New targets of TetR‐type regulator SLCG_2919 for controlling lincomycin biosynthesis in Streptomyces lincolnensis

The transcription factor (TF)‐mediated regulatory network controlling lincomycin production in Streptomyces lincolnensis is yet to be fully elucidated despite several types of associated TFs having been reported. SLCG_2919, a tetracycline repressor (TetR)‐type regulator, was the first TF to be characterized outside the lincomycin biosynthetic cluster to directly suppress the lincomycin biosynthesis in S. lincolnensis. In this study, improved genomic systematic evolution of ligands by exponential enrichment (gSELEX), an in vitro technique, was adopted to capture additional SLCG_2919‐targeted sequences harboring the promoter regions of SLCG_6675, SLCG_4123‐4124, SLCG_6579, and SLCG_0139‐0140. The four DNA fragments were confirmed by electrophoretic mobility shift assays (EMSAs). Reverse‐transcription quantitative polymerase chain reaction (RT‐qPCR) showed that the corresponding target genes SLCG_6675 (anthranilate synthase), SLCG_0139 (LysR family transcriptional regulator), SLCG_0140 (beta‐lactamase), SLCG_6579 (cytochrome P450), SLCG_4123 (bifunctional DNA primase/polymerase), and SLCG_4124 (magnesium or magnesium‐dependent protein phosphatase) in ΔSLCGL_2919 were differentially increased by 3.3‐, 4.2‐, 3.2‐, 2.5‐, 4.6‐, and 2.2‐fold relative to those in the parental strain S. lincolnensis LCGL. Furthermore, the individual inactivation of these target genes in LCGL reduced the lincomycin yield to varying degrees. This investigation expands on the known DNA targets of SLCG_2919 to control lincomycin production and lays the foundation for improving industrial lincomycin yields via genetic engineering of this regulatory network.


| INTRODUCTION
Actinomycetes are a group of unicellular filamentous bacteria, which are an important source of antibiotic drugs and encode a wide variety of transcription factors (TFs) that finely control the biosynthesis of secondary metabolites [1].Typically, these TFs include global or pleiotropic regulators and cluster-situated regulators (CSRs) which exert diverse regulatory functions [2].Although 70% of the known natural antibiotics, such as lincomycin, spinosad, and avermectins are produced by actinomycetes and have important application values in the fields of medicine and agriculture [3], these secondary metabolites still exhibit low industrial fermentation yields.Rewiring the regulatory network of TFs in actinomycetes has manifested great potential for the industrial overproduction of antibiotics [4].Therefore, the definition of these complicated regulatory mechanisms will help overcome the bottleneck of low outputs of secondary metabolites.
Lincomycin, a lincosamide antibiotic produced by Streptomyces lincolnensis, is generally effective against gram-positive bacteria [5].It is widely used in clinics because of its favorable side effect profile and strong ability to penetrate tissues and cells [5].The lincomycin biosynthetic (lmb) gene cluster, spanning over 35 kb of DNA, contains 25 structural genes, two resistance genes (lmrA, lmrB), and lmrC with dual characteristics in resistance and regulation, and one CSR gene lmbU [6].In recent years, great progress has been made in research on the lincomycin biosynthetic pathway [7], while investigation of the regulatory mechanism of lincomycin biosynthesis has been relatively slow [8,9].In addition to the CSR LmbU [8], several regulatory TFs outside the lmb cluster, such as BldD, AdpA, and SLCG_Lrp have been confirmed as having central roles in the control of lincomycin production.BldD is a positive regulator with direct control over lincomycin biosynthesis in S. lincolnensis [10].AdpA has been implicated as being central to a cascade regulation of lincomycin biosynthesis by AdpA lin , LmbU, and BldA [11].We found that an Lrp family regulator SLCG_Lrp in S. lincolnensis not only could promote lincomycin biosynthesis by directly activating lmb genes but also promote nitrate assimilation and inhibit fatty acid metabolism [12].A comparative genomics approach revealed that large fragment deletion and gene mutation, especially that of regulatory genes, are crucial for lincomycin overproduction [13].These studies not only highlight the complexity of lincomycin biosynthetic regulation but also seem to provide a viable way to upgrade the industrial yield of lincomycin by engineering the regulatory system of S. lincolnensis.
The tetracycline repressor (TetR) family transcriptional regulator (TFR) plays an important role in the secondary metabolism of actinomycetes [14].Various TFRs inside or outside antibiotic biosynthetic clusters in actinomycetes regulate the biosynthesis of secondary metabolites by binding to the promoter regions of their target genes [15][16][17].We first identified the TFR SLCG_2919 in S. lincolnensis and demonstrated that it could modulate lincomycin biosynthesis by repressing the majority of the genes within lmb cluster [16].However, the SLCG_2919-mediated regulatory network's role during lincomycin biosynthesis remains to be elucidated.
Genomic systematic evolution of ligands by exponential enrichment (gSELEX) is a method derived from SELEX in which DNA fragments from a genomic library interact with a protein are screened out through successive rounds of binding, partitioning, and amplification [18].We previously improved the gSELEX method and used it to capture DNA targets of BldD in Saccharopolyspora erythraea [19].In this study, using this method, we identified four additional SLCG_2919targeted sequences.Furthermore, the aforementioned target genes were individually inactivated in LCGL and were confirmed to be associated with lincomycin production.This project lays the foundation for understanding the regulatory network of lincomycin biosynthesis in S. lincolnensis.

| Strains, plasmids, and cultivation conditions
All strains and plasmids are listed in Table 1.Escherichia coli strains were cultured at 37°C in Luria Bertani (LB) medium and supplemented with antibiotics as required [22].S. lincolnensis strains were cultured at 30°C with shaking at 220 rpm in a liquid TSBY medium for genomic DNA extraction [23].S. lincolnensis strains were grown on an MGM medium for spore production [16].

| Improved genomic SELEX method (gSELEX)
The gSELEX workflow was performed according to the method described by Wu et al. [19].Briefly, the genomic DNA of S. lincolnensis was digested with Sau3AI to generate DNA fragments of suitable size (<1 kb).Then, 2 μg digested DNA fragments were mixed with 2 μM purified SLCG_2919 for 30 min at 20°C in a 20 μL binding buffer (10 mM Tris-HCl, pH 8), (60 mM KCl, 50 mM ethylene diaminetetraacetic acid [EDTA], 5 mM MgCl 2 , 10 mM dithiothreitol [DTT], and 10% v/v glycerol).Potential SLCG_2919-binding DNA fragments were ligated using adapter primers (Supporting Information: Table S1) and amplified by polymerase chain reaction (PCR).The enriched DNA fragments were cloned into the pGEM-T vector for sequencing.BLAST was used to locate these sequences in the S. lincolnensis genome.
T A B L E 1 Bacterial strains and plasmids used in this study.

DH5α
After incubation of the mixture at 30°C for 15 min, the samples were separated on 6% native polyacrylamide gel electrophoresis gels in ice-cold 1× tris-acetate-EDTA buffer at 50 mA for approximately 50 min.

| Reverse-transcription quantitative polymerase chain reaction (RT-qPCR) assay
The relative transcriptional levels of SLCG_6675, SLCG_4123, SLCG_4124, SLCG_6579, SLCG_0139, and SLCG_0140 were analyzed by RT-qPCR with the specific primers listed in Supporting Information: Table S1.An RNA extraction/purification kit (Transgen) was used to isolate total RNA from S. lincolnensis LCGL and its derivatives after 24 h of growth in a liquid fermentation medium and quantified using a microplate reader (BioTek).Isolated RNA (500 ng) was treated with DNase I (Vazyme), and reverse transcription was performed using a complementary DNA (cDNA) synthesis kit (Vazyme).RT-qPCR was exerted on the Applied Biosystems QuantStudio 6 Flex system with Maxima™ SYBR Green/ROX qPCR Master Mix (Vazyme).
Melting curve was inserted ramping from 65°C to 95°C (increment 0.5°C/5 s) to verify the specificity of primer amplification based on the presence of a single and sharp peak.Each experiment was carried out with three independent biological replicates and three experimental replicates.The rpoD gene in S. lincolnensis acted as an internal control, and relative transcription was quantified using the comparative cycle threshold method [25].

| Construction of gene deletion mutants in S. lincolnensis
To construct the SLCG_6675 deletion mutant ΔSLCGL_6675, a 1.4-kb HindIII/XbaI fragment of the left flanking region was amplified using primer pairs (Supporting Information: Table S1), and a 1.5-kb EcoRI/KpnI fragment of the right flanking region was PCR-amplified with primer pairs 6675-P1/P2 and 6675-P3/P4 (Supporting Information: Table S1).These two fragments were ligated at the corresponding sites of pUCTSR [26] to generate the plasmid pUCTSRΔ6675.Then, a 4.3 kb DNA fragment was digested with EcoRI/HindIII from pUCTSRΔ6675, and ligated into the same sites of pKC1139, generating pKC1139Δ6675.Using polyethylene glycol-mediated protoplast transformation, pKC1139Δ6675 was introduced into LCGL.Through chromosomal homologous recombination, a 720-nt fragment within SLCG_6675 was replaced by the thiostrepton resistance gene (tsr) in LCGL.The desired ΔSLCGL_6675 mutant was confirmed using PCR with the primers 6675-P5 and 6675-P6 (Supporting Information: Table S1).

| Fermentation and lincomycin A (Lin-A) determination
Flask fermentation of S. lincolnensis strains was performed as described in our previous study [16].Briefly, spores of S. lincolnensis and its derivatives were inoculated into 30 mL of seed medium at 30°C on an orbital shaker at 240 rpm for 2 days.Then, 2 mL of the seed culture was transferred into 30 mL industrial fermentation medium and was incubated at 30°C with 240 rpm for 7 days.Lin-A production was measured as described previously [16].

| Statistical analysis
All acquired data are shown as mean ± SD and were analyzed using Student's t test.The differences were considered statistically significant at *p < 0.05, **p < 0.01 and ***p < 0.001.

| gSELEX-based exploiting of potentially new targets of SLCG_2919
To establish the SLCG_2919-mediated regulatory network, we employed the previously improved gSELEX method to identify target sequences of SLCG_2919.Genomic DNA of S. lincolnensis LCGL was initially digested with Sau3AI to generate <1 kb DNA fragments (Supporting Information: Figure S1A).After two rounds of SELEX, DNA fragments potentially binding to SLCG_2919 were cloned into the T vector for sequencing (Supporting Information: Figure S1B).When the SLCG_2919 concentration was 2 μM, four independent clones were obtained and sequenced (Supporting Information: Table S2).DNA segments corresponding to the promoter regions of SLCG_6675, SLCG_4123-4124, SLCG_6579, and SLCG_0139-0140 were captured at least twice.According to the genome annotation of S. lincolnensis, SLCG_6675, SLCG_0139, SLCG_0140, SLCG_6579, SLCG_ 4123, and SLCG_4124 encode anthranilate synthase, LysR family TFR, beta-lactamase, cytochrome P450, bifunctional DNA primase/polymerase, and magnesium or magnesiumdependent protein phosphatase, respectively.In summary, these results imply that SLCG_2919 participates in the control of primary metabolism, transcriptional regulation, oxidoreduction, and anabolism for lincomycin biosynthesis.

| SLCG_2919 binds specifically to four captured DNA fragments
To identify the four DNA fragments potentially interacting with SLCG_2919, we expressed His 6 -tagged SLCG_2919 in E. coli BL21 (DE3), as previously described [16], and tested its DNA-binding affinity for the promoter regions of  SLCG_6675, SLCG_4123-4124, SLCG_0139-0140, and SLCG_6579.Because the length of P 4123-4124 is 711 nt, we designed two DNA fragments P 4123-4124-1 (356 nt) and P 4123- 4124-2 (355 nt), without overlapping sequences.Results from the EMSAs showed notable shifts when different amounts of His 6 -SLCG_2919 were added.Then, 50-fold unlabeled probes or poly dIdC were individually added to the reaction system to evaluate the binding specificities.We found that these labeled probes were able to pull down shifted bands, whereas poly dIdC were not (Figure 1a).This indicated that SLCG_2919 binds specifically to the above DNA fragments.
Using the defined AT-rich binding site of SLCG_2919 [16], we scanned the AT-rich motif in the genome of S. lincolnensis LC-G using PREDetector (http://www.montefiore.ulg.ac.be/~hiard/PreDetector/PreDetector.php).Five DNA fragments were predicted within the promoter regions of the target genes (Figure 1b).In addition, we used the motif-finding program MEME (http://meme-suite.org/) to evaluate the predicted sequences of targeted genes and a conserved SLCG_2919-binding motif (TcaTTc, c: A, C, T; a: A, C) was identified (Figure 1b).

| SLCG_2919 disruption increases the transcription of its target genes
To confirm whether SLCG_2919 transcriptionally regulates the above-mentioned target genes, we used RT-qPCR to compare the transcription of these genes between LCGL and ΔSLCGL_2919.The results showed that the transcriptional levels of SLCG_6675, SLCG_0139, SLCG_0140, SLCG_6579, SLCG_4123, and SLCG_4124 increased by 3.3-, 4.2-, 3.2-, 2.5-, 4.6-, and 2.2-fold, respectively, in ΔSLCGL_2919 compared with those in LCGL (Figure 2).It could be concluded that SLCG_2919 acts as a negative regulator to modulate the transcriptional expression of the six new targeted genes.

| New targets of SLCG_2919 positively correlate with lincomycin yield
To investigate the relevance of the aforementioned target genes to lincomycin production, SLCG_6675, SLCG_0139, SLCG_0140, SLCG_6579, SLCG_4123, and SLCG_4124 were individually disrupted with tsr replacement in LCGL, and the corresponding mutants were obtained and confirmed by PCR analyses (Figure 3a,b).As shown in Figure 3c, the mutants ΔSLCGL_6675, ΔSLCGL_0139, ΔSLCGL_0140, ΔSLCGL_6579, ΔSLCGL_4123, and ΔSLCGL_4124 displayed 24%, 20%, 10%, 26%, 24%, and 26% reductions, respectively, in Lin-A yield compared with the parental strain LCGL.Thus, our findings indicated that the six new target genes of SLCG_2919 had a positive effect on lincomycin production.

| DISCUSSION
TFRs, which are widely present in prokaryotes, are generally involved in highly regulated biosynthesis such as that of multidrug efflux pumps, antibiotic biosynthesis, and osmotic stress response proteins [14].We previously reported that SLCG_2919 was associated with lincomycin production by directly repressing most genes within the lmb cluster [16].However, knowledge of the SLCG_2919-mediated regulatory network in lincomycin biosynthesis remains limited.In the present study, we utilized the previously improved gSELEX technique, combined with EMSA, RT-qPCR, and gene inactivation, to facilitate the identification of new targets of SLCG_2919 in S. lincolnensis.
gSELEX is an in vitro strategy for the direct capture of DNA-binding TFs; however, it is not very efficient when applied to identify targets from the high-GC actinomycete genome [19].We previously improved this method by circumventing genomic library construction and enriching TF targets with PCR in each cycle of SELEX [19].In this study, improved gSELEX was utilized to screen the target sequences of SLCG_2919.RT-qPCR and EMSA results showed that SLCG_2919 negatively regulated all the targeted genes, including SLCG_6675, SLCG_0139, SLCG_0140, SLCG_6579, SLCG_4123, and SLCG_4124.We found that the targets of SLCG_2919 have an AT-rich binding site, similar to the AT-rich motif previously defined by DNase I footprinting assays and EMSAs [16].However, there were no intergenic regions amenable to interacting with this motif among the targets of SLCG_2919 isolated in this study from the lmb cluster.We suspect that this might be due to restriction enzyme digestion or the experimental conditions.Further experiments are required to optimize this methodology and identify these intergenic regions for further clarification.
The six identified gene targets were demonstrated to be positively correlated with lincomycin production in that their knockout resulted in a substantial reduction in Lin-A yield.SLCG_6675, encoding anthranilate synthase, can transform chorismate into anthranilic acid [27].SLCG_4123, a bifunctional DNA primase/polymerase, consumes excess ATP for DNA synthesis.These two enzymes might play a significant role in the primary metabolism of S. lincolnensis.SLCG_0140 encodes a betalactamase, and its inactivation in S. lincolnensis might lower antibiotic resistance, possibly resulting in adverse effects on lincomycin biosynthesis.SLCG_6579, which encodes cytochrome P450, plays a key role in the reductive activation of molecular oxygen, and is possibly involved in the biosynthesis of lincomycin [28].SLCG_4124, a manganese-or magnesium-dependent protein phosphatase, can transfer ATP or GTP phosphate groups to amino acid residues.The LysR family TFR FkbR1 is a positive regulator of ascomycin production [29].In our study, the deletion of SLCG_0139, encoding a LysR family TFR, also decreased lincomycin yield in S. lincolnensis.Together with previously determined targets, we have proposed a general framework of SLCG_2919-mediated regulatory network (Figure 4).SLCG_2919 may participate in primary metabolism, transcriptional regulation, oxidoreduction, and anabolism, illustrating the importance of TFR in the control of the complex regulatory network of lincomycin biosynthesis.

AUTHOR CONTRIBUTIONS
two 1.6-kbfragments, the upstream and downstream regions of SLCG_4123 This study pUCTSRΔ4124 pUCTSR derivative containing two 2.0-kb fragments, the upstream and downstream regions of SLCG_4124 This study pUCTSRΔ6579 pUCTSR derivative containing two 1.5-kb fragments, the upstream and downstream regions of SLCG_6579 This study pUCTSRΔ6675 pUCTSR derivative containing two 1.5-kb fragments, the upstream and downstream regions of SLCG_6675 This study pKC1139 ori (pSG5), aac(3)IV, lacZ

F I G U R E 1
Confirmation of SLCG_2919 binding to its captured sequences.(a) Electrophoretic mobility shift assays (EMSAs) of His 6 -SLCG_2919 with promoter regions of SLCG_6675, SLCG_0139-0140, SLCG_6579, and SLCG_4123-4124.Each lane contained 100 ng DNA probes.S, unlabeled specific probe (50-fold) was added; N, nonspecific probe poly dIdC (50-fold) was added.(b) SLCG_2919-binding motif within the five target sequences.The standard code of the Weblogo server is shown at the bottom using online MEME software (http://meme.nbcr.net/meme/).

F I G U R E 4
Regulatory network of SLCG_2919 for modulating lincomycin biosynthesis in Streptomyces lincolnensis.Flat-headed arrows indicate repression.Thick red line represents SLCG_2919 controlling all promoters within lmb cluster.