyces bacteria respond to various environmental signals using a complex transcriptional regulation mechanism. Understanding regulation of their gene expression is crucial for Streptomyces application as industrial organisms. Here, on the basis of the results of extensive transcriptomics analyses, we describe the concerted gene regulation by global DNA supercoiling and novel two-component system. Our data indicate that regulated genes encode growth and sporulation regulators. Thus, we demonstrate that Streptomyces bacteria link the global regulatory strategies to adjust life cycle to unfavorable conditions.

INTRODUCTION

In all organisms, gene expression is precisely controlled, primarily at the level of transcription initiation. The main transcriptional regulatory factors include promoter DNA sequences and trans-acting transcriptional regulators. Bacterial genomes encode numerous transcriptional regulators, among which the key players are DNA-binding proteins like sigma factors, which recruit RNA polymerase (RNAP) to promoters, and transcription factors (TFs), which may act as repressors or activators by affecting RNAP binding (1, 2). Notably, the binding of regulatory proteins in bacteria is in turn controlled by systems that adjust transcription in response to external and internal cell conditions. Examples include changes in chromosome topology to modify promoter accessibility to regulatory factors, as well as the modification of regulatory protein activity itself (3).

The activity of specific sigma factors or regulatory proteins can be modulated through partner protein or ligand binding, as well as through proteolysis or covalent modifications like phosphorylation. The importance of phosphorylation has been well established for regulators that are a part of two-component systems (TCSs). Canonical TCSs consist of a transmembrane sensor histidine kinase (HK) and cytoplasmic response regulator (RR), which detects environmental signals and triggers intracellular responses, respectively (4). Upon signal sensing, the kinase in classical TCSs undergoes autophosphorylation and subsequently transfers the phosphate moiety to its cognate response regulator, which promotes DNA binding and transcriptional control of its target genes. While the regulatory targets and biological functions of many regulatory proteins have been well described (5), a plethora of them remain unexplored.

One of the crucial factors influencing regulatory protein binding to DNA is chromosome topology, determined by chromosome supercoiling and nucleoid-associated proteins (NAPs). The global chromosome topology of bacteria depends on growth phase and environmental conditions and adjusts transcription in response to both extra- and intracellular conditions. Overall, bacterial chromosome supercoiling is controlled by enzymes called topoisomerases, mainly the TopA type I topoisomerase, which relaxes DNA (removes negative supercoils), and gyrase, which in contrast introduces negative supercoils (6). Inhibiting topoisomerase activity or altering topoisomerase levels leads to changes in chromosome topology and affects DNA transactions, including replication and transcription. To date, studies on various bacterial species (Streptococcus pneumoniae, Haemophilus influenzae, Escherichia coli, Salmonella enterica, and Streptomyces coelicolor) have shown that disturbances in the topological balance affect the transcription of a significant fraction of so-called supercoiling-sensitive genes (7–13).

The binding of NAPs also depends on chromosome topology, with NAPs in turn affecting the binding of other transcription factors (3, 14–18). However, little is known about the cross talk between chromosome supercoiling and other regulatory systems controlling gene transcription, particularly in response to changes in environmental conditions.

Soil-dwelling bacteria such as Streptomyces frequently encounter environmental stress. Streptomyces adaptations to the soil environment include their mycelial growth and complex developmental life cycle, which encompasses both spore formation and exploratory growth (19, 20). Vegetatively growing Streptomyces cells elongate and branch to generate a network of multicellular hyphae. In response to environmental stimuli, particularly nutrient depletion, sporulation is triggered. Sporulation starts with raising aerial hyphae, within which spore chains subsequently develop. The conversion of multigenomic hyphal cells to chains of unigenomic spores requires chromosome condensation and segregation, accompanied by synchronous septation (19, 21). The progression of the Streptomyces life cycle is governed by a set of well-described regulatory proteins (such as those encoded by the whi or bld genes) (22, 23); however, numerous reports indicate an abundance of less-studied regulators and other proteins that also contribute to sporulation regulation (24–29).

Streptomyces bacteria use a repertoire of biologically active secondary metabolites to thrive in their environmental niche, including numerous antibiotics (approximately 60% of natural antibiotics are produced by Streptomyces), immunosuppressants, and cytostatics (30). The production of secondary metabolites remains under the control of complex regulatory networks and is coordinated with developmental programs (31–34). As a free-living organism, Streptomyces responds to highly variable conditions using a large number of transcriptional regulators, many of which remain uncharacterized (31). The number of transcription factors encoded by streptomycete genomes ranges from 471 to 1,101, and among these, depending on the species, there are 315 to 691 transcriptional regulators and 31 to 76 sigma factors (26). Compared to other bacterial genera, Streptomyces genomes also encode numerous TCSs, the number of which varies depending on the species, ranging from 59 and 117, alongside 13 to 21 orphan response regulators and 17 to 39 unpaired/uncharacterized sensor kinases (35–37).

As in other bacteria, chromosome topology plays a critical role in the regulation of gene expression in Streptomyces. In contrast to many bacteria, the model Streptomyces species S. coelicolor possesses only one type I topoisomerase, TopA, which is essential for viability (38). TopA depletion in Streptomyces results in increased DNA supercoiling and altered gene expression, leading to severe growth retardation and sporulation blockage (38, 39). Moreover, disturbances in global DNA supercoiling affect the transcription of up to 7% of Streptomyces genes (12). Numerous supercoiling-sensitive genes are grouped into discrete clusters, with one cluster in particular, named SHC (supercoiling-hypersensitive cluster), exhibiting extreme DNA supercoiling sensitivity. This region encodes many proteins of unassigned function but also appears to include a two-component system, anti-sigma factors and probable transcriptional regulators. Interestingly, most of the SHC genes are poorly transcribed under standard conditions but are upregulated in response to increased DNA supercoiling (12). Having established that altered DNA supercoiling significantly impacts transcription in S. coelicolor, we predicted that altered gene expression may contribute to the sporulation inhibition observed for the TopA-depleted strain.

To identify the genes responsible for sporulation and growth inhibition under high supercoiling conditions, we performed random transposon mutagenesis of the TopA-depleted S. coelicolor strain and screened for strains with mutations that suppressed the sporulation blockage associated with high DNA supercoiling. We found that mutations in genes encoding a two-component system named SatKR (SCO3390-89) led to altered transcription of the SHC cluster. We established that the activated response regulator SatR (SCO3389) inhibited growth and sporulation by inducing transcription of SHC genes independently of high DNA supercoiling. Moreover, we confirmed that mutations within SHC prevented the activation of genes within this region and restored growth and sporulation to the TopA-depleted strain. Thus, our results reveal a unique interplay between the two-component system SatKR and chromosome supercoiling in regulating SHC gene expression, with the SHC products subsequently impacting S. coelicolor growth and sporulation.

RESULTS

Screening for suppressors of supercoiling-induced sporulation blockage.

TopA is the only type I S. coelicolor topoisomerase and consequently is essential for viability. Its depletion in the TopA-controlled strain (PS04, in which the topA gene expression is under the control of the thiostrepton-inducible promoter tipA, allowing for an up to 20-fold depletion of TopA levels) leads to increased negative DNA supercoiling (38). Elevated negative DNA supercoiling in turn results in changes in global gene expression and affects the growth rate, sporulation, and secondary metabolism of S. coelicolor (12, 38). During differentiation of wild-type S. coelicolor, white sporogenic (aerial) hyphae mature into chains of gray spores; in contrast, the development of a TopA-depleted strain is inhibited at the aerial hyphal stage, resulting in a “white colony phenotype.” We speculated that inhibition of aerial hypha maturation may result from changes in the expression of supercoiling-sensitive genes encoding sporulation regulators (12). To identify any such sporulation regulators, we searched for transposon mutations that were able to suppress the TopA depletion phenotype and restore sporulation (gray colonies). To ensure that the transposon insertion frequency was sufficient to cover all 7,825 predicted S. coelicolor genes (40), we aimed to obtain a mutant library containing approximately 16,000 clones. Having obtained the representative transposon library (PS04-Tnlib), we searched for mutants that formed gray colonies under TopA-depleted conditions. We identified seven transposants exhibiting this phenotype, and among them, one transposant, termed MGHM5, additionally exhibited a partially restored growth rate upon TopA depletion (with effective depletion being confirmed by Western blotting and reverse transcription-quantitative PCR [RT-qPCR]; see Fig. S1 in the supplemental material), both on solid medium and in liquid medium, compared with its TopA-depleted parental strain (Fig. 1A and B). Unlike the TopA-depleted parental strain, which overproduced blue actinorhodin, the TopA-depleted transposon strain did not produce either of the pigmented antibiotics made by S. coelicolor (blue actinorhodin or red undecylprodigiosin) (Fig. 1A). Microscopic analysis of spores produced by the TopA-depleted transposon strain confirmed the presence of spore chains, although these were detectable only after prolonged incubation (approximately 53 h compared to the 48 h needed to sporulate in the wild-type strain, Fig. 1C); spore chains could not be detected in the TopA-depleted parental strain (Fig. S2). Interestingly, the spores produced by the TopA-depleted transposon strain were of various sizes compared with the wild-type strain and the parental strain in which the TopA level was restored to that of the wild-type strain (PS04, 1 μg/ml thiostrepton). Moreover, spores produced by the TopA-depleted transposon strain were highly sensitive to 5% sodium dodecyl sulfate (SDS): only 2.5% survived 1 h of SDS exposure compared with the 76% spore survival of the same strain with restored TopA levels (Fig. 1D).

FIG 1

Phenotype of the TopA-depleted transposon strain MGHM5 (Tn TopA↓). (A) Growth of the TopA-depleted transposon strain (Tn TopA↓) on solid MS agar (top panel) and antibiotic production in R2 liquid medium (bottom panel) compared with wild-type (WT) and TopA-depleted PS04 (TopA↓) strains. The cultures were grown for 72 h. (B) Growth curves of the TopA-depleted transposon strain (Tn TopA↓) in liquid 79 medium compared with WT and TopA-depleted PS04 (TopA↓) strain growth. The growth rate was measured in triplicate using a Bioscreen C instrument for 48 h. (C) Spores produced by the TopA-depleted transposon strain (Tn TopA↓). (Top) Phase-contrast microscopy images demonstrating representative spore chains of the TopA-depleted transposon strain (Tn TopA↓) and its parental strain PS04 with restored TopA levels (induced with 1 μg/ml thiostrepton) and the wild-type strain after 53 h of growth in MM minimal medium (with 1% mannitol). (Bottom) Spore size distribution. Asterisks indicate the significance of the P value (*, P ≤ 0.05; ***, P ≤ 0.001) when comparing mean spore sizes. (D) Viability of spores of the TopA-depleted transposon strain (Tn TopA↓) after SDS treatment compared with the wild-type strain, PS04 with restored TopA levels, and transposon strain (Tn) with TopA level restored. Spores were collected and incubated for 1 h in 5% SDS at room temperature. The viability percentage was calculated as a ratio of the colony number grown from spores treated and untreated with disrupting agent. (E) DNA supercoiling of the reporter plasmid pWHM3Hyg or pWHM3Spec isolated from the wild-type strain derivative MS10 (WT), TopA-depleted strain derivative MS11 (TopA↓) and TopA-depleted transposon strain derivative (MGHM5_RP, Tn TopA↓) cultured for 24 h in liquid 79 medium. The distribution of the reporter plasmid topoisomers was analyzed by agarose gel electrophoresis. Black vertical lines indicate the most abundant topoisomers.

Next, we tested whether the increased growth rate and sporulation of the TopA-depleted transposon strain resulted from a restoration of wild-type levels of chromosome supercoiling. To achieve this goal, we isolated a supercoiling reporter plasmid (pWHM3Hyg) from a derivative of the TopA-depleted transposon strain (MGHM5_RP) and established that its negative supercoiling level was similar to that of plasmid isolated from the parental TopA-depleted strain derivative, indicating similar chromosome supercoiling (Fig. 1E).

Thus, we successfully identified a transposon mutant in which sporulation and growth rate defects of the parental TopA-depleted strain were restored to wild-type levels, and the observed phenotypic effect did not result from restored negative supercoiling.

Transposon insertion in two-component system-encoding genes influences the expression of a supercoiling-sensitive cluster.

We mapped the transposon insertion loci in the MGHM5 strain (by sequencing rescue plasmids and genome sequencing), identifying two transposons: one in sco3390 and one in sco2474. In both cases, the orientation of the aadA(1) gene within the transposon cassette was the same as that of the disrupted gene. In the first locus (sco3390), the transposon cassette was inserted 292 nucleotides downstream of the start codon (Fig. 2). The sco3390 gene (1,206-bp length) was annotated as encoding a putative two-component system kinase, while the genes downstream of this, in a presumable operon, were annotated as encoding a probable cognate response regulator (sco3389) and TrmB-like protein (sco3388) (41, 42). The second transposition site was located 1,020 nucleotides downstream of the start codon of the sco2474 gene (1,644 bp in length), which encodes a putative secreted metalloproteinase (Fig. S3A). Importantly, previously performed transcriptome sequencing (RNA-seq) analysis (12) showed that while the sco3388-3390 genes were transcribed during S. coelicolor vegetative growth in a supercoiling-insensitive manner (Fig. 3A, left panel), the sco2474 gene was not expressed during vegetative growth, either in normal or in high DNA supercoiling conditions (i.e., in the TopA-depleted PS04 strain) (Fig. S3B). Moreover, sco2475, located downstream of the disrupted gene, was transcribed in the transposon strain, and its expression was not changed due to transposition, suggesting a lack of polar effects associated with this transposon insertion (Fig. S3B). Since sco2474 was not expressed under any tested conditions, and we knew that transposition in MGHM5 affected not only sporulation but also the vegetative growth rate, we excluded the disruption of sco2474 as a reason for the restored growth of the TopA-depleted transposon strain and focused our attention on the sco3390 gene/operon.

FIG 2

Position of the Himar1 transposon insertion site in the MGHM5 strain. The green triangle shows the insertion site within the sco3390 gene with the orientation of the inserted aadA(1) gene. ori, origin of replication.

FIG 3

Transcriptional changes in the TopA-depleted transposon mutant MGHM5 (Tn TopA↓) compared with its TopA-depleted parental strain (PS04) and wild-type strain. (A) Normalized transcription level of sco3390-sco3388 genes in the TopA-depleted transposon strain (Tn TopA↓), transposon strain with TopA level restored (Tn TopA restored), its parental TopA-depleted PS04 strain (TopA↓), and the wild type (WT), based on RNA-seq (left) and the relative transcription level analyzed by RT-qPCR (right). Asterisks in the RNA-seq analysis indicate statistical significance of the q value and log₂ fold (q ≤ 0.01 and 1.5 < log₂ fold < −1.5). (B) Volcano plots based on the RNA-seq experiments showing changes in global gene expression in the TopA-depleted transposon strain (Tn TopA↓ strain) compared with its parental TopA-depleted PS04 strain (TopA↓) (left), as well as genes affected by supercoiling (changes between the wild-type strain and TopA-depleted PS04 strain [TopA↓]) (right panel, data obtained earlier [12]). Significantly altered transcripts (q ≤ 0.01 and 1.5 < log₂ fold < −1.5) are depicted in red, SHC genes are shown in blue, and genes with unchanged transcription are marked in green. NS, not significant. (C) Transcriptional profile of the SHC region in the wild-type strain (blue), TopA-depleted transposon mutant (Tn TopA↓; green) and TopA-depleted parental PS04 strain (TopA↓; orange). SHC genes are depicted in yellow. (D) Silencing of SHC gene transcription in the TopA-depleted transposon strain (Tn TopA↓) compared with its parental TopA-depleted PS04 strain (TopA↓) and wild-type strain analyzed by RT-qPCR. Primers for amplification of the two genes, the first and last genes of the SHC cluster (sco4667 and sco4699), were used.

Transposon insertion in the kinase-encoding sco3390 gene was expected to affect the level of expression of the downstream response regulator-encoding sco3389 gene. Moreover, since the transposon insertion was expected to inactivate the kinase-encoding gene, we further predicted that this mutation would modify the phosphorylation state and activity of its cognate regulator. Transcriptional analysis (RNA-seq, confirmed by RT-qPCR analysis) performed on RNA isolated from liquid cultures of the TopA-depleted MGHM5 transposon strain showed significant sco3389 downregulation compared with the parental TopA-depleted strain (Fig. 3A and Fig. S4). Additionally, we established that the expression of 100 genes changed significantly in the TopA-depleted MGHM5 transposon strain compared to its parental TopA-depleted strain. The majority of differentially expressed genes (72 genes) were downregulated in the transposon mutant, with only 28 genes being upregulated compared to the parental TopA-depleted strain (Data Set S1, Tab 1; Fig. 3B, left panel). Surprisingly, 30 of 72 downregulated genes were concentrated in one region of the chromosome: the supercoiling-hypersensitive cluster (SHC) encompassing 34 genes (sco4667-sco4700). The majority of the SHC genes (26 of 34 genes) have been previously shown to not be transcribed or transcribed at very low levels under optimal growth conditions but highly induced upon TopA depletion (12) (Fig. 3B, right panel, and Fig. 3C). We confirmed the decreased transcription of SHC genes in the TopA-depleted transposon strain compared with the TopA-depleted parental strain using RT-PCR with primers specific for the first (sco4667) and penultimate gene (sco4699) of the SHC cluster (Fig. 3D). According to the RT-qPCR results, SHC gene expression was reduced despite TopA depletion, although not to the extent observed in the RNA-seq analysis, which might be due to the different approaches to measure the transcript levels (over the whole gene length versus in a certain position).

In summary, we determined that our TopA-depleted strain carrying a transposon integrated into sco3390 exhibited an altered transcriptional landscape compared with its parent TopA-depleted strain, with reverted induction of the supercoiling-sensitive SHC genes. This implicated the sco3389-sco3390-encoding two-component system in controlling the transcription of the supercoiling-sensitive cluster and led us to term these gene products SatKR (for SHC activity controlling two-component system kinase/regulator).

Changes in SatKR levels affect Streptomyces growth and sporulation.

To confirm that the SatKR two-component system regulated growth and sporulation in S. coelicolor in coordination with chromosome supercoiling, we analyzed the phenotypic effects of satKR gene deletion and satR overexpression in different genetic backgrounds.

Unexpectedly, deleting satKR did not affect vegetative growth in liquid medium or differentiation on solid medium in either the wild-type or TopA-depleted backgrounds (MGM12 and MGP12 strains, respectively) compared to parental strains (Fig. 4A). Sporulation of the ΔsatKR mutant was also unaffected in the wild-type background, while in the TopA-depleted background, sporulation was not restored, based on plate analyses, and was confirmed microscopically (Fig. S5). Thus, eliminating both components of the SatKR system had different phenotypic effects than the inactivation of the SatK kinase and lowering of the SatR transcript levels in transposon strains.

FIG 4

Phenotypes of satKR mutant strains. (A) The phenotype of the MGM12 strain with satKR deletion in the wild-type background (ΔsatKR) and the MGP12 strain carrying the satKR deletion in the TopA-controlled background (TopA-depleted ΔsatKR TopA↓ and TopA restored with the addition of 0.5 μg/ml thiostrepton inducer [ΔsatKR TopA restored]). (Left) Scheme of the mutant strain genotypes. (Right, top) Growth and differentiation of strains carrying the ΔsatKR deletion in the wild-type and TopA-depleted background on MS agar after 48 and 72 h of growth, respectively. (Bottom) Growth curves in liquid 79 broth obtained using a Bioscreen C instrument. Measurements were performed in triplicate every 20 min for 42 h. (B) The phenotype of strain MGM11 overexpressing satR in the wild-type background (satR↑) and strain MGP11 overexpressing satR in the TopA-controlled background (TopA-depleted: satR↑TopA↓ and TopA restored with the addition of 0.5 μg/ml thiostrepton inducer: satR↑ TopA restored). (Left) Scheme of the mutant strain genotypes. (Right, top) Growth and differentiation of the analyzed strains on MS agar after 48 or 72 h of growth, respectively. (Bottom) Growth curves of the analyzed strain in liquid 79 broth obtained using a Bioscreen C instrument. Measurements were performed in triplicate every 20 min for 42 h. (C) The phenotype of the MGM14 strain overexpressing satR in the satK deletion background (satR↑ΔsatK) and the MGP14 strain overexpressing satR in the TopA-controlled satK deletion background (TopA depleted: satR↑ΔsatK TopA↓ or TopA restored with the addition of 0.5 μg/ml thiostrepton inducer: satR↑ΔsatK TopA restored). (Left) Scheme of the mutant strain genotypes. (Right, top) Growth of analyzed strains on MS agar for 48 or 72 h. (Bottom) Growth curves of the analyzed strain in liquid 79 broth obtained using a Bioscreen C instrument. Measurements were performed in triplicate every 20 min for 42 h.

To further explore the role of SatR, we overexpressed satR in the presence of its cognate kinase (using an additional copy of the satR gene under the control of a strong constitutive p_ermE promoter, with overexpression being confirmed using RNA-seq and RT-qPCR analysis [Fig. S6]) in both the wild-type strain and the TopA-controlled strain (MGM11 and MGP11 strains, respectively; Fig. 4B). Elevated satR transcription slightly retarded vegetative growth in liquid cultures and delayed differentiation on solid medium in both the wild-type and TopA-depleted backgrounds compared to the parental strains (Fig. 4B and Fig. S5). Moreover, the satR-overexpressing strain with restored TopA levels exhibited somewhat inhibited growth compared with the TopA-restored parental strain (Fig. 4B). These observations suggested that overexpression of satR in the presence of the cognate kinase SatK impaired cell growth.

To assess the importance of its cognate kinase on SatR activity, we inactivated satK by frameshift mutation in the wild-type and TopA-controlled backgrounds and overexpressed satR in the obtained strains (MGM14 and MGP14 strains, respectively). In contrast to satR overexpression in the presence of kinase, elevated satR transcript levels in the absence of cognate kinase did not affect the growth rate and development either in liquid medium or on solid medium (Fig. 4C) compared with the parental strains. These observations suggested that SatK was crucial for SatR activity and its influence on growth and sporulation.

In summary, overexpression of satR in the presence of its intact cognate kinase gene (satK) led to growth inhibition and delayed sporulation; these effects were lost when satK was inactivated.

The SatKR regulon includes SHC genes.

To better understand the biological function of SatKR, we investigated how modulating satKR gene expression in different genetic backgrounds affected global gene transcription.

Transcriptional analysis showed that deletion of satKR in the wild-type background (MGM12) altered the expression of 126 genes (1.5% of all S. coelicolor genes) (see Data Set S1, Tab 2, in the supplemental material). Within these genes, there were nine SHC genes (seven of which were supercoiling sensitive), which we observed to be further downregulated in the satKR mutant (despite their very low expression in the wild type), indicating that the SatKR system also influenced the expression of the SHC genes under normal supercoiling conditions. Changes in gene expression were also observed when satKR was deleted in the TopA-depleted background (strain MGP12), where 67 genes, or 0.76% of all S. coelicolor genes, had significantly altered expression (Data Set S1, Tab 3 and Tab 4). However, the expression of SHC genes activated by high DNA supercoiling in this mutant was unaffected, indicating that the presence of SatR and absence of SatK (as in the transposon mutant) are required for this inhibition.

Next, we established the impact of satR overexpression on transcription in the wild-type and TopA-depleted backgrounds (MGM11 and MGP11 strains, respectively). We determined that a high level of SatR in the presence of SatK under normal supercoiling conditions significantly affected the expression of 452 genes (approximately 5% of all genes) (Data Set S1, Tab 5). Among these genes, 351 (nearly 78%) were activated due to satR upregulation, while the expression of 101 genes was reduced (Fig. 5A). A similar effect was noted in the TopA depletion background (Data Set S1, Tab 6, 409 genes with altered transcription); however, only 55 genes were equally changed in both wild-type and TopA-depleted conditions. satR overexpression in the presence of SatK and normal DNA supercoiling (MGM11) was also found to activate the transcription of 16 genes from the SHC region (Fig. 5B). We confirmed the activation of sco4667 (first SHC gene) and sco4699 (penultimate SHC gene) using RT-qPCR (Fig. 5C). In the TopA-depleted background (MGP11 strain), satR overexpression did not significantly affect SHC transcription (with one exception being sco4677), as under high DNA supercoiling conditions, the transcription of these genes was already highly induced.

FIG 5

Global transcriptome analysis of strains overexpressing satR in the presence of SatK (MGM11) and in the satK background (MGM14). (A) Volcano plot based on RNA-seq experiments showing changes in global gene expression between the wild-type (WT) and satR-overexpressing strain MGM11 (satR↑). Significantly altered transcripts (q ≤ 0.01 and 1.5 < log₂ fold < −1.5) are depicted in red, SHC genes are shown in blue, and nonsignificant (NS) changes are marked in green. (B) Transcriptional profile of the SHC region in the MGM11 strain overexpressing satR (satR↑; yellow) compared with its wild-type parental strain (WT; blue) and TopA-depleted PS04 strain (TopA↓; orange). (C) RT-qPCR analysis of sco4667 and sco4699 gene transcription levels in the WT, TopA-depleted PS04 strain (TopA↓), TopA-depleted MGP12 strain carrying a deletion of satKR (ΔsatKR TopA↓), MGM11 strain overexpressing satR in the wild-type background (satR↑), MGP11 strain overexpressing satR in the TopA-depleted background (satR↑ TopA↓), and MGP14 strain overexpressing satR in the TopA-depleted satK deletion (satR↑ ΔsatK TopA↓) background. (D) Volcano plot based on RNA-seq experiments showing changes in global gene expression between strain MGM14 overexpressing satR in the absence of SatK (satR↑ ΔsatK) and strain MGM11 overexpressing SatR in the wild-type background (satR↑). Significantly altered transcripts (q ≤ 0.01 and 1.5 < log₂ fold < −1.5) are depicted in red, SHC genes are shown in blue, and nonsignificant changes are marked in green.

While overexpression of SatR in the presence of the SatK kinase activated numerous genes, including SHC even under normal supercoiling conditions, its overexpression in the absence of kinase resulted in much less pronounced changes in transcription both in the wild-type and increased supercoiling backgrounds (MGM14 and MGP14, respectively) (Data Set S1, Tab 7 and Tab 8). In conditions of normal supercoiling, overexpression of satR in a satK mutant background led to transcriptional activation of only 15 genes and repression of 17 genes compared with the wild type. Similarly, overexpressing satR in the satK mutant, TopA-depleted background led to changes in the expression of 19 genes (14 activated and 5 repressed) compared with the TopA-depleted strain. Furthermore, when comparing the effect of satR overexpression in the wild-type and satK deletion backgrounds (MGP11 and MGP14, respectively), we found that all SHC genes activated by satR overexpression in the presence of SatK were silenced when the kinase was absent (Fig. 5D, confirmed by RT-qPCR [Fig. 5C]). This evidence strongly suggested that SatK-phosphorylated SatR functions to activate—either directly or indirectly—the SHC genes.

In summary, we confirmed that the SatKR system contributed to the activation of SHC gene transcription. The results of our transcriptional analyses suggest that SatR functions predominantly as a transcriptional activator, and its activity requires the presence of SatK. Interestingly, our data suggested that both SatKR and supercoiling were sufficient to independently activate SHC transcription. Moreover, the activation of SHC genes either by supercoiling or SatR activity was correlated with slower growth and sporulation inhibition.

A single transposon insertion in the SHC induces sporulation under TopA depletion.

Our findings suggested that increased SHC expression may be contributing to sporulation defects, given the reduced sporulation observed in both a TopA-depleted strain (high supercoiling, high SHC expression) and an satR overexpression strain (high SHC expression). Consistent with this possibility was our finding that one of our original transposon mutant strains that restored sporulation to a TopA-depleted strain had an insertion in the penultimate SHC gene sco4699 (strain MGHM14). sco4699 encodes a homologue of the Rhs protein from E. coli, a secreted toxin that mediates cellular competition (43). In this mutant strain, the Himar1 transposon was inserted 311 nucleotides downstream of the sco4699 start codon (Fig. 6A).

FIG 6

Phenotype of the SHC transposon strain MGHM14. (A) Himar1 transposon insertion site of the transposon in MGHM14 (Tn-SHC). The green triangle shows the insertion site with the orientation of the inserted aadA(1) gene. (B) Growth and differentiation of the TopA-depleted transposon MGHM14 (Tn-SHC TopA↓) compared with the wild-type (WT) and TopA-depleted PS04 strain (TopA↓) on MS agar after 72 h of growth. (C) Growth curve of TopA-depleted MGHM14 (Tn-SHC TopA↓) compared with the wild-type (WT) and TopA-depleted strain PS04 (TopA↓) in liquid 79 broth obtained using a Bioscreen C instrument. Measurements were performed in triplicate every 20 min for 42 h. (D) Spore size distribution in the TopA-depleted MGHM14 strain (Tn-SHC TopA↓), its parental strain PS04 with TopA levels restored by induction with 1 μg/ml thiostrepton (TopA restored) and the wild-type (WT) strain. Asterisks indicate the significance of the P value (*, P ≤ 0.05; ****, P ≤ 0.0001) when comparing mean spore sizes. (E) RT-qPCR analysis of sco4667 transcription in TopA-depleted MGHM14 (Tn-SHC TopA↓) and TopA-depleted sco4699 mutant strains (sco4699::hyg TopA↓) compared with the wild-type (WT) and TopA-depleted PS04 strains (TopA↓).

As was seen for the satKR-associated transposon mutant, the growth rate of the TopA-depleted MGHM14 strain (where TopA depletion was confirmed by Western blot analysis [Fig. S7]) was partially restored, both on solid medium and in liquid medium, compared with its TopA-depleted parental strain (Fig. 6B and C). We observed abundant spore chains in the TopA-depleted SHC transposon mutant, while in its parental TopA-depleted strain (PS04), no spore chains could be detected (Fig. 6B). Moreover, similar to what was observed when satKR was disrupted by transposon insertion (in the MGHM5 strain), transposon integration into sco4699 also resulted in formation of spores of varied sizes (Fig. 6D).

Since sco4699 product supposedly does not act as transcriptional regulator, we wondered whether the supercoiling-dependent transcriptional regulation of the SHC region might be modified by transposon insertion in sco4699, providing a rationale for the restored sporulation. Thus, we used RT-qPCR to test the transcription level of the first SHC gene (sco4667) in this mutant background. We found that transposon insertion in sco4699 silenced expression of the cluster even when TopA was depleted (Fig. 6E), similarly to what had been observed for the satKR mutant.

Next, to confirm that the downregulation of the first gene of the SHC resulted from transposon insertion in the last gene of the cluster, we reconstructed the SHC transposon strain by inserting a hygromycin resistance cassette in sco4699 (at the transposon insertion site) in the TopA-controlled background (MGP20 strain). RT-qPCR analysis confirmed that transcription of the first gene within the SHC region (sco4667) was silenced in the TopA-depleted strain (Fig. 6E). This analysis suggested that supercoiling-dependent activation of the SHC genes might be disrupted by insertion in a distant region of the gene cluster.

In summary, the mutation of sco4699 within the SHC region inhibited cluster activation by supercoiling. Moreover, mutations in sco4699 reinforced the idea that silencing of the SHC under conditions of high supercoiling restores sporulation.

Deletion of SHC genes (sco4667-4668) encoding a two-component system SitKR restores sporulation in conditions of high supercoiling.

To further probe the contribution of the SHC genes to the inhibition of sporulation under conditions of high chromosome supercoiling, we sought to test whether specific genes within the SHC region might be responsible. As the first gene of the SHC region (sco4667) was annotated as encoding a kinase of another putative two-component system and formed the operon with the sco4668 gene, encoding its cognate response regulator, we deleted sco4667-4668 together in the wild-type and TopA-controlled backgrounds (MB01 and MB02 strains, respectively).

We found that the sco4667-4668 deletion—like the satK transposon insertion and mutation of sco4699—partially restored the growth of the TopA-depleted strain in liquid cultures compared with its TopA-depleted parental strain, while having no effect on growth in the wild-type background (Fig. 7A). Sporulation of the sco4667-4668 mutant strain in the TopA-depleted background was also restored; this was confirmed microscopically but was observed only after prolonged incubation (120 h of growth on MS agar) (Fig. 7B). As before, spores of the TopA-depleted sco4667-4668 mutant also exhibited a diversity in size, with a greater mean length than that of the wild type (Fig. 7C). Given the ability of this novel two-component system encoded within the SHC to influence growth and sporulation under conditions of high supercoiling, it was named SitKR (sporulation-inhibiting two-component system kinase and regulator).

FIG 7

Phenotype of Δsco4667-4668 mutants in wild-type (MB01) and TopA-depleted backgrounds (MB02). (A) Growth curves of MB01 (Δsco4667-4668) and the TopA-depleted MB02 strain (Δsco4667-4668 TopA↓) compared with the wild-type (WT) and TopA-depleted strain PS04 (TopA↓) in liquid 79 broth obtained using a Bioscreen C instrument. Measurements were performed every 20 min for 42 h. (B) Growth and differentiation of the MB01 (Δsco4667-4668) and TopA-depleted MB02 (Δsco4667-4668 TopA↓) strains, respectively, compared with the wild-type (WT) and TopA-depleted strain PS04 (TopA↓) after 120 h of growth on MS agar. A microscopic image of the spore chain of the TopA MB02 strain (Δsco4667-4668 TopA↓) after 48 h of growth in MM minimal medium (with 1% mannitol) is shown to the right of panel B. (C) Spore size distribution in TopA-depleted MB02 (Δsco4667-4668 TopA↓) compared with the parental PS04 strain with restored TopA levels induced with 1 μg/ml thiostrepton (TopA restored) and with wild type (WT). Asterisks indicate the significance of the P value (*, P ≤ 0.05; ****, P ≤ 0.0001) when comparing mean spore sizes. (D) RT-qPCR analysis of the transcription of sco4699 located at the end of the SHC region in the TopA-depleted MB02 strain (Δsco4667-4668 TopA↓) compared with the TopA-depleted PS04 (TopA↓) and wild-type (WT) strain. RNA was isolated from a 24-h culture in liquid 79 broth.

Since we found that insertion in the second to last SHC gene abolished TopA depletion-dependent activation of the first SHC gene sco4667, we tested whether sitKR deletion affected transcription of the penultimate SHC gene. RT-qPCR analysis confirmed that the deletion of sitKR genes caused a downregulation of sco4699 under high supercoiling conditions (Fig. 7D).

This analysis reinforced the idea that mutations of genes within the SHC region can have profound effects on the supercoiling-dependent transcription of more distant genes. Moreover, these results further support the notion that inhibition of SHC gene induction under elevated supercoiling conditions restores the sporulation of TopA-depleted strains. Finally, we have shown that a second two-component system, SitKR, inhibits sporulation when transcriptionally induced by SatKR.

DISCUSSION

TopA depletion in Streptomyces inhibits growth and sporulation. Before, we showed that parB deletion in the TopA-depleted strain partially suppresses the white phenotype and restores spore formation in S. coelicolor (38). This indicated that topological tension generated during formation of ParB complexes could inhibit their segregation and impair progress of sporulation at TopA depletion (38). However, we expected that additional explanation of the white phenotype of TopA-depleted mutant could be supercoiling-induced changes in the transcription of unknown sporulation/growth regulators. The application of random transposon mutagenesis in the TopA-depleted S. coelicolor strain allowed us to identify novel proteins engaged in sporulation regulation. Among the newly identified sporulation regulators was a two-component system named SatKR conserved in Streptomyces. The SatKR regulon includes a supercoiling-hypersensitive cluster that is also independently activated by TopA depletion. Transposon insertion in the satK gene encoding the kinase led to lower levels of SatR and was observed to revert the TopA depletion-induced activation of SHC and restored growth and sporulation. In contrast, increased SatR levels in the presence of its cognate kinase at wild-type supercoiling levels led to the activation of transcription of numerous SHC genes, inhibiting Streptomyces development.

The two-component system SatKR controls growth and differentiation by regulating SHC genes.

Streptomyces two-component systems are known to control antibiotic production, central metabolism, morphology, and differentiation (36). Although numerous TCSs in Streptomyces affecting these processes have been well described (e.g., CepRS [44], MacRS [45], PhoRP [46, 47], OsdKR [48], AfsQ1/Q2 [49], and DraRK [50]), the identification of signals to which TCSs respond has been successful for only a few TCSs (e.g., phosphate concentrations for PhoPR [46], nitrogen concentrations for AfsQ1/Q2 [49] and DraRK [50], iron availability for AbrA1/A2 [51], and redox stress for SenSR [52]). Here, we have established a role for SatKR in regulating growth rate, differentiation, and antibiotic production in S. coelicolor, but we were unable to predict the SatKR activation signal on the basis of our data. It should be noted that satKR gene expression is unresponsive to changes of DNA supercoiling; however, this does not exclude a signal related to changes of DNA topology.

Our results suggested that both the SatR levels in the cell and its activation by the cognate kinase SatK influenced the growth and development of S. coelicolor. SatR is a sequence homologue of the DegU response regulator from the Bacillus subtilis DegUS two-component system (37.2% identity with 223 amino acids [aa] of overlap). In B. subtilis, DegU controls flagellum synthesis, antibiotic production, and biofilm formation, inducing the expression of genes involved in matrix formation, competition, and nutrient acquisition (53). Interestingly, the S. coelicolor genome encodes another DegUS homologue, SCO5784-5785, with a similar percentage identity to DegUS proteins as SatKR. Moreover, SCO5784-5785 has also been shown to influence sporulation as well as antibiotic production; however, its effect on Streptomyces differentiation seems to be opposite that of SatKR (54). Markedly, some DegU-regulated genes are induced independently of its phosphorylation (55). Given that SatR stimulates SHC transcription only in the presence of a functional cognate kinase SatK and that reduced levels of SatR in the absence of SatK (in the transposon mutant) lead to the inhibition of SHC transcription despite high DNA supercoiling conditions, we inferred that both SatR states (phosphorylated and unphosphorylated [see below]) were able to regulate transcription, although with different effects on the regulated genes. An interesting feature is a localization of the sco3388 gene downstream of the satKR genes, presumably forming an operon. sco3388 is similar to the trmB gene from B. subtilis; however, it was previously shown that, in contrast to its homolog, sco3388 does not determine tunicamycin resistance (56). SCO3388 was indicated to control cell wall integrity and influence spore viability (56); thus, its proximity to satKR genes may suggest their cooperation in controlling sporulation in S. coelicolor.

Since the SatR regulon includes the SHC cluster and SHC is induced in the TopA-depleted strain (12), we concluded that the activation of one or more of the 26 supercoiling-sensitive SHC genes may be responsible for the growth and sporulation inhibition. The SHC region is unique to S. coelicolor and shows low synteny among Streptomyces, even though homologues of individual SHC genes can be found in other species (see Fig. S8 and Data Set S1, Tab 9, in the supplemental material). The functions of many of the SHC genes remain unknown, but those genes with predicted functions encode two-component systems (sco4667-4668), two probable regulators (sco4671 and sco4673), a putative Rhs protein (sco4699), and the RsfA anti-sigma factor (sco4677) (57) that represses SigF, a sporulation-specific sigma factor (58, 59).

The first SHC operon, sco4667-4668, encoded another novel two-component system named SitKR, which seems to be critical for sporulation and growth. SitKR is conserved among Streptomyces species; however, its chromosomal location differs among species. Interestingly, the regulation of sitKR gene expression by SatKR is a phenomenon of a two-component system signaling cascade that has been observed for only a few bacterial species, e.g., SsrAB regulation by OmpR/EnvZ and PmrAB by PhoPQ in S. enterica (60, 61) and RseDE regulation by PhoPR in B. subtilis (62). sitKR genes are among the most significantly induced by TopA depletion and satR overexpression. Whereas deletion of satKR did not restore sporulation in the TopA-depleted strain, a strain in which the sitKR genes had been deleted was capable of producing spores. The regulon of sitKR has not been established; however, we found that operon deletion abolished supercoiling induction of other genes in the SHC (e.g., sco4699). While SitR may directly control the expression of sco4699 (and possibly other SHC genes), there may also be synchronized regulation between (at least) the first (sitKR) and last genes (sco4699-4700) within the SHC region. We cannot exclude the possibility that upregulation of sco4677 (encoding anti-sigF) by SatR (and possibly by SitKR) or by increased DNA supercoiling can significantly contribute to sporulation inhibition.

The significance of SHC gene induction for the inhibition of growth and sporulation of the TopA-depleted strain was reinforced by the identification of transposon mutations in the next to the last gene of the SHC region (sco4699), which led to restored sporulation in the TopA-depleted strain. sco4699 encodes a homologue of the Rhs protein from E. coli, where Rhs is a secreted toxin that mediates cellular competition (43). Importantly, transposon insertion in sco4699 also abolished the expression of the first SHC gene, sitK. It is possible that the observed effect on sporulation and growth in the SHC transposon strain was due to Rhs inactivation, but we consider the abolished induction of sitKR by TopA depletion to be a more likely explanation for the restored sporulation in the transposon strain.

Thus, abolished induction of the SHC region upon TopA depletion in three distinct mutant strains (transposon in satK, transposon in rhs, and deletion of sitKR) restored growth and sporulation despite lowered TopA levels and high chromosomal supercoiling. Notably, spores produced by all the strains with high DNA supercoiling and inhibited SHC transcription had varied sizes and impaired resistance to damaging agents. This phenomenon could result from a lack of additional factors that are needed for proper spore formation and maturation, which may be encoded outside the SHC region. Nevertheless, we established that genes within the SHC region were under the control of the SatKR two-component system and included sporulation and growth rate inhibitors. We speculate that in response to unidentified factors, SatKR induces SHC genes to support slower growth and inhibit sporulation under unfavorable, possibly DNA topology-affecting conditions. Given that a similar induction of SHC genes results from increased supercoiling, we hypothesize that both factors, i.e., changes in chromosome supercoiling and SatKR activation, would independently prevent sporulation under specific environmental conditions, possibly connected with chromosome damage.

The two-component system SatKR collaborates with chromosome supercoiling in the regulation of gene expression.

SHC genes are subject to no or low transcription in the wild-type S. coelicolor strain under standard culture conditions, but they are activated by high DNA supercoiling as a result of TopA depletion. However, the presumed low levels of SatR resulting from kinase inactivation (transposon mutant, with low satR transcript levels) completely abolished SHC induction by high DNA supercoiling. Importantly, the deletion of satKR genes together did not abolish SHC activation by increased DNA supercoiling. Thus, we suggest that in addition to the presence of the kinase, the SatR phosphorylation state may also depend on SatR protein level. Taking into account the possibility of nonspecific activation of SatR by other cellular factors (e.g., acetyl phosphate [AcP] [63] or noncognate TCS kinases [64]), we suggest that SatR overproduction results in inefficient or nonspecific modification of SatR. Additionally, considering the potential phosphatase activity associated with many kinases, at this stage we cannot predict the phosphorylation state of SatR in the presence and absence of SatK; however, we assume that it is different in those two genetic backgrounds. Since we established that deletion of satKR genes had different effects on growth and gene expression than a transposon inserted in satK, we speculate that upon satK inactivation, SatR in its inactive state may also bind to DNA and control the expression of SHC genes, reversing their activation by high supercoiling. This could be achieved either by different DNA binding specificities/affinities for phosphorylated and unphosphorylated SatR or different abilities to form higher-order complexes or interact with RNA polymerase. Our speculation here is supported by evidence showing that other two-component system response regulators can act when unphosphorylated (65), e.g., OmpR (66, 67), PhoP (68), SsrB (69), and DegU (55).

Possible explanations for the coordinated regulation of gene transcription by DNA supercoiling and transcription factor activity could include the formation of a supercoiled DNA loop due to transcription factor binding, or supercoiling-dependent binding of transcription factors. Synchronized activation of the SHC region could suggest formation of the loop encompassing this region, stabilized by SatR. Moreover, loop formation could depend on global DNA supercoiling. There is published evidence for DNA looping by some repressor proteins promoted by supercoiling, e.g., LacI in E. coli or bacteriophage λ (70, 71), because DNA compaction increases the likelihood of bringing together distant binding sites. Therefore, the affinity of SatR for DNA could be altered not only by its phosphorylation but also by DNA supercoiling. Indeed, the binding of some response regulators has been shown to be influenced by DNA supercoiling, e.g., OmpR binding in S. enterica (72).

Elevated satR transcription in the presence of kinase activated SHC transcription in the wild-type strain, showing that the SatKR system could control the SHC region independently of DNA supercoiling. In other bacterial species, it has also been suggested that two-component systems might influence the DNA supercoiling state by interacting with topoisomerases or altering their activity. In Staphylococcus aureus, deletion of the ArlSR two-component system elevates DNA supercoiling (73), while the E. coli RstB response regulator interacts with TopA and increases its activity (74). Although we observed elevated supercoiling in the TopA-depleted strain with a transposon insertion in satK, and lowered satR transcript levels, we cannot exclude the possibility that local DNA topology changes when SatR binds.

We propose a model in which SatR and DNA supercoiling (caused by TopA depletion), both independently and cooperatively, regulate SHC transcription (Fig. 8). SatR, as a two-component system regulator in the presence of SatK kinase, activates SHC transcription. However, in the absence of kinase, SatR also regulates the transcription of SHC genes, reversing their activation by high supercoiling. Induction of SitKR, an SHC-encoded two-component system, inhibits sporulation; however, it is still unclear whether this sporulation defect is a direct effect or whether other SHC-encoded proteins are involved. Our results are indicating that there is an unprecedented complex regulatory interplay that connects DNA supercoiling with a cascade of two-component regulatory systems in controlling growth and sporulation in Streptomyces.

FIG 8

Proposed model of cooperation of SatKR, SitKR, and DNA supercoiling in SHC transcription regulation and sporulation inhibition. SHC genes are activated by elevated DNA supercoiling. In the absence of SatK and at low level, SatR inhibits supercoiling-dependent activation of SHC transcription. At high intracellular concentration and when activated by the cognate kinase SatK, SatR induces transcription of SHC genes, independently of supercoiling. Among SHC products is SitKR two-component system, which is involved in sporulation inhibition. Contribution of other SHC gene-encoded proteins in the sporulation inhibition is also probable.

MATERIALS AND METHODS

Bacterial strains, plasmids, and growth conditions.

Basic DNA manipulations were performed according to standard protocols (75). Unless otherwise stated, all enzymes and isolation kits were supplied by Thermo Fisher Scientific (Waltham, MA, USA) and New England Biolabs (Ipswitch, MA, USA). Bacterial media and antibiotics were purchased from Difco Laboratories (Detroit, MI, USA) and Carl Roth (Karlsruhe, Germany), respectively. The S. coelicolor strains used in this study are listed in Text S1A in the supplemental material. Strain construction details are provided in Text S1B. The growth conditions, antibiotic concentrations, and S. coelicolor conjugation procedure as described by Kieser et al. (76) were used. To restore TopA levels in the TopA-controlled strain (PS04) (during growth analysis and spore sensitivity assay), the growth medium was supplemented with 0.5 to 1 μg/ml thiostrepton (77). During conjugation with the PS04 strain, thiostrepton concentrations of 5 μg/ml were used, unless otherwise stated (it was earlier shown that induction with thiostrepton at a concentration higher than 2 μg/ml increases TopA levels only slightly above the level of the wild type).

For growth rate analyses, S. coelicolor cultures were inoculated with spores diluted to an optical density at 600 nm (OD₆₀₀) of 0.01/ml in 79 medium (peptone 1% v/o, casein hydrolysate 0.2% v/o, yeast extract 0.2%, NaCl 0.1 M, pH 7.2 to 7.4). To determine the growth rate, cultures were grown for 48 to 55 h in microplates in a final volume of 300 μl using a Bioscreen C instrument (Oy Growth Curves Ab Ltd., Helsinki, Finland), with optical density (OD₆₀₀) measurements being taken at 20-min intervals. To analyze S. coelicolor differentiation, strains were plated on solid MS agar medium (76) and cultured for 3 to 7 days.

Transposon mutagenesis.

Random transposon mutagenesis was performed on the TopA-controlled strain (PS04) using the synthetic Himar1 transposon (3,276 bp in length, containing a spectinomycin resistance gene [aadA(1) and R6Kγ ori flanked with ITRs (inverted terminal repeats)] (78). Exconjugants were selected using hygromycin and spectinomycin, in addition to 0.2 μg/ml thiostrepton, to limit transposase induction but increase the TopA level. Spores of exconjugants were collected and inoculated into liquid cultures with thiostrepton (0.2 μg/ml); these were cultivated at 39°C overnight to eliminate the pHSM plasmid. The mutant library was then spread for single colonies (to obtain at least 16,000 mutants) on MS agar plates supplemented with spectinomycin but no thiostrepton (the PS04 strain has a “white phenotype” under these conditions), and gray colonies were screened for. The transposon insertion sites in the selected transposon library clones were identified using a rescue plasmid approach (78). In the MGHM5 strain (PS04 sco3390::Himar1 sco2474::Himar1), insertion sites were additionally confirmed by whole-genome sequencing (Genomed, Warsaw, Poland).

Supercoiling reporter plasmid isolation.

The pWHM3Hyg/pWHM3Spec plasmid, which served as a probe of the DNA supercoiling state in vivo, was isolated according to a previously described procedure (77) from S. coelicolor strains (MGHM5_RP, MS10, and MS11—derivatives of analyzed mutants modified by pWHM3Hyg introduction), where these strains were cultivated in liquid 79 medium for 24 h at 30°C. The isolated plasmid DNA was resolved on a 0.8% agarose gel with 2.32 μg/ml chloroquine in Tris-acetate-EDTA (TAE) buffer. To visualize topoisomers, the gel was stained with ethidium bromide. The topoisomer distribution was analyzed using ImageJ Software.

RNA-Seq and data analysis.

For the RNA-seq experiments, RNA was isolated from S. coelicolor mycelia obtained from 18-h cultures in 30 ml YEME/tryptic soy broth (TSB) liquid medium (76). The mycelia were collected by centrifugation, frozen, and stored at −70°C for subsequent RNA isolation. RNA was isolated using the procedure described previously by Moody et al. (84), after which the preparations were subjected to digestion with TURBO DNase I (Invitrogen, Waltham, MA, USA) and checked using PCR to ensure the samples were free of chromosomal DNA contamination. Strand-specific cDNA libraries with an average fragment size of 250 bp were constructed and sequenced using a MiSeq kit (Illumina, San Diego, CA, USA) at the Farncombe Metagenomics Facility at McMaster University (Hamilton, Canada). Paired-end 76-bp reads were subsequently mapped against the S. coelicolor chromosome using Rockhopper software (79), achieving 1.0 × 10⁶ to 1.5 × 10⁶ successfully aligned reads per sample. For data visualization, Integrated Genomics Viewer (IGV) software was used (80, 81). The analysis of differentially regulated genes was based on the data generated by Rockhopper software. To calculate the fold change in gene transcription, the normalized gene expression in the control strain (wild type [WT] or PS04) was divided by the normalized gene expression under particular experimental conditions, delivering information on the fold change, and subsequently, the log₂ value of the fold change was calculated. The genes with a q value (Rockhopper adjusted P value) greater than or equal to 0.01 and a log₂ fold change in the range from −1.5 to 1.5 were rejected from the subsequent analysis as not significant. Volcano plots were prepared using R Studio software and the EnhancedVolcano package (R package version 1.10.0; https://github.com/kevinblighe/EnhancedVolcano).

RT-qPCR.

For RT-qPCR analyses, RNA was isolated from S. coelicolor mycelia obtained from 24-h cultures growing in 5 ml of liquid 79 medium. Transcription was arrested by adding STOP solution (95% ethanol [EtOH] [vol/vol], 5% phenol [vol/vol]) (82), and mycelia were harvested by centrifugation and frozen at −80°C. Total RNA was isolated using TRI-Reagent (Sigma-Aldrich, Saint Louis, MO, USA) according to the manufacturer’s procedure. Homogenization was performed in a FastPrep-24 instrument (MP Biomedicals, Irvine, CA, USA) (6 m/s, 2 cycles × 45 s). After centrifugation, RNA was isolated by chloroform extraction, purified on a column (Total Mini RNA, AA Biotechnology, Gdańsk, Poland), and eluted with 50 μl of ultrapure water. The isolated RNA was digested with TURBO DNase (Invitrogen) according to the manufacturer’s instructions at 37°C for 30 min. Then, RNA was purified and concentrated using a CleanUp RNA Concentrator (AA Biotechnology) and eluted with 17 μl of ultrapure water. Five hundred nanograms of RNA was used for cDNA synthesis using a Maxima First Strand cDNA synthesis kit (Thermo Fisher Scientific) in a final volume of 20 μl. The original manufacturer’s protocol was modified for GC-rich transcripts by increasing the temperature of the first-strand synthesis to 65°C and elongation time up to 30 min. Subsequently, the obtained cDNA was diluted five times and used directly for quantitative PCRs performed with PowerUp SYBR green Master Mix (Applied Biosystems, Waltham, MA, USA). The relative level of a particular transcript was quantified using the comparative ΔΔC_T method, and the hrdB gene was used as the endogenous control (StepOne Plus real-time PCR system; Applied Biosystems). The sequences of optimized oligonucleotides used in this study are given in Text S1C.

Microscopic analyses.

For analysis of spore formation, the tested strains were inoculated at the acute-angled junctions of coverslips inserted at 45° into minimal medium agar plates supplemented with 1% mannitol and cultured for 53 h to ensure sporulation of all mutant strains. Sporulation of the TopA-controlled (PS04) strain was induced by the addition of 1 μg/ml thiostrepton. Coverslips were fixed with methanol and then mounted using 50% glycerol solution in phosphate-buffered saline (PBS). DNA was stained with a 2-μg/ml 4′,6′-diamidino-2-phenylindole (DAPI) solution (Molecular Probes, Eugene, OR, USA). Microscopic analyses were performed using a Leica microscope (Leica Microsystems, Wetzlar, Germany) with phase-contrast imaging. Images were analyzed using ImageJ software. The statistical analysis of spore length was performed using RStudio (83) and the Student’s t test for paired samples.

Spore sensitivity assay.

To test spore viability, S. coelicolor strains were cultured on MS agar plates for 5 days. Next, the spores were collected and incubated in 5% SDS (sodium dodecyl sulfate) solution for 1.5 h, washed twice with ultrapure water, and resuspended in 0.5 ml of water. Next, serial dilutions were spread on MS agar plates to obtain single colonies. Subsequently, the number of colonies grown after SDS treatment was compared with the negative control (spores of the same strain collected and incubated in water). Spore viability was calculated as a ratio of the number of colonies obtained for spores treated and untreated with SDS.

Data availability.

RNA-seq data are available at ArrayExpress (EMBL-EBI) accession number E-MTAB-10835.