All strains and plasmids are listed in Table 1. E. coli DH5α (Life Technologies) was used for routine subcloning. E. coli ET12567 harboring the conjugative plasmid pUB307 (provided by C. P. Smith, UMIST, Manchester, UK) was used to perform intergeneric conjugation from E. coli to Streptomyces species Flett et al. 1997; Luzhetskyy et al. 2006. For plasmid and total DNA isolation, E. coli and S. antibioticus strains were grown as described by Sambrook and Russell (2001), and Kieser et al. (2000). For simocyclinone production, S. antibioticus strains were grown in liquid NL5 medium (NaCl 1 g l^-1, KH₂PO₄ 1 g l^-1, MgSO4 × 7H₂O 0.5 g l^-1, glycerol 25 g l^-1, L-glutamin - 5.84 g l^-1, trace elements - 2.0 ml, pH 7.3 prior to sterilization) at 30°C. For conjugation, spores of S. antibioticus strains were harvested from a sporulated lawn grown on soya-mannitol or oatmeal medium Kieser et al. 2000Luzhetskyy et al. 2006. When it was necessary, bacterial strains were grown in the presence of respective antibiotics. X-gal and IPTG were used for blue-white colony selection in the case of the pBluescript, pSET152, pKC1139, pKC1218E vectors as described elsewhere Kieser et al., 2000; Sambrook et al., 2001.

Isolation of genomic DNA from streptomycetes and plasmid DNA from E. coli were carried out using standard protocols Kieser et al. 2000. Restriction enzymes and molecular biology reagents were used according to the recommendation of suppliers (NEB, MBI Fermentas, Promega). DIG DNA labeling and Southern hybridization analyses were performed according to the DIG DNA labeling and detection kit (Roche Applied Science).

A 4.3 kb BamHI fragment carrying the entire simReg1 gene and its flanking regions was cloned from 5JH10 (Table 1) into pUC19 to yield pUCsimR1 with an unique BsaAI site within the coding region of the simReg1 gene. The plasmid pUCsimR1 was digested with BsaAI and ligated to the hygromycin resistance cassette hyg, retrieved as an EcoRV fragment from pHYG1 (Table 1). The resulting plasmid pUCsimR1-hyg was digested with BamHI and the fragment containing the simReg1::hyg mutant allele was cloned into the shuttle vector pKC1139 to yield pKCsimR1-hyg.

The gene disruption plasmid pKCsimR1-hyg was conjugally transferred from E. coli into S. antibioticus Tü6040. Exconjugants were selected for resistance to apramycin (10 μg ml^-1). To generate S. antibioticus ΔsimReg1 strain, single-crossover mutants were obtained by cultivation of the respective exconjugants at 39°C for 3 days with a further screen for the loss of apramycin resistance as a consequence of a secondary crossover.

The simReg1 gene with flanking regions was retrieved from the plasmid pKCEsimR1 Rebets et al. 2008 as a 2.3 kb BamHI fragment and cloned into the BamHI sites of pSET152 to yield pSsimR1. A 1.4 kb SmaI fragment harboring only simReg1 with its promoter region was retrieved from pSsimR1 and cloned into EcoRV linearized pSET152 to yield pSsimR1-1.

A 0.5 kb DNA fragment, containing promoter of the simD4 gene (P_D4) was amplified from the chromosome S. antibioticus Tü6040 using primers simD4_for_script and simD4_rev_script (Table 2). The PCR product was digested with XbaI/KpnI and cloned into the respective sites of pGUS Myronovskyi et al. 2011, giving pSimD4script. In this plasmid transcription of the gusA gene is under the control P_SD4 promoter.

For measurement of GusA activity, mycelium of the S. lividans strain harboring both pSimD4script and pUWLsimReg1 plasmids, the control strains S. lividans 1326 × pSimD4script, S. lividans 1326 × pGUS, and S. lividans 1326 × pGUS/pUWLsimReg1 were grown in liquid TSB medium (100 ml) for 2 days at 30°C in a rotary shaker (180 rpm). 1 ml of the pre-culture was inoculated into liquid TSB medium (100 ml) and grown for 5 days at 30°C in a rotary shaker. Mycelium was harvested, washed with distilled water, then resuspended in lysis buffer (50 mM phosphate buffer [pH 7.0], 0.1% Triton X-100, 5 mM DTT, 4 mg ml^-1 lysozyme) and incubated for 30 min at 37°C. Lysates were centrifuged for 10 min at 5000 rpm. Then, 0.5 ml of lysate was mixed with 0.5 ml of dilution buffer (50 mM phosphate buffer [pH 7.0], 5 mM DTT, 0.1% Triton X-100) supplemented with 5 μl 0.2 M p-nitrophenyl-β-D-glucuronide and used for measuring optical density at λ = 415 nm every minute during 20 min of incubation at 37°C. As a reference, a 1:1 mixture of lysate and dilution buffer was used.

Streptomyces strains were grown in liquid TSB medium (50 ml) for 2 days at 30°C in a rotary shaker (180 rpm). Five ml of the pre-cultures were inoculated into liquid NL5 medium (100 ml) and the cultures were grown for 5 days at 30°C in a rotary shaker. The culture broths were extracted three times with 100 ml of ethyl acetate. The extracts were dried in vacuum and dissolved in methanol (200-400 μl). The metabolites were analyzed by high-pressure liquid chromatography-mass spectrometry (HPLC-MS) Schimana et al. 2001. 10 ml of each culture were taken and lyophilized. The dry weight of each sample was measured. In all cases amounts of antibiotic were referred back to equal amounts of biomass (dry weight) and are mean values from at least three independent experiments.

The codon-optimized copy of the simReg1 gene, named simReg1s, was synthesized by Mr. GENE Company (Heidelberg, Germany) and was provided on the plasmid pMA-simR1. Gene simReg1s was amplified from pMA-simR1 using primers SSR1F and SSR1R (Table 2). The PCR product was cloned into the pET21d NcoI-EcoRI sites, giving pETSR1c-15.

E. coli BL21(DE3) (pLysS) harboring the pETSR1c-15 plasmid was grown overnight at 37°C. LB (400 mL) containing 50 μg/mL of ampicillin was inoculated with 2 mL of the overnight culture and incubated at 21°C until the OD_{600 nm} reached 0.7. SimReg1 expression was induced with 1 mM IPTG. After incubation for an additional 16 h, the cells were harvested by centrifugation and washed with ice-cold column buffer (20 mM Tris-HCl [pH 8.0], 50 mM NaCl). Cell lysis and purification of SimReg1 with His-tag-binding resins were performed according to Novagen instructions. SimReg1 was eluted with column buffer containing 200 mM imidazole. The purest fractions (as determined by SDS-PAGE and Coomassie blue staining) were pooled, washed with storage buffer (50 mM potassium phosphate [pH 8.0], 300 mM NaCl, 10% glycerol), concentrated using Amicon Ultra (Millipore). Aliquots of SimReg1 fusion protein in storage buffer were stored at - 80°C, or used immediately in DNA-binding assays.

DNA fragments containing putative promoters of simD4 (P_D4, 513 bp), simReg1 (P_R1, 490 bp), simD3 (P_D3, 300 bp), simX4 (P_X4, 350 bp), simA7 (P_A7, 300 bp), simEx2 (P_Ex2, 550 bp), simB7 (P_SR3, 319 bp), simX (P_SEx1, 280 bp), simR (P_SR2, 300 bp), and the putative promoter region between simX and simR genes (P_R2Ex, 780 bp) (Figure 2) were used in EMSA. Indicated promoter regions were amplified from the chromosomal DNA of S. antibioticus using primer pairs listed in Table 2. Each EMSA contained 50 ng of a target DNA and 0.9 μg, 1.8 μg, 2.7 μg, 3.6 μg, 4.5 μg of the His-SimReg1 protein in a total volume of 20 μL in a binding buffer (20 mM Tris HCl [pH 8.0], 1 mM EDTA, 1 mM DTT, 100 mM KCl, 1 mM MgCl₂, 10% glycerol). After incubation for 25 min at room temperature, protein-bound and free DNA were separated by electrophoresis at 4°C on a 4.5% nondenaturing polyacrylamide gel in 0.5 × TBE buffer. The gel was stained with ethidium bromide and analyzed using a UV-imaging system (Fluorochem 5330). A negative control assay was carried out in the presence of the part of the simD4 coding region, amplified with the use of primers D4For and D4Rev (Table 2). Extracts from the strain S. antibioticus Tü6040 × pSsimR1-1, containing more then 95% of simocyclinones (Additional file 1), dissolved in methanol (5% and 10% - final volume in a reaction mixture) were tested as SimReg1 ligands.

The putative product of the simReg1 gene is a 251 aa protein with a molecular mass 27.94 kDa. As evident from BLAST and CDD search results, putative amino acid sequence of the protein has significant similarity to response regulators in two component control systems. The closest homologues of SimReg1 are proteins that act as positive regulators for angucycline-like biosynthesis, including JadR1 from S. venezuelae (60% similarity) Wang et al. 2009, LanI from S. cyanogenus (58% similarity) Rebets et al. 2008 and LndI from S. globisporus (58% similarity) Rebets et al. 2003; Rebets et al. 2005. Analysis of the SimReg1 amino acid sequence using ExPASy Proteomics Server http://expasy.org revealed a putative signal receiver domain (the REC domain, aa 15-123) located at the N-terminal part of the protein and a DNA-binding domain at the C-terminus (aa 167-239). The latter is predicted to interact with short conserved regions of the target DNA and with the RNA polymerase. The secondary structure of the C-terminal DNA-binding domain of SimReg1 was similar to that of OmpR (E. coli) and PhoB (E. coli), which adopt a winged helix-turn-helix (HTH) moiety. In the REC domain of the regulatory protein PhoB, six conserved amino acid residues are believed to be vital for phosphorylation and consequence response Sola-Landa et al. 2003; Wang et al. 2009; Dyer and Dahlquist 2006, but only three of them are present in SimReg1 (Figure 3). Also, no protein kinase encoding genes have been found within the sim cluster. Thus, we suppose that SimReg1 belongs to "atypical" response regulators (ARR), like its close homolog JadR1 Wang et al. 2009.

In order to investigate the function of simReg1, the chromosomal copy of the gene was replaced by the mutant allele containing a hygromycin resistance cassette (hyg) (Figure 4a). Inactivation of the simReg1 gene was proven by Southern hybridization. BamHI digested chromosomal DNA of the wild type and S. antibioticus ΔsimReg1 strains were probed with the DIG-labeled fragment containing simReg1, obtained as a KpnI fragment from the plasmid pKCEsimR1 Rebets et al. 2008. A single hybridization signal of the expected size (4.3 kb) was detected in the case of the wild type strain and a 6.3 kb fragment was detected in the ΔsimReg1 mutant (Figure 4b). The S. antibioticus ΔsimReg1 mutant had growth and morphological characteristics identical to those of the wild type. HPLC and TLC analysis (Figure 5a) of the extracts from the mutant strain ΔsimReg1 revealed no simocyclinone and its precursors, indicating that this gene is essential for antibiotic production.

To exclude any possibility of polar effects and to confirm that the cessation of simocyclinone production was caused by the inactivation of the simReg1, complementation experiment was carried. For this purpose, we used the pSSimR1-1 plasmid (Table 1), which contains the simReg1 gene under its own promoter cloned in the integrative vector pSET152. This plasmid was transferred into S. antibioticus wild type strain by means of conjugation. The recombinant strain S. antibioticus ΔsimReg1 × pSSimR1-1 was found to accumulate simocyclinone at a level comparable to those of the wild type (Figure 5b).

It is known that very often overexpression of the positive pathway-specific regulators lead to overproduction of antibiotics (Bibb 2005; Novakova et al., 2011). To analyze the effect of additional copies of simReg1 gene on simocyclinone biosynthesis, we introduced the plasmid pSsimR1-1 that contains simReg1 gene under its own promoter, into the wild type strain. Recombinant strain S. antibioticus Tü6040 × pSSimR1-1 produced in average 2.5 times more simocyclinone then the wild type.

In order to prove the DNA binding activity of SimReg1, gel mobility-shift assays were carried out. His-SimReg1 was purified (Additional file 2) and an in vitro binding assay was performed using His-SimReg1 and DNA fragments containing putative promoters of the regulator gene simReg1 (P_R1), the 3-keto-acyl-reductase gene simD4 (P_D4), the oxygenase gene simA7 (P_A7), the transporter gene simEx2 (P_Ex2), the 3-keto-acyl-reductase gene simD3 (P_D3), the putative gene simX4 (P_X4), the putative olivosyltransferase gene simB7 (P_SR3), and the intergenic region between simR and the transporter gene simEx1 (hereafter simX) (P_R2Ex) (Figure 2). Shifted bands were detected using the promoter regions of the enzyme encoding genes (Figure 6a, c, d, f), the transporter gene simEx2 (Figure 6g) and the regulatory gene simReg3, which is likely co-transcribed with the genes simB7, simB5, simB4, simX5 and simX7 (Figure 6h). Furthermore, DNA retardation occurred (Figure 6b) when the promoter of the simReg1 gene was used in the binding assay, indicating that SimReg1 is an autoregulatory protein. We carried out a set of control assays to demonstrate the specificity of the SimReg1 binding. For instance, none of the compounds in the crude extract of E. coli BL21(DE3) binds to any of the putative promoters (data not shown). We also showed that randomly chosen DNA did not interact with SimReg1 (Additional file 3).

SimReg1 was found to bind to the DNA fragment containing the simR/simX intergenic region (Figure 6e). However, it was not known whether SimReg1 interacts with the promoters of both genes. A 67 bp fragment located in front of the start codon of simR (P_SR2) and a 69 bp fragment located in front of simX (P_SEx1) (Figure 7a) were used for additional EMSA analysis. No binding was identified with the P_SR2 promoter, whereas DNA retardation occurred when the P_SEx1 promoter was used in the assay (Figure 7b). These results indicate that SimReg1 is capable of binding to the promoter region of simX.

Since DNA binding ability of JadR1, which also belongs to ARR and is very similar to SimReg1 (60% similarity), is regulated by jadomycin B Wang et al. 2009, we studied the effects of simocyclinone extracts from the S. antibioticus Tü6040 × pSSimR1-1 on the DNA binding activity of SimReg1. For this purpose the culture broth of S. antibioticus Tü6040 × pSSimR1 strain grown for 72 hours was extracted with an ethyl acetate, dried and dissolved in methanol. In overall the percentage of different types of simocyclinone in such an extract was more than 95% (Additional file 1). Presence of these extracts could dissociate His-SimReg1 from the promoter regions P_R1 and P_A7, as a result no shifted bands occurred (Figure 8). This effect was not due to methanol, the simocyclinone D8 solvent, as equivalent amounts of methanol had no effect on His-SimReg1-DNA complex formation (Figure 8).

On the basis of the gene inactivation, overexpression and EMSA results we suppose that SimReg1 is a positive regulator of simocyclinone production. To investigate whether SimReg1 can activate the expression of the structural genes under heterologous conditions, a reporter system on the basis of gusA was used. For these purpose, we constructed two plasmids pSimD4script and pUWLsimReg1 (Table 1). In the first plasmid the promoter region of the putative ketoreductase gene simD4 (P_D4) was fused with the coding sequence of the gusA gene. As a result expression of the reporter gusA is controlled by P_D4. In the plasmid pUWLsimReg1 intact gene simReg1 was cloned under the control of erythromycin resistance gene promoter to make the expression of the regulatory gene constitutive. As it is evident from the EMSA analysis SimReg1 binds to the promoter of the gene simD4 (Figure 6a) this means that SimReg1 should influence expression from this promoter. To verify this assumption, both plasmids were transferred into heterologous host S. lividans 1326 to avoid influence of two other regulatory proteins SimR and SimReg3 Trefzer et al., 2002. We obtained two strains: S. lividans harboring only pSimD4script and S. lividans harboring both plasmids pSimD4script and pUWLsimReg1. As a negative control we used strains: S. lividans 1326 × pGUS to show that there is no GusA activity when gusA gene contains no promoter and S. lividans 1326 harboring both plasmids pGUS (Table 1) and pUWLsimReg1 to demonstrate that SimReg1 specifically binds only to simD4 promoter region and that SimReg1 can't influence gusA expression in the absence of this promoter. Aforementioned four strains were grown in liquid TSB medium for 5 days and samples of the strains were used for GusA activity measurement as described in Materials and Methods. In the control strains the activity of GusA was approximately 0.25 ± 0.06 (Figure 9). In the case of the S. lividans strain that contains gusA gene under P_D4 activity was in average 3.3 ± 0.24 (Figure 9). In the strain containing both gusA gene under P_D4 promoter and the SimReg1 protein the activity was 6.25 ± 0.43 (Figure 9). It is in overall two times higher than without the protein. On the basis of these results, we may conclude that SimReg1 binds to the simD4 promoter region.