Gene sco1417 encodes a positive regulator of the de novo biosynthesis of pyridoxal 5ʹ-phosphate (vitamin B6) in Streptomyces coelicolor M145

Abstract

Background

Actinomycetes of the genus Streptomyces are renowned for their highly developed and diverse specialized metaboliс pathways, and there is an extensive body of data on their specific and pleiotropic levels of regulation. Much less is known about routes leading to essential metabolites in this genus. In this work, we focused on elucidating the function of the highly conserved across Streptomyces gene SCO1417 for GntR type regulator in the model strain S. coelicolor A3(2).

Results

Combining the results of knockout and promoter probe experiments, we show that the gene SCO1417 controls pyridoxal 5ʹ-phosphate (PLP; vitamin B6) biosynthesis, and thus is a member of the PdxR group of the transcriptional regulators. The Sco1417 protein is a transcriptional repressor of its gene and an activator of the expression of the PLP synthase genes, SCO1523 (pdxS) and SCO1522 (pdxT). According to electrophoretic mobility shift assays, out of several tested B6 vitamers, only PLP served as a Sco1417 effector molecule. We also provide data on the location of the Sco1417 binding site within the promoter region of pdxST.

Conclusions

Our work portrays for the first time an evidence-based picture of the genetic control of vitamin B6 biosynthesis in S. coelicolor M145. Given the high conservancy and synteny of pdx homologs in the other streptomycetes, we suggest that the described genetic circuit is a general feature for the entire genus.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12866-025-04108-y.

Introduction

Species within the actinomycete genus Streptomyces are known for their sizable genomes harboring numerous biosynthetic gene clusters for specialized metabolism [1, 2]. There is much interest in studying these gene clusters and regulating their expression, as it will help to overproduce practically valuable small molecules and discover novel ones [3]. Specialized metabolism is intimately connected to the primary metabolic pathways serving as a source of energy, cofactors, and initial building blocks for small molecule biosynthesis [4–6]. A detailed knowledge of primary metabolism is therefore needed if one aims to obtain a comprehensive overview of the biosynthesis of specialized metabolites.

During the studies of genes for GntR family of transcriptional factors (TFs) in Streptomyces we revealed 12 TFs whose genetic determinants are invariably present in the genomes of this genus [7]. We denote these genes as “core GntRs” to underscore our suggestion about an essential role for these genes in Streptomyces physiology or morphogenesis. One core GntR gene in the model strain S. coelicolor A3(2), SCO1417, falls into the MocR subfamily of TFs. The MocR members are notable for their a two-domain architecture with an N-terminal helix-turn-helix motif (HTH), a hallmark of DNA-binding proteins, and an aspartate aminotransferase-like domain (AAT) at the C-terminus. The latter serves as an element to recognize pyridoxal phosphate (PLP; vitamin B6) and other amino compounds, and thus modulates the activity of HTH [8]. SCO1417 is similar to the pdxR gene of S. venezuelae ISP5230, which has been shown to complement PLP auxotrophy in S. lividans TK24 [9]. In pairwise alignment, Sco1417 exhibits around 40% similarity to PdxR (cg0897) of Corynebacterium glutamicum ATCC13032, the TF involved in the upregulation of adjacent PLP synthase genes pdxST for de novo biosynthesis of PLP [10]. Hence, it is likely that Sco1417 is a member of a group of PdxR TFs that control the production of PLP from pentose and triose sugars (the pdxST-dependent pathway), as found in most eubacteria. However, the SCO1417 gene is not physically linked to pdxST homologs, unlike the situation in C. glutamicum. The exact mechanism of action of Sco1417 and the scope of effector molecules it recognizes (e.g., different B6 vitamers – pyridoxine, pyridoxamine, pyridoxal, and their phosphates) remain unknown. We therefore carried out a set of experiments to clarify the function of SCO1417. Our data confirm the involvement of SCO1417 in PLP biosynthesis as a positive regulator of expression of pdxS (SCO1523) and pdxT (SCO1522) homologs. We also provide information on the ligand preferences of the Sco1417 protein and the nucleotide sequence of its operator site. As SCO1417 orthologs are omnipresent in Streptomyces genomes, it is reasonable to expect that de novo vitamin B6 biosynthesis in all streptomycetes will follow the logic described here for S. coelicolor A3(2). This fills one more gap in our understanding of essential metabolism of Streptomyces.

Materials and methods

Bacterial strains, plasmids and culture conditions

Bacterial strains and plasmids used in this work are listed in Table 1. E. coli strains were used as described in the Table 1 and respective references. The Streptomyces strains were grown at 28 °C and E. coli strains were grown at 37 °C in Luria-Bertani (LB) medium [11], unless otherwise stated. Soya flour agar medium SFM [12] was used to obtain sporulating Streptomyces lawns and to plate Streptomyces – E. coli conjugation mixtures. Vitamin B6 auxotrophic phenotype of generated mutants was screened on minimal medium MM [12]. Appropriate antibiotics were added to growth media where needed. PLP and hydrochlorides of pyridoxal (PL), pyridoxine (PN) and pyridoxamine (PM) were purchased from Sigma, of highest possible purity.

Table 1.

Bacterial strains and plasmids used in this work

Strain or plasmid	Description	Source or reference
S. coelicolor M145	SCP1–, SCP2– derivative of A3(2)	[11]
S. coelicolor pdxR::Am	SCO1417 replaced with apramycin resistance gene (Amr)	This work
S. coelicolor ΔpdxR	pdxR::Am strain with excised Amr	This work
S. coelicolor ΔpdxR pIJ-pdxR+	ΔpdxR complemented with SCO1417	This work
S. coelicolor ΔpdxR pIJ82+	ΔpdxR carrying empty pIJ82	This work
S. coelicolor pdxST::Am	SCO152-1523 replaced with Amr	This work
S. coelicolor ΔpdxST	pdxST::Am strain with excised Amr	This work
S. coelicolor ΔpdxST pSET-pdxST	ΔpdxST complemented with SCO1522-1523	This work
S. coelicolor ΔpdxST pSET152+	ΔpdxST with empty vector	This work
S. coelicolor pGUS+	M145 carrying pGUS	This work
S. coelicolor pGUS-pdxRp+	M145 carrying pGUS-pdxRp	This work
S. coelicolor ΔpdxR pGUS-pdxRp+	ΔpdxR carrying pGUS-pdxRp	This work
S. coelicolor pGUS-pdxSTp+	M145 carrying pGUS-pdxSTp	This work
Escherichia coli DH5α	Routine cloning host	Life Technologies
E. coli ET12567 pUZ8002+	Host for conjugative DNA transfer	[11]
E. coli BW25113 pIJ790+	Host for recombineering experiments	[12]
E. coli BL21 (DE3) STAR pRARE	Strain for recombinant protein production. The pRARE plasmid was isolated from Rosetta™(DE3) (Novagen) and introduced into BL21 (DE3) STAR™ (Thermo Fisher Scientific)	This work
pIJ790	ts-plasmid carrying genes for λ-RED recombination, chloramphenicol resistant (Cmr)	[12]
St6D7	Cosmid containing pdxR genomic region, kanamycin resistance (Kmr)	JIC
pIJ774	Carrying oriT-acc(3)IV flanked by loxP-sites, Amr and ampicillin resistance (Apr)	[7]
pLERECJ	Carrying acc(3)IV flanked by loxP-sites, Amr and Apr	Prof. A. Luzhetskyy
pUWLCre	Replicative plasmid (reppIJ101) harboring recombinase gene cre under ermEp	Prof. A. Luzhetskyy
pIJ82	Integrative φC31-based vector, hygromycin resistance (Hyr)	[11]
pKC0702	pKC1139 derivative, Amr replaced with Hyr marker	Prof. A. Luzhetskyy
pSET152	Integrative φC31-based vector, Amr	[11]
St6D7-pdxR::Am	Cosmid with inactivated pdxR; pdxR::aac(3)IV (Kmr and Amr)	This work
pIJ-pdxR	pIJ82 harboring pdxR with promoter region; Hyr	This work
pKC0702-pdxST	pKC0702 harboring pdxST with 2-kb flanking region, Hyr	This work
pKC0702-pdxST::Am	pdxST knockout construct; pdxST::aac(3)IV (Hyr and Amr)	This work
pSET-pdxST	pSET152 harboring pdxST with promoter region, Amr	This work
pGUS	The gusA-based reporter plasmid, Amr	[13]
pGUS-pdxRp	pGUS with pdxRp-gusA fusion, Amr	This work
pGUS-pdxSTp	pGUS with pdxSTp-gusA fusion, Amr	This work
pET28a	Vector for histidine-tagged protein production, pET-system	Novagen
pET28a-pdxR	PdxR-6His protein production	This work

For the gene transcription profiling, S. coelicolor wild type and pdxR null mutant were grown in TSB [12] medium for 36 h. Then, the cells were collected by centrifugation, washed three times with water, and inoculated in SMM [12] supplemented with 20 µM PL. The cells were then allowed to grow for another 36 h. To investigate the influence of PL and PLP on the transcriptional activity of the pdxR and pdxST promoters in S. coelicolor M145, strains were grown likewise in SMM with 20 µM PL and PLP. To study the auto-regulatory function of PdxR, strains were cultivated in TSB medium for 36 h.

Plasmid and strain construction

Routine molecular biology techniques (DNA isolation, transformation, intergeneric matings, PCR, etc.) were carried out as described in [11]. Molecular biology enzymes were used according to manufacturer instructions. The oligonucleotides used in this work are listed in Table 2. All plasmids were verified by DNA sequencing.

Table 2.

Primers used in this work

Primer name	Sequence	Purpose/PCR product
pdxR_acc_f	GTCAACCATGGCCAATTTGGGGAAGGTGGACTGATTTCCATGATTCCGGGGATCCGTCGACC	To amplify aac(3)IV for pdxR replacement
pdxR_acc_r	GCGGGCGGAGGTCCCCCGGCGGCCCCGCCGGACGCGCCCTATGTAGGCTGGAGCTGCTTC
pdxST_f	AATAAAAGCTTACAAGACGTCTGTCATGACA	To amplify pdxST with 3-kb flanks
pdxST_r	AATAATCTAGACCAGTTCACTTGATTCTCCT
pdxST_acc_f	CGTTGCTGCACAACCCCCCTTCTCCCCAGTGAGGTACCCGTGGATATCTCTAGATACCG	To amplify aac(3)IV for pdxST replacement
pdxST_acc_r	CTTCGCGTAACCCATCTCCTGCAACGAACGCAGGAATCCTAAACAAAAGCTGGAGCTC
pdxR-cmpl-f	ATTGAAAGGATCCCATCGACAGCCCCGTGTGCT	To clone pdxR for complementation
pdxR-cmpl-r	ATTGAAATCTAGAGCGCCCTAGGCCACGAACGT
pdxST-cmpl-f	AATAAGGATCCCTGCGCGAGCACCGCAAGGT	To clone pdxST for complementation
pdxST-cmpl-r	AATAATCTAGAATCTGCCTGTCTCCTTCGCG
Gus-R-f	TCTGAATCTAGATTCAGGACGATGCCCGCCAC	pdxRp for pGUS
Gus-R-r	TCTGAAGGTACCCCAAATTGGCCATGGTTGAC
Gus-ST-f	TCTGAATCTAGAAGGAACTCGCCACCTCCGGC	pdxSTp for pGUS
Gus-ST-r	TCTGAAGGTACCACTGGGGAGAAGGGGGGTTG
RT_1523_f	ACATGATCGAGGGCATCATC	To check the trans-cription of pdxS
RT_1523_r	ATCTCGTTCTTGATCTGGCG
RThrdB_f	CGAGGACGAGGCGACCGAGGAG	Positive RT control (hrdB)
RThrdB_r	CAGCTTGTCCTCGGCGAACAGA
SCO1417NcoI-f	AAAACCATGGCGCAGTGGACCTCGGCCGTG	pdxR-ORF for protein production
SCO1417BamHIhis-r	AAAGGATCCTCAGTGGTGGTGGTGGTGGTGGGCCACGAACGTCCGCG
R-f	CTGGTGCAGCACGTTCCA	pdxR promoter for EMSAs
R-r	ACTGCGCCATGGAAATCAGT
ST-f	CCGCCGCTTCCACAACGA	pdxST promoter for EMSA
ST-r	TGCTGGACACGGGTACCTCA
OTR1-f	TGGGGTGGTGGTAGTGGACTGCCACATGACACCCTGTGGCTTCAGAA	To generate ds-oligonucleotides with predicted rPdxR binding sites (bs) in pdxRp
OTR1-r	TTCTGAAGCCACAGGGTGTCATGTGGCAGTCCACTACCACCACCCCA
OTR2-f	TCAACCATGGCCAATTTGGGGAAGGTGGACTGATTT
OTR2-r	AAATCAGTCCACCTTCCCCAAATTGGCCATGGTTGA
OTST-f	ACGAAAACGATCCACCGCCTTCTCATTGGCCCTTGCTGTGGCCTGCT	To generate ds-oligonucleotides carrying putative rPdxR bs in pdxSTp
OTST-r	AGCAGGCCACAGCAAGGGCCAATGAGAAGGCGGTGGATCGTTTTCGT
OTST2-f	CTTCTCATTGGCCCTTGCTGTGGCCTGCTC
OTST2-r	GAGCAGGCCACAGCAAGGGCCAATGAGAAG
OTR1a-f	TGGGGTGGTGGTAAAAAACTGCCACATGACACCCTGTGGCTTCAGAA	To generate ds-oligonucleotides with mutated rPdxR bs in pdxRp
OTR1a-r	TTCTGAAGCCACAGGGTGTCATGTGGCAGTTTTTTACCACCACCCCA
OTR1b-f	TGGGGTGGTGGTAGTGGACTGCCACATGACACCCTAAAACTTCAGAA
OTR1b-r	TTCTGAAGTTTTAGGGTGTCATGTGGCAGTCCACTACCACCACCCCA
OTR1c-f	TGGGGTGGTGGTAAAAAACTGCCACATGACACCCTAAAACTTCAGAA
OTR1c-r	TTCTGAAGTTTTAGGGTGTCATGTGGCAGTTTTTTACCACCACCCCA
OTR2a-f	TCAACCATGGCCAATTTGGGGAAGAAAAACTGATTT
OTR2a-r	AAATCAGTTTTTCTTCCCCAAATTGGCCATGGTTGA
OTR2b-f	TCAACCAAAACCAATTTGGGGAAGGTGGACTGATTT
OTR2b-r	AAATCAGTCCACCTTCCCCAAATTGGTTTTGGTTGA
OTR2c-f	TCAACCAAAACCAATTTGGGGAAGAAAAACTGATTT
OTR2c-r	AAATCAGTTTTTCTTCCCCAAATTGGTTTTGGTTGA
OTST2a-f	CTTCTCATTGGCCCTTGCTAAAACCTGCTC	To generate ds-oligonucleotides carrying matated rPdxR bs in pdxSTp
OTST2a-r	GAGCAGGTTTTAGCAAGGGCCAATGAGAAG
OTST2b-f	CTTCTCAAAAACCCTTGCTGTGGCCTGCTC
OTST2b-r	GAGCAGGCCACAGCAAGGGTTTTTGAGAAG
OTST2c-f	CTTCTCAAAAACCCTTGCTAAAACCTGCTC
OTST2c-r	GAGCAGGTTTTAGCAAGGGTTTTTGAGAAG
Pnat-f	CAGGTGATCGTTACGGAGCGTAGAGAACCTATGTCCCTGAGTGACGATACTGG	To generate unspecific oligonucleotide probe
Pnat-r	CCAGTATCGTCACTCAGGGACATAGGTTCTCTACGCTCCGTAACGATCACCTG

An in-frame deletion mutant, S. coelicolor ΔpdxR, was constructed using recombineering approach [12]. The cosmid St6D7, which carries the pdxR gene, was introduced into E. coli strain BW25113 pIJ790⁺. The aac(3)IV cassette containing oriT was amplified from pIJ774 using primers pdxR_acc_f/pdxR_acc_r with specific homology extension to pdxR. The cassette was then electroporated into arabinose-induced BW25113 pIJ790⁺ St6D7⁺. The final construct, St6D7-pdxR::Am, was selected on plates supplemented with kanamycin and apramycin. The plasmid was conjugally transferred from E. coli ET12567 pUZ8002⁺ into S. coelicolor M145. The double crossover mutant was selected for apramycin resistance and kanamycin sensitivity, and confirmed by PCR.

To inactivate pdxST, both genes along with flanking regions (approximately 2 kb long) were amplified from S. coelicolor chromosome with primers pdxST-f/pdxST-r. The PCR product was digested with HindIII and XbaI, purified, and ligated into the respective sites of pKC0702. The resulting plasmid pKC0702-pdxST was transformed into E. coli strain BW25113 pIJ790⁺, thereby facilitating the replacement of pdxST with an apramycin resistance cassette. The aac(3)IV cassette with pdxST-specific extensions was amplified from pLERECJ using primers pdxST_acc_f/pdxST_acc_r. Subsequently, the plasmid pKC0702-pdxST::Am was introduced into S. coelicolor M145 by intergeneric conjugation with E. coli ET12567 pUZ8002⁺, which harbored the respective construct. To generate single-crossover Am^r Hy^r mutants, initial exconjugants were incubated at 40 °C for 3 days (a nonpermissive temperature for the pSG5 replication of pKC0702), and then screened for an apramycin resistance and hygromycin sensitivity (an indicative of vector loss and double crossover).

The apramycin resistance cassettes for pdxR and pdxST replacement are flanked with loxP sites. The marker genes were evicted from the genomes of S. coelicolor pdxR::Am and S. coelicolor pdxST::Am through the expression of a site-specific recombinase cre from a plasmid pUWLCre. The pdxR and pdxST knockouts and marker excisions were confirmed via PCR (primers pdxR_cmpl-f/r, pdxST_ cmpl-f/r).

The coding sequences of pdxR and pdxST along with the 300-/225-bp promoter regions were amplified from chromosomal DNA using primers pdxR-cmpl-f/r and pdxST-cmpl-f/r, respectively. The PCR products were digested with XbaI and BamHI and cloned into the respective sites of integrative plasmids pIJ82 and pSET152, resulting in pIJ-pdxR and pSET-pdxST.

The promoter regions of pdxR (pdxRp; 472 bp) and pdxST (pdxSTp; 500 bp) were amplified with primers Gus-R-f/Gus-R-r and Gus-ST-f/Gus-ST-r, respectively, and then digested and cloned into the XbaI and KpnI sites of the pGUS vector, giving pGUS-pdxRp and pGUS-pdxSTp.

RT-PCR

The RNA used for semiquantitative RT-PCR was isolated as described in [13]. Subsequently, 200 ng of cDNA were used as a template for the PCR analysis using primers listed in Table 2.

Analysis of β-glucuronidase activity

The β-glucuronidase activity of the strains carrying the reporter plasmid was assayed according to [14, 15].

Production and purification of His-tagged Sco1417 (rPdxR)

The coding sequence of pdxR was cloned into the pET28a expression vector (primers SCO1417NcoI-f and SCO1417BamHIhis-r) to give pET28a-pdxR. To produce the C-terminally hexahistidine-tagged PdxR protein (rPdxR), E. coli BL21 (DE3) STAR pRARE carrying pET28a-pdxR was grown in LB medium supplemented with chloramphenicol and kanamycin until an OD₆₀₀ of 0.5 was reached. The culture was then induced with 0.25 mM IPTG (isopropyl thiogalactoside) and incubated for 18 h at 20°C. The cells were collected by centrifugation and resuspended in a lysis buffer (50 mM TrisHCl, 300 mM NaCl, 1 mM DTT, 5% glycerol, 20 mM imidazole, pH 7.5) containing a protease inhibitor cocktail (Roche). The lysis of the cells was achieved by two consecutive passages through a French press (American Instrument Corporation) at 1000 psi. The cell lysate was centrifuged at 16000 rpm for 30 min and the soluble fraction was applied to a Ni-NTA agarose resin (Qiagen), washed two times with a wash buffer (50 mM TrisHCl, 300 mM NaCl, 1 mM DTT, 5% glycerol, 50 mM imidazole, pH 7.5). The protein was eluted with increasing concentrations of imidazole (100–200 mM) and subsequently dialyzed against a storage buffer (50 mM TrisHCl, 300 mM NaCl, 1 mM DTT, 5% glycerol, pH 7.5). The concentration of rPdxR was determined by Bradford assay.

Electrophoretic mobility shift assay (EMSA) of DNA-protein complexes

The putative promoter regions pdxRp and pdxSTp were amplified from the S. coelicolor genome with the primers listed in Table 2. The PCR products were labelled at 5’-end with [γ-³³P]-ATP using T4 polynucleotide kinase. The labeled probes (20 fmol) were incubated with purified rPdxR in concentrations ranging from 0.05 to 1.00 µM, in binding buffer (20 mM Tris–HCl pH 7.9, 100 mM KCl, 10 mM MgCl₂, 1 mM EDTA, 1 mM DTT, 12.5% glycerol supplemented with 1 µg poly(dI-dC) and 4.5 µg BSA)) for 20 min at 30°C. The DNA-protein complexes were resolved on 8% nondenaturing polyacrylamide gels in 0.5xTBE buffer at 150 V. DNA bands were visualized by phosphor imaging on Typhoon FLA7000. For competition assays, an unlabelled DNA probe was added into the reaction mixture in 10- to 500-fold molar excess. Full-length unedited versions of all EMSA gels present in this work are given in Supplementary Materials, Figs. S7-S10.

To probe different B6 vitamers as potential ligands of rPdxR, they were incubated with the protein (50 nM) for 5 min at 30 °C prior to the addition of pdxRp or pdxSTp probes. The concentrations of vitamin B6 vitamers that were tested are indicated in the main text and in additional information Fig. S4.

To identify the putative binding sites of rPdxR within the pdxR and pdxST promoters, a number of double-stranded oligonucleotides R1, R2, ST, ST2 and their derivatives (mutated versions), were generated as described in [16] and used in EMSA.

Results

Regulatory and structural genes involved in PLP biosynthesis in S. coelicolor A3(2)

The putative translation product of the SCO1417 gene is a typical TF of the MocR subfamily and harbors the all expected motifs involved in DNA and ligand recognition. In particular, when compared to the structurally studied MocR-type TF GabR involved in regulation of genes for γ-aminobutyric acid degradation in Bacillus [8], Sco1417 retains amino acid residues important for DNA binding (correspond to Arg43 and Ser152 within the HTH motif of GabR) and PLP recognition (Ser311, Lys312 and Arg319 within the AAT domain; Fig. S1). The SCO1417 in A3(2) genome is flanked by genes that do not appear to be related to vitamin B6 metabolism (e.g. its closest neighbors encode membrane proteins of unknown function). Using the pdxS and pdxT genes of C. glutamicum as a queries in BLAST searches, we readily identified their orthologs, SCO1523 and SCO1522, respectively, about a hundred genes away from SCO1417. We note here that disjoint location of pdxR and the genes it controls has been described for some bacteria, e.g. Streptococcus [17]. The coding sequences of the aforementioned genes are separated by a 12-bp gap, an indicative of their transcriptional coupling. BLASTP analysis revealed that Streptomyces genomes invariably encode orthologs of Sco1523 and Sco1522. Overall, our initial scrutiny of available data supports the hypothesis that Sco1417 is PdxR involved in regulation of the expression of genes for the production of PLP from ribose 5-phosphate, glyceraldehyde 3-phosphate and glutamine (Fig. 1).

Fig. 1.

PLP biosynthesis in S. coelicolor A3(2): genes (A) and the proposed pathway (B). The gene SCO1417 (pdxR) is located separately (gap marked with dots) from the pdxST genes (SCO1523, SCO1522). Bidirectional arrows point to orthologous genes in S. coelicolor and C. glutamicum. Dashed rectangle (bottom of the figure) marks known (unphosphorylated) B6 vitamers. SCO genes colored in grey are likely not related to vitamin B6 metabolism

Inactivation of SCO1417 leads to vitamin B6 auxotrophy and impaired transcription of the PLP synthase genes in S. coelicolor

To corroborate the regulatory function of Sco1417 (designated PdxR) in PLP biosynthesis, we constructed a knockout strain S. coelicolor ∆pdxR. First, the coding sequence of the gene on a chromosome of S. coelicolor M145 was replaced by an apramycin resistance cassette and further excised to generate a marker-free strain. In comparison to the wild type, the resulting mutant was unable to grow on minimal medium unless vitamin B6 vitamers were supplemented (PM, PN, PL or PLP were tested, Fig. 2A). In trans expression of pdxR gene from its own promoter in S. coelicolor ∆pdxR fully complemented the vitamin B6 auxotrophic phenotype, restoring the ability of the strain to grow on minimal medium.

Fig. 2.

The gene SCO1417 (pdxR) is involved in the regulation of the de novo biosynthesis of PLP in S. coelicolor M145. Growth of wild type (M145), pdxR (A) and pdxST (B) null mutants. Photos of the plates were taken after 120 h of growth on MM or MM supplemented with 10 µM pyridoxal. Introduction of an intact copy of pdxR or pdxST into the respective mutants restored their ability to grow on MM compared to mutants carrying empty vectors pIJ82 and pSET152. Knockout of pdxR abrogated the transcription of pdxST genes (C). The hrdB gene served as a positive control for sqRT-PCR. The cDNA wt, cDNA prepared from total RNA of M145; cDNA mut, cDNA of ∆pdxR; DNA – chromosomal DNA isolated from M145

Next, we wanted to confirm that SCO1523 (pdxS) and SCO1522 (pdxT) encode subunits of PdxST-complex that catalyzes de novo synthesis of PLP. For this purpose, S. coelicolor ∆pdxST mutant was generated and characterized by growth assay on minimal medium. As shown in Fig. 2B, the mutant did not grow on minimal medium without B6 vitamers, whereas the wild type and the complemented mutant grew abundantly. This concludes that pdxST encode a functional PLP synthase complex that is solely responsible for PLP production in S. coelicolor M145. In addition, utilization of different vitamers indicates a functional salvage pathway in the strain.

The inability of S. coelicolor ∆pdxR to grow on MM was the first evidence that PdxR is a positive regulator of pdxST transcription. To confirm this hypothesis, the transcriptional profile of pdxST genes was compared in M145 and ∆pdxR grown in MM supplemented with pyridoxal. As shown in Fig. 2C, no pdxST transcript was detected in the mutant compared to the wild-type strain. Altogether, our data show that ∆pdxR is indeed impaired in the gene encoding a TF involved in the regulation of PLP biosynthesis, and its absence abolishes the expression of pdxST genes.

Properties of Sco1417: ligand preferences and negative regulation of its own gene

Recombinant C-terminally 6His-tagged Sco1417 (rPdxR) was purified to homogeneity from E. coli (Fig. S2) and subjected to a series of EMSAs. First, rPdxR was shown to bind promoter regions of its own gene (pdxRp) and that of pdxST (pdxSTp; Fig. 3A, B and Fig. S3), directly confirming its role as a TF for these genes. Second, we revealed that, in vitro, only PLP was able to interact with rPdxR and prevent its binding to the promoter regions of pdxR and pdxST genes (Fig. 3C, D), while PL, PN and PM did not impact rPdxR-DNA interaction (Fig. S4).

Fig. 3.

EMSA confirms the interaction of rPdxR with the pdxR (A) and pdxST (B) promoter regions. The first lanes in gels A and B (marked as 0) correspond to free labeled DNA. PLP prevents the binding of rPdxR to pdxRp (C) and pdxSTp (D). Assay conditions: rPdxR (50 nM) was pre-incubated with increasing concentrations of PLP for 5 min; then, a labeled DNA probe was added, and reactions were run for 20 min. C⁻, labeled DNA without rPdxR and PLP; C⁺, DNA + rPdxR

To assess the biological consequence of the presence of vitamin B6 on the expression of pdxR and pdxST, we employed plasmids where the gusA reporter gene [14] was fused to either pdxRp or pdxSTp. The plasmids were introduced into M145 and glucuronidase (GusA) activity was measured in minimal medium in either the absence or presence of PL and PLP. Both vitamers had no influence on the transcription from the pdxR promoter, while their presence led to a strong (almost threefold) decrease in GusA activity from the pdxST promoter (Fig. 4A). The results suggest that PdxR represses the transcription of its own gene and upregulates pdxST genes. The MocR-like TFs are known to act as auto-repressors. To find an evidence of negative regulation of pdxR by PdxR, we compared the GusA activity of the pdxRp-gusA transcriptional fusion in ∆pdxR and parental strain M145. Indeed, the aforementioned reporter plasmid exhibited higher activity in the mutant than in the wild-type strain (Fig. 4B), in accord with auto-repressing function of PdxR.

Fig. 4.

PL and PLP act as in vivo effectors of PdxR which is a repressor of its own gene and an activator of pdxST genes. Glucuronidase (GusA) activity of cell lysates prepared from M145 strain grown in minimal medium for 36 h (A). Lanes: pGUS. M145 harboring empty gusA vector pGUS; pdxRp, M145 harboring pdxRp-gusA fusion in the absence of vitamers; pdxRp-PL, M145 harboring pdxRp-gusA fusion in the presence of PL; pdxRp-PLP, M145 harboring pdxRp-gusA fusion in the presence of PLP; pdxSpT, M145 harboring pdxSTp-gusA fusion in the absence of vitamers; pdxSTp-PL, M145 harboring pdxSTp-gusA fusion in the presence of PL; pdxSTp-PLP, M145 harboring pdxSTp-gusA fusion in the presence of PLP. Comparison of GusA activity of M145 (WT) and ∆pdxR strains harboring pdxRp-gusA fusion (B). Error bars, ±2SD (three independent biological replicates)

Mapping of the PdxR binding site within the pdxR and pdxST promoters

Multiple alignment of pdxRp with sequences found upstream of the pdxR orthologs in the other streptomycetes identified two adjacent regions, referred to as R1 and R2, carrying putative binding motifs of PdxR (Fig. S5). R2 is located immediately upstream of the pdxR start codon (Fig. 5A). The double-stranded oligomers containing one of the two sequences were used in EMSA. Both R1 and R2 are bound by rPdxR (Fig. 5). Close examination of the R1 and R2 sequences led to the identification of conserved 6-nt direct repeats, the substitution of which by the hexa-A sequence perturbed rPdxR binding (Fig. 5A-C). Complete cessation of DNA binding was only observed when both repeats were mutated.

Fig. 5.

The PdxR binding sites within the promoter of the pdxR gene. The nucleotide sequence of the promoter (A) comprised two conserved motifs, R1 and R2 (marked with bold line), each containing 6-nt imperfect repeats, underlined and marked with a and b. The start codon of pdxR is shown on green background. EMSA of rPdxR binding to the native and mutated (substituted with AAAAAA) versions of R1 (B) and R2 (C). Labels at the bottom of the gels: R1, native sequence, R1a, mutated repeat a of R1; R1b, mutated repeat b of R1; R1a + b – both a and b are mutated within R1. The versions of R2 are labeled analogously. Pnat, unspecific DNA (ssfg_02181). The first lane in each gel is free DNA, grey triangles mark increasing amounts of rPdxR (0.1, 0.25, 0.5 and 1.0 µM)

Analysis of the pdxSTp region also revealed a highly conserved sequence approximately 50 bp upstream of the pdxS start codon (Fig. S6), referred to as ST (Fig. 6A). Both ST and its subfragment ST2 were bound by rPdxR, whereas the mutation to tetra-A either of the two 4-nt repeats within ST2 completely blocked rPdxR binding to DNA (Fig. 6B).

Fig. 6.

The PdxR binding sites within the promoter of the pdxS gene. The nucleotide sequence of the promoter (A) contains the conserved motif ST, a subfragment of which, ST2 (marked with bold line) was further tested in a series of EMSAs. ST2 harbors 4-nt imperfect repeats, underlined and marked as a and b. The start codon of pdxS is shown on the green background. EMSA of rPdxR binding to ST2, ST and mutated (substituted with AAAAAA) versions of ST2 (B). Labels at the bottom of the gels: ST2a, mutated repeat a of ST2; ST2b, mutated repeat b of ST2; ST2a + b – both a and b are mutated within ST2. The first lane in each gel is free DNA, grey triangles mark increasing amounts of rPdxR (0.1, 0.25, 0.5 and 1.0 µM)

Discussion

This work provides experimental evidence that the gene SCO1417 in S. coelicolor A3(2) encodes the PdxR protein involved in the regulation of the de novo biosynthesis of PLP. Similar to other PdxR-governed mechanisms investigated in Gram⁺ bacteria, this transcriptional regulator controls the expression of genes SCO1523-SCO1522 encoding subunits of the PLP synthase complex (PdxST), whose proposed biochemical functions are summarized in Fig. 1. Therefore, we propose the names pdxR and pdxST for SCO1417 and SCO1523-1522, respectively, by analogy to the studied precedents.

Our current understanding of PdxR-controlled circuitry is summarized in Fig. 7 and detailed below. PdxR exerts a positive regulatory control on the pdxST transcription while negatively regulating the expression of its own gene in S. coelicolor. The differences in the biological effects of PdxR on the expression of pdxST and pdxR genes is likely underlined by the dissimilarity in the relative location of PdxR binding sites within the pdxR and pdxST promoters, as well as their sequence identity. The MocR-like regulators act as homodimers with a head-to-tail arrangement that allows them to interact with direct DNA repeats. However, the binding of the ligand alters their DNA-binding preferences towards inverted repeats. The PdxR binding sites, which were identified in the pdxR-ST intergeneric region of C. glutamicum and B. clausii, consist of two direct and one inverted repeats. Likewise, detailed analysis of the pdxR promoter identified at least four conserved imperfect direct hexa-nucleotide repeats RTGGVY possibly recognized by PdxR in S. coelicolor. According to the mutational analysis, two motifs R1b (GTGGCT) and R2a (GTGGAC) are indeed important for the DNA-binding activity of PdxR. An additional inverted repeat (motif 3) was identified ten nucleotides upstream of the motif 2, and might be required for binding by holo-PdxR. In the pdxR promoter region, the − 10 and − 35 regulatory sequences are located between PdxR recognition sites. Such an arrangement most likely prevents the binding of the RNA-polymerase to pdxRp irrespective of the interaction of PLP with PdxR, thus explaining no differences in the gusA expression from the promoter region of pdxR. Contrary to C. glutamicum and B. clausii, where the expression of pdxR and pdxST is governed by the binding of PdxR to the same motifs, in S.coelicolor the regulatory and the biosynthetic genes are transcribed from their own promoters. In contrast to pdxRp, in silico analysis of pdxSTp did not revealed any prominent candidates for the PdxR recognition. Introduction of mutations into two conserved 4-bp sequences in pdxSTp blocked PdxR binding in EMSA, indicating their importance for the interaction. However, additional experimental analyses are required to identify the regulatory sequences. We note here that the structure of the repeats recognized by PdxR from the probiotic bacterium Bacillus clausii [8] differs from the sites revealed in our study, suggesting diversity in the PdxR mechanism. Similarly, the role of PLP in the nuances of PdxR function in S. coelicolor, as recently reported for B. clausii PdxR [18], awaits further investigation.

Fig. 7.

Proposed regulatory mechanism governed by PdxR. In B. clausii (A), apo-PdxR preferably binds to the direct repeats (motif1 and motif2) exerting a positive and negative regulatory control on the transcription of the pdxST (green arrow lines) and pdxR (orange lines) genes, respectively. Holo-PdxR binds to the inverted motifs 1 and 3, and represses the expression of both genes (orange dashed lines). Similarly, in S. coelicolor (B), PdxR binds to the predicted DNA-binding sites (motifs 1–3) in the pdxR promoter region. Sequences (R1a, R1b, R2a and R2b) tested in EMSA are underlined, and the − 10 and − 35 RNA-polymerase binding hexamers in the nucleotide sequence of pdxRp are highlighted in grey. The predicted pdxR translational start codon (ATG) is indicated by + 1. Motifs 1 (M1) and 2 (M2) correspond to the boxed sequences R2a and R1b, whereas motif 3 (M3) represents the boxed TGCCAC inverted repeat. Apo-PdxR is shown as two grey ovals, holo-PdxR – as grey ovals with yellow dots

Under in vitro conditions only PLP showed the ability to preclude PdxR from binding to DNA, while in vivo (promoter probe) tests showed that PL is also able to affect PdxR. We suggest that PL undergoes phosphorylation in the cells, and thus becomes a PdxR effector molecule. The feedback inhibition of the pdxST transcription tightly regulates levels of PLP, maintaining sufficient amounts of the cofactor in the cell pool but below its toxic concentrations. Indeed, vitamin B6 salvage pathways are well known for bacteria [19, 20], although their identity in Streptomyces remains unknown. Another interesting line of research that was left unaddressed in our study is the effect of the pdxR mutation on the specialized metabolism of S. coelicolor. The production of the two most prominent specialized metabolites of A3(2), the red pigment undecylprodigiosin and the deep blue actinorhodin, did not seem to be affected by the pdxR deletion when pyridoxal is abundant (data not shown). However, a rather narrow range of media was tested; it remains unknown how manipulations of vitamin B6 levels (in the ∆pdxR mutant) would impact specialized metabolism. Given the abundance of pyridoxal-dependent enzymes across tree of Life, some other important phenomena might be missed out [21]. We therefore invite further experimental scrutiny of the described here mutant in order to portray a comprehensive picture of the links between PLP biosynthesis and the other aspects of Streptomyces physiology.

Conclusions

Our results showed that the SCO1417 gene of S. coelicolor M145 encodes the TF PdxR, which is involved in the positive regulation of the expression of the genes SCO1523 (pdxS) and SCO1522 (pdxT), homologs of enzymes catalyzing the de novo formation of PLP. PdxR negatively regulates the transcription of pdxR. PLP serves as an effector molecule for PdxR under in vitro and in vivo conditions, while PL was shown to affect pdxST expression under in vivo conditions. PdxR binding sites were mapped to short stretches of the pdxR and pdxST promoters and crucial repeat sequences were pinpointed by mutagenesis and EMSAs.

Abbreviations

PLP

Pyridoxal 5ʹ–phosphate

TF

Transcriptional factor

PL

Pyridoxal

PN

Pyridoxine

PM

Pyridoxamine

HTH

Helix–turn–helix motif

AAT

Aspartate aminotransferase domain

Authors’ contributions

O.T., K.F., A.M., A.B., W.W. and B.O. conceived and designed the experiments. O.T., R.M. and E.G. performed the experiments. B.O. and O.T. wrote the paper. All authors reviewed the manuscript.

Funding

This work was supported by the Research Council of Lithuania (LMTLT) under agreement No. S-LU-24-14; by the Ministry of Education and Science of Ukraine (MESU) within the framework of the bilateral Ukrainian-Lithuanian R&D projects for the period of 2024–2025 years (M/51-2024). This work was also partially supported by grant BG-21F and BG-19F from MESU (both to B.O.). In addition, the research work was funded by Visby and DAAD fellowships awarded to O.T.

Data availability

All data are present in main text of the article and Additional file 1. Materials are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.