Correspondence: Jan Kormanec, Institute of Molecular Biology, Slovak Academy of Sciences, Dubravska cesta 21, 845 51 Bratislava, Slovak Republic. Tel.: +421 2 59307419; fax: +421 2 59307416; e-mail: email@example.com
Three regulators, Aur1P, Aur1R and a SARP-family Aur1PR3, have been previously found to control expression of the aur1 cluster for the angucycline antibiotic auricin in Streptomyces aureofaciens CCM 3239. Here, we describe an additional regulatory gene, aur1PR4, encoding a homologue from the SARP-family regulators. Its role in auricin regulation was confirmed by its disruption that dramatically affected auricin production. However, transcription from the aur1Ap promoter, directing expression of 22 auricin biosynthetic genes, was not substantially affected in the Δaur1PR4 mutant. A new promoter, sa13p, directing transcription of four putative auricin tailoring genes, was found to be dependent on aur1PR4. Moreover, analysis of the sa13p promoter region revealed the presence of three heptameric repeat sequences corresponding to putative SARP-binding sites. Expression of aur1PR4 is directed by a single promoter, aur1PR4p, which is induced after entry into stationary phase. Transcription from aur1PR4p was absent in a S. aureofaciens Δaur1P mutant strain, and Aur1P was shown to bind specifically to the aur1PR4p promoter. These results indicate a complex network of regulation of the auricin gene cluster. Both Aur1P and Aur1PR3 are involved in regulation of the core aur1A-U biosynthetic genes, and Aur1PR4 in regulation of putative auricin tailoring genes.
The Gram-positive soil bacteria of the genus Streptomyces are one of the most important groups of microorganisms, which produce bioactive secondary metabolites, including many known antibiotics. The genes required for antibiotic biosynthesis are generally clustered and they are coregulated by pathway-specific transcriptional activators that are located in these clusters. The expression of these regulatory genes is controlled by complex regulatory mechanisms that determine the onset of antibiotic production (Bibb, 2005; Martin & Liras, 2010). Pathway-specific transcriptional activators have been classified into different families. One of the most prominent of them is the Streptomyces antibiotic regulatory protein (SARP) family. The members of this family are activators characterized by an N-terminal OmpR-type winged helix-turn-helix (HTH) DNA-binding domain (Wietzorreck & Bibb, 1997). This family is represented by several antibiotic pathway-specific activators, like ActII-ORF4 and RedD for actinorhodin and undecylprodigiosin in Streptomyces coelicolor (Fernandez-Moreno et al., 1991; Takano et al., 1992), DnrI for daunorubicin in Streptomyces peucetius (Sheldon et al., 2002), CcaR for clavulanic acid and cephamycin in Streptomyces clavuligerus (Santamarta et al., 2011), FdmR1 for fredericamycin in Streptomyces griseus (Chen et al., 2008), and TylS for tylosin in Streptomyces fradiae (Cundliffe, 2008). However, it contains also well-studied pleiotropic regulatory protein AfsR (Tanaka et al., 2007). Studies of the interactions between SARP-family activators with their target genes identified heptameric direct repeats (5′-TCGAGXX-3′) with 4-bp spacers, located stringently 8 bp upstream of the -10 regions of promoters (Wietzorreck & Bibb, 1997; Arias et al., 1999; Sheldon et al., 2002; Tanaka et al., 2007).
We previously identified a polyketide synthase gene cluster, aur1, responsible for the production of the angucycline antibiotic auricin in Streptomyces aureofaciens CCM 3239 (Novakova et al., 2002). Auricin, which is a unique angucycline polyketide conjugated to an amino deoxyhexose d-forosamine, is produced during a very narrow time period of several hours following entry into stationary phase. This unusual phenomenon likely arises from a strict, but complex regulation of auricin biosynthesis, involving several predicted transcriptional regulators (Kutas et al., 2013). Three of them have already been characterized, and their role in regulation of auricin biosynthesis is described in the Fig. 6. The aur1P gene, which encodes a protein belonging to a family of ‘atypical’ response regulators, has been shown to specifically activate expression of the auricin biosynthesis genes by directly binding and activating the aur1Ap promoter, triggering the expression of 22 structural genes (aur1A-aur1U) of the auricin cluster (Novakova et al., 2005, 2011a). The product of the aur1R gene, which encodes a protein homologous to the transcriptional repressors of the TetR family, has been shown to repress expression of the aur1P gene by directly binding the aur1Pp promoter (Novakova et al., 2010a). Another regulatory gene, aur1PR3, encoding a homologue of the SARP family of transcriptional activators, has also been found to be involved in auricin regulation, and it is also under the direct control of Aur1R (Novakova et al., 2011b).
The auricin cluster is located on pSA3239, a large linear plasmid. Although the main part of the auricin-specific genes is clustered, several genes encoding auricin tailoring enzymes are scattered across a rather large aur1-flanking region (Novakova et al., 2013). The aur1 cluster is even interrupted by eight genes encoding proteins with no auricin-specific function, including bpsA, a gene involved in the biosynthesis of the blue pigment indigoidine (Novakova et al., 2010b). Intriguingly, disrupting of bpsA, by replacing it with the resistance marker gene aac3(IV), causes increased auricin production. A possible explanation of this increase may be a polar effect of the aac3(IV) gene on the downstream sa9 gene encoding a SARP regulator (Novakova et al., 2010b). In the present study, we describe the characterization of sa9 (aur1PR4). We provide evidence that it is essential for auricin production in S. aureofaciens CCM 3239. We describe the identification of a putative target promoter of this regulator. In addition, we investigate the transcriptional regulation of aur1PR4 and show that the auricin-specific activator Aur1P directly regulates its expression.
Materials and methods
Bacterial strains, plasmids and culture conditions
The S. aureofaciens CCM 3239 wild type was from the Czechoslovak Collection of Microorganisms, Brno, Czech Republic. The S. aureofaciens Δaur1P mutant strain was described in Novakova et al. (2005), the S. aureofaciens Δaur1R mutant strain in Novakova et al. (2010a) and the S. aureofaciens Δaur1PR3 mutant strain in Novakova et al. (2011b). Escherichia coli SURE and a pBluescript II SK- plasmid (Stratagene) were used for E. coli-cloning experiments. Escherichia coli BW25113/pIJ790 was the host for PCR-targeted gene disruption using the plasmid pIJ773 and E. coli ET12567/pUZ8002 as a nonmethylating host for conjugation (Gust et al., 2003). The Streptomyces integrative plasmid pPM927 (Smokvina et al., 1990) was used for complementation studies. The growth of S. aureofaciens CCM 3239 strains was carried out in Bennet medium (Horinouchi et al., 1983) as described in Novakova et al. (2010a). For RNA isolation, spores (5 × 108 CFU) of the particular S. aureofaciens CCM 3239 strain being studied were inoculated in 50 mL of liquid Bennet medium and grown at 28 °C to different growth phases. Conditions for E. coli growth and transformation are described by Ausubel et al. (1995).
Disruption of the S. aureofaciens CCM 3239 aur1PR4 gene
The S. aureofaciens CCM 3239 cosmid library (comprising Sau3AI partially digested DNA fragments in the BamHI site of SuperCos-1) is described in Novakova et al. (2010a). Deletion of the whole coding region of the S. aureofaciens CCM 3239 aur1PR4 gene (aside from its start and stop codons) in the cosmid pCosSA19 (Novakova et al., 2010b) was achieved using the PCR-targeted REDIRECT method (Gust et al., 2003). In brief, the apramycin resistance cassette was PCR amplified using the primers SARP5DIR (5′- AATTGCCGCT GTTCTGCTGACGAGGAGTCATCACCCATGTGTAGGCTGGAGCTGCTTC-3′) and SARP5REV (5′-CCGATACCGGGACTCCTTCGTGTCGTGCGCCGGTGCTCAATT CCGGGGATCCGTCGACC -3′) and the template plasmid pIJ773. The resulting PCR product was used to electroporate E. coli BW25113/pIJ790 containing cosmid pCosSA19.
The replacement of the coding region in the resulting SuperCos1-derived cosmids pCosSA19-aur1PR4/AmR was verified by restriction mapping. This cosmid was transformed into the nonmethylating E. coli ET12567/pUZ8002 strain and introduced into S. aureofaciens CCM 3239 by conjugation. Four apramycin-resistant and kanamycin-sensitive colonies were selected from the conjugation and the double-cross event verified by Southern blot hybridization (data not shown). One representative strain, S. aureofaciens Δaur1PR4 was chosen for further study. The plasmid pPM-aur1PR4, used for complementation of the Δaur1PR4 mutation, was prepared by inserting a 1280-bp Acc65I-BsiWI fragment of the cosmid pCosSA19 (containing the aur1PR4 gene with its promoter) in pPM927 digested with Acc65I.
RNA isolation and S1-nuclease mapping
Isolation of total RNA and high-resolution S1 nuclease mapping were performed according to Kormanec (2001). 40-g samples of RNA were hybridized to c. 0.015 pmol of a corresponding DNA probe labelled at one 5′ end with [γ-32P] ATP (c. 106 d.p.m. pmol−1 of probe). The S1 probes were prepared as follows: probe 1 (for the aur1PR4p promoter, Fig. 1a) was prepared by PCR amplification from the cosmid pCosSA19 using the 5′ end-labelled reverse primer SRP5REV (5′-CGTGTG AGCAGCAGTGCCAGC-3′) from the aur1PR4-coding region and the unlabelled direct primer SRP5DIR (5′-CAACGGTTACTCCGCCGAGCC-3′) from aur1PR4 upstream region; probe 2 (for the sa13p promoter) was prepared similarly by PCR amplification using 5′ end-labelled primer Sa13S1rev (5′- GCCAGCTCCTGCGCCTTGACC -3′) from the sa13-coding region and the unlabelled primer Sa13S1dir (5′- GCTTGTCGTAGACGGGCAGCAGC -3′) from sa13 upstream region; the probes for the aur1Ap promoter and the control hrdBp2 promoter were described previously (Kormanec & Farkasovsky, 1993; Novakova et al., 2005). The primers were labelled at the 5′ end with [γ-32P] ATP (ICN, 4500 Ci mmol−1) and T4 polynucleotide kinase following the method of Ausubel et al. (1995). The protected DNA fragments were analysed on DNA sequencing gels together with G+A and T+C sequencing ladders derived from the end-labelled fragments (Maxam & Gilbert, 1980).
Electrophoretic mobility-shift assay (EMSA)
A 500-bp DNA fragment (probe 1, Fig. 1a) comprising the aur1PR4p promoter region was prepared by PCR amplification from pCosSA19 using the 5′ end-labelled primer SRP5REV and the unlabelled primer SRP5DIR. EMSA with purified Aur1P protein was performed as described in Novakova et al. (2005). The 32P-labelled DNA fragment (0.4 ng, c. 5000 d.p.m.) was incubated with increasing amounts of the Aur1P protein for 15 min at 30 °C, and protein-bound and free DNA were resolved on a nondenaturing polyacrylamide gel as described by Novakova et al. (2005).
Analysis of auricin production
Spores (5 × 108 CFU) of S. aureofaciens CCM 3239, ΔaurPR4 and complemented strains were inoculated into 50 mL Bennet medium in 250-mL Erlenmeyer flasks, and the cultures grown on an orbital shaker at 270 r.p.m. and 28 °C. 4 mL aliquots were taken from the cultures at various time points, and auricin was extracted and analysed by TLC and HPLC as described previously (Novakova et al., 2011b).
An additional SARP gene is present in the auricin cluster
We previously found that disruption of the bpsA gene, responsible for indigoidine biosynthesis in S. aureofaciens CCM3239, resulted in increased auricin production (Novakova et al., 2010b). A possible explanation of this increase may be a polar effect on the downstream sa9 gene caused by replacement of bpsA by the resistance marker gene aac3(IV) (Fig. 1a). This gene encodes a protein with sequence similarity to the SARP family of Streptomyces antibiotic regulatory proteins (Fig. 1b). Many members of this family are essential for the positive regulation of secondary metabolism in streptomycetes and are characterized by an N-terminal OmpR-type winged (HTH) DNA-binding domain, which contains two helical segments (α2 and α3) separated by an 11-residue loop (Wietzorreck & Bibb, 1997). This secondary structure motif and the DNA-binding region are also conserved in the protein produced by sa9, leading us to change the name of the gene to aur1PR4.
The aur1PR4 gene is essential for auricin production
In order to confirm the proposed role of the aur1PR4 gene in auricin biosynthesis, the coding region of aur1PR4 (aside from the start and stop codons) was deleted from the chromosome using a PCR-targeting system designed for the disruption of Streptomyces genes and replaced by a resistance cassette (Gust et al., 2003). This mutation did not affect growth and differentiation. The auricin production of this S. aureofaciens Δaur1PR4 mutant was investigated as described previously (Novakova et al., 2010a). The deletion of aur1PR4 had a dramatic effect on auricin production. The inhibition zones corresponding to auricin in the mutant were much smaller than for the wild-type S. aureofaciens CCM 3239 strain for all time points examined (Fig. 2b and c). The quantities of auricin produced were determined by HPLC of the ethyl acetate extracts (Fig. 2d). Deletion of aur1PR4 caused a dramatic, 20-fold decrease in auricin production. To verify that this phenotype was due to the deletion of aur1PR4, the S. aureofaciens Δaur1PR4 strain was complemented in trans by transformation with the plasmid pPM-aur1PR4, which contained the aur1PR4 gene, including its promoter. The level of auricin produced by the complemented strain was similar to that of the wild-type strain, confirming that the decrease in auricin production by S. aureofaciens Δaur1PR4 is indeed due to aur1PR4 deletion. These results suggest that aur1PR4 has an important role in the positive regulation of auricin biosynthesis.
We previously characterized the aur1Ap promoter that directed expression of 22 structural genes (aur1A-aur1U) of the auricin cluster (Novakova et al., 2005, 2011a).
Dependence of the aur1Ap transcription on aur1PR4 was investigated by S1-nuclease mapping with RNA isolated from S. aureofaciens CCM3239 wild-type and the aur1PR4-mutant strains cultivated in liquid Bennet medium. In addition, RNA from the aur1PR3-mutant strain with deleted gene encoding previously characterized SARP-regulator Aur1PR3 (Novakova et al., 2011b) was used in this experiment. In contrast to the aur1PR3 mutant, where the level of aur1Ap mRNA was substantially lower (Novakova et al., 2011b), the level of aur1Ap mRNA from all the time points was not substantially affected in the aur1PR4 mutant (Fig. 3a). We investigated expression of additional putative auricin biosynthetic genes flanking to the aur1 cluster and their proposed dependence on aur1PR4 by S1-nuclease mapping. A putative operon of four genes, sa13, sa12, sa11, sa10, encoding proteins with similarity to several polyketide tailoring enzymes (Novakova et al., 2010b) is located just downstream of the aur1PR4 gene (Fig. 1a). S1-nuclease mapping was performed using a probe 2 (Fig. 1a) and the same RNA samples as for the aur1Ap promoter. As shown in Fig. 3a, an RNA-protected fragment was identified that corresponded to the sa13p promoter with the transcriptional start point (TSP) at a G 50-bp upstream from the ATG initiation codon of sa13 (Fig. 3c). The promoter was inactive during exponential phase (9 h) and was induced at the beginning of stationary phase (12 h) that correlates with onset of auricin production. The level of sa13p mRNA decreased during late stationary phase. No RNA-protected fragment was identified using the tRNA control. The level of sa13p mRNA was dramatically lower in the Δaur1PR4 mutant. However, its level was only partially reduced in the Δaur1PR3 mutant. As an internal control, S1-nuclease mapping was performed with the same RNA samples using a probe fragment specific for the S. aureofaciens hrdBp2 promoter, which is expressed fairly constantly during growth and differentiation (Kormanec & Farkasovsky, 1993). RNA-protected fragments of similar intensities corresponding to the hrdBp2 promoter were identified in all RNA samples (Fig. 3b). As shown in Fig. 3d, the upstream region of the sa13p promoter contained three heptameric direct repeats, which are similar to the consensus sequence of typical SARP-binding sites (5′-TCGAGXX-3′) and located correctly 8 bp upstream of the -10 region. These results indicate that the sa13p promoter is dependent on the aur1PR4 gene, and likely all four genes are under the control of this SARP regulator. In addition, these genes are also partially dependent upon other SARP-regulator Aur1PR3.
Transcriptional analysis of the aur1PR4 gene
To investigate the expression of the aur1PR4 gene, S1-nuclease mapping was performed with RNA isolated from the S. aureofaciens CCM3239 wild-type strain cultivated to various growth stages in liquid Bennet medium. We previously characterized two auricin-specific regulators, an atypical response regulator Aur1P (Novakova et al., 2005) and a TetR-family repressor Aur1R (Novakova et al., 2010a). To investigate whether these regulatory genes have an effect on the transcription of aur1PR4, RNA was also isolated from S. aureofaciens mutant strains Δaur1P and Δaur1R from the same time points as wild-type strain. A single RNA-protected fragment was identified that corresponded to the aur1PR4p promoter with the (TSP) at an A 88-bp upstream from the ATG initiation codon of aur1PR4 (Fig. 4a and c). The promoter had similar time course as the sa13p promoter. The level of aur1PR4p mRNA was not substantially different in the Δaur1R mutant; however, the mRNA level was dramatically lower in the Δaur1P mutant. RNA-protected fragments of similar intensities corresponding to the control hrdBp2 promoter were identified in all RNA samples (Fig. 4b). The putative -35 and -10 regions of the aur1PR4p promoter (Fig. 4c) showed similarity to the consensus sequence (TTGaCA-N16–18-TAgaaT) recognized by the major Streptomyces sigma factor σHrdB (Kang et al., 1997). Therefore, it is likely that it is transcribed by an RNA polymerase holoenzyme containing the principal sigma factor σHrdB. These results indicate that the aur1PR4p promoter is dependent on the auricin-specific transcriptional activator Aur1P.
The auricin-specific positive regulator Aur1P binds to the aur1PR4p promoter
To investigate whether Aur1P directly regulates aur1PR4p transcription, a purified N-terminal His-tagged Aur1P (Novakova et al., 2005) was examined for its binding to aur1RP4p by EMSA using a 500-bp 32P-labelled DNA fragment encompassing the aur1PR4p promoter (nucleotide positions −306 to + 195 in respect to the TSP). As shown in Fig. 5, a single retarded band was identified with increasing concentration of Aur1P. The specificity of the interaction was demonstrated by the competitive binding of the unlabelled fragment (Fig. 5, lane 6). These results indicate that Aur1P is capable of binding to the aur1PR4p promoter and that this promoter is under the direct control of this auricin-specific activator. We similarly investigated the binding of the auricin-specific repressor Aur1R to the aur1PR4p promoter; however, no binding was found (data not shown).
It has already been suggested that the auricin gene cluster aur1 in S. aureofaciens CCM 3239 is under the control of a complex regulatory mechanism, which is responsible for the rather brief period of auricin production after entry into the stationary phase (Kutas et al., 2013). Two auricin-specific positive regulators, the ‘atypical’ response regulator Aur1P and the SARP-family Aur1PR3, are under the direct control of a negative TetR-family regulator, Aur1R, which ensures that the transcription of the auricin biosynthetic genes is correctly timed (Novakova et al., 2005, 2010a, 2011b) (Fig. 6). In the present paper, we have described an additional regulatory gene, aur1PR4, which encodes a member of the SARP family of antibiotics regulators. This suggestion is based on the comparison of its deduced protein product (278 aa) with databases, which revealed significant similarity only to the SARP-family regulatory proteins, with the highest similarity (45% identity, 65% similarity) to ActII-ORF4, from the well-characterized representatives of this family (Fernandez-Moreno et al., 1991). In addition, domain prediction revealed both SARP-family typical domains, the N-terminal Trans_reg_C DNA binding domain (pfam00486, aa 23–94) and the C-terminal bacterial transcriptional activator domain (pfam03704, aa 103–248). The secondary structure of the N-terminal DNA-binding domain of Aur1PR4 was also very similar to that of the OmpR-like DNA-binding domain of the SARP family, containing a winged HTH region composed of two α-helices (α1 and α2) separated by an 11-residues loop (Fig. 1b).
Deletion of aur1PR4 dramatically reduces auricin production. Therefore, along with Aur1PR3, Aur1PR4 constitutes an additional auricin pathway-specific activator of the SARP family. However, in contrast to aur1PR3, deletion of aur1PR4 does not substantially affect transcription of the biosynthetic promoter aur1Ap, which controls the expression of 22 auricin biosynthetic genes in the aur1 operon (Novakova et al., 2005, 2011a). However, a single promoter, sa13p, located upstream of a putative operon of four genes, sa13, sa12, sa11, sa10 (Fig. 1a) encoding proteins with similarity to several polyketide tailoring enzymes (Novakova et al., 2010b), was found to be dependent on aur1PR4 (Fig. 6). The presence of correctly located three heptameric sequences similar to the SARP consensus sequence (Wietzorreck & Bibb, 1997; Arias et al., 1999; Sheldon et al., 2002; Tanaka et al., 2007) suggested that the promoter is under the control of SARP-family regulator. Therefore, these four genes are likely targets for Aur1PR4. Comparison of these heptameric sequences with several predicted or proven binding sites for SARP regulators (Fig. 3d) revealed the highest similarity to the binding sites for ActII-ORF4 regulator for the activation of actinorhodin gene cluster in S. coelicolor (Arias et al., 1999). This is consistent with the highest similarity of the Aur1PR4 DNA-binding region with those for ActII-ORF4 (Fig. 1b). All the SARP-binding sites were most conserved in the first four positions of the heptameric sequences, which was previously identified as a TCGA logo for the SARP consensus sequence (Santamarta et al., 2011). However, there are some exceptions from this conservation. Two recently characterized SARP regulators, NanR1 and NanR2, directing biosynthesis of nanchangmycin in S. nanchangensis likely recognize different binding sites with the consensus motif TTAG (N)6 TT(A/T)AG (Yu et al., 2012).
Interestingly, in contrast to aur1PR3, the aur1PR4 gene is not under the direct control of Aur1R, but is directly regulated by Aur1P. Moreover, aur1PR4 was induced later during growth than aur1PR3. This makes the regulation of auricin even more complex, involving two differently regulated SARP-family regulators (Fig. 6).
A single pathway-specific transcriptional activator is usually responsible for the activation of biosynthetic genes in antibiotic clusters. However, there is an increasing number of cases in which several pathway-specific activators are involved in the regulation of antibiotic clusters. For instance, two SARP-family regulators, TylS and TylT, and three additional regulatory proteins, TylR, TylU and TylQ, are involved in the regulation of the macrolide antibiotic tylosin in S. fradiae (Cundliffe, 2008). However, only TylS and TylR are required for tylosin production. A similar complex regulation has been described for the kinamycin-producing alp clusters in S. ambofaciens, where five regulatory genes, encoding three SARP homologues and two TetR-family homologues, have been identified, and the alpV gene, encoding one of the SARP regulators, appears to be involved in kinamycin regulation (Bunet et al., 2011). Two SARP homologues, SrrY and SrrZ, also appear to be involved in lankamycin regulation in S. rochei; one of them, SsrZ, is directly controlled by the other, SsrY (Suzuki et al., 2010). Two other SARP homologues, NanR1 and NanR2, are required for the transcription of nanchangmycin polyketide genes; the authors suggested that they might form a heterodimer to bind and activate the genes (Yu et al., 2012). The situation of auricin regulation is more complex as outlined in Fig. 6. In addition to the auricin-specific regulator Aur1P, which directly controls expression of the aur1 core biosynthetic genes aur1A-aur1U, two additional SARP-family regulators, Aur1PR3 and Aur1PR4, have roles in auricin regulation. Aur1PR3 is involved in an indirect activation of the biosynthetic aur1Ap promoter, and Aur1PR4 likely directly controls additional transcriptional unit comprising four genes encoding putative auricin tailoring enzymes. A partial dependence of the sa13p promoter on aur1PR3 indicates dependence also on this SARP regulator. However, additional binding experiments of both regulators to the sa13p promoter region will reveal whether this effect is direct and also way of binding to the three identified heptameric repeat sequences in the promoter. These experiments and also further investigation of the role of both SARP homologues in regulation of auricin cluster are in progress.
We are grateful to Dr Mark Buttner (John Innes Centre, Norwich, UK) for the plasmid pPM927 and to Dr Bertold Gust (John Innes Centre, Norwich, UK) for kindly providing all the plasmids and strains used in the PCR-targeting system; the system itself was supplied by Plant Bioscience Ltd. (Norwich, UK). We thank Ludovit Skultety for his assistance with the mass spectrometric analysis. This work was supported by the Slovak Research and Development Agency under contract No. APVV-0203-11. This contribution is also a result of the TRANSMED (ITMS: 26240120008) and TRANSMED 2 (ITMS: 26240120030) projects, supported by the Research & Development Operational Programme funded by the ERDF.