TetR-Type Regulator SLCG_2919 Is a Negative Regulator of Lincomycin Biosynthesis in Streptomyces lincolnensis

TetR family transcriptional regulators (TFRs) are generally found to regulate diverse cellular processes in bacteria, especially antibiotic biosynthesis in Streptomyces species. However, knowledge of their function in lincomycin biosynthesis in S. lincolnensis remains unknown. The present study provides a new insight into the regulation of lincomycin biosynthesis through a TFR, SLCG_2919, that directly modulates lincomycin production and resistance. Intriguingly, SLCG_2919 and its adjoining gene, SLCG_2920, which encodes an ATP/GTP binding protein, were extensively distributed in diverse Streptomyces species. In addition, we revealed a new TFR binding motif, in which SLCG_2919 binds to the promoter region of SLCG_2920, dependent on the intervening AT-rich sequence rather than on the flanking inverted repeats found in the binding sites of other TFRs. These insights into transcriptional regulation of lincomycin biosynthesis by SLCG_2919 will be valuable in paving the way for genetic engineering of regulatory elements in Streptomyces species to improve antibiotic production.

ABSTRACT

Lincomycin A (Lin-A) is a widely used antibacterial antibiotic fermented by Streptomyces lincolnensis. However, the transcriptional regulatory mechanisms underlying lincomycin biosynthesis have seldom been investigated. Here, we first identified a TetR family transcriptional regulator (TFR), SLCG_2919, which negatively modulates lincomycin biosynthesis in S. lincolnensis LCGL. SLCG_2919 was found to specifically bind to promoter regions of the lincomycin biosynthetic gene cluster (lin cluster), including 25 structural genes, three resistance genes, and one regulatory gene, and to inhibit the transcription of these genes, demonstrating a directly regulatory role in lincomycin biosynthesis. Furthermore, we found that SLCG_2919 was not autoregulated, but directly repressed its adjacent gene, SLCG_2920, which encodes an ATP/GTP binding protein whose overexpression increased resistance against lincomycin and Lin-A yields in S. lincolnensis. The precise SLCG_2919 binding site within the promoter region of SLCG_2920 was determined by a DNase I footprinting assay and by electrophoretic mobility shift assays (EMSAs) based on base substitution mutagenesis, with the internal 10-nucleotide (nt) AT-rich sequence (AAATTATTTA) shown to be essential for SLCG_2919 binding. Our findings indicate that SLCG_2919 is a negative regulator for controlling lincomycin biosynthesis in S. lincolnensis. The present study improves our understanding of molecular regulation for lincomycin biosynthesis.

IMPORTANCE TetR family transcriptional regulators (TFRs) are generally found to regulate diverse cellular processes in bacteria, especially antibiotic biosynthesis in Streptomyces species. However, knowledge of their function in lincomycin biosynthesis in S. lincolnensis remains unknown. The present study provides a new insight into the regulation of lincomycin biosynthesis through a TFR, SLCG_2919, that directly modulates lincomycin production and resistance. Intriguingly, SLCG_2919 and its adjoining gene, SLCG_2920, which encodes an ATP/GTP binding protein, were extensively distributed in diverse Streptomyces species. In addition, we revealed a new TFR binding motif, in which SLCG_2919 binds to the promoter region of SLCG_2920, dependent on the intervening AT-rich sequence rather than on the flanking inverted repeats found in the binding sites of other TFRs. These insights into transcriptional regulation of lincomycin biosynthesis by SLCG_2919 will be valuable in paving the way for genetic engineering of regulatory elements in Streptomyces species to improve antibiotic production.

INTRODUCTION

Streptomyces lincolnensis is a Gram-positive actinomycete that is generally utilized for the industrial-scale production of a lincosamide antibiotic, lincomycin A (Lin-A), which consists of an α-methylthiolincosaminide (MTL) and an amino acid derivative, propylproline (PPL) (1). Lin-A and its semisynthetic derivative clindamycin are used clinically for the treatment of infective diseases caused by Gram-positive bacteria (2). Clindamycin can also be used for the treatment of protozoal diseases, e.g., malaria (2). The lincomycin biosynthetic gene cluster (lin cluster) contains 29 genes that encode proteins for biosynthesis, resistance, and regulation, spanning over 35 kb of DNA in S. lincolnensis (3, 4). In recent years, genetic and biochemical strategies have been performed to explore the molecular mechanisms of lincomycin biosynthesis (5–7). Through heterologous expression, key gene inactivation, and in vitro combinatorial biosynthesis, some lincosamide derivatives with relevant bioactivities have been obtained in the engineered strains (7, 8).

Recently, Meng et al. confirmed that the ABC1 transporter gene and the lincomycin export gene, lmrA, were transcriptionally activated by the global regulator GlnR with nitrate supplementation (9). Subsequently, Hou et al. showed that LmbU from the lin cluster served as a cryptic cluster-situated regulator (CSR), promoting lincomycin production (10). Yet, the transcriptional regulatory mechanisms underlying lincomycin biosynthesis remain obscure, limiting production improvements for Lin-A.

The TetR family transcriptional regulators (TFRs) are widespread in bacteria, playing important roles in a series of diverse processes (11, 12). It was reported that TFRs bind to consensus or apparent palindromic DNA sequences and could regulate the biosynthesis of secondary metabolites in actinomycetes (11, 12). It is worthwhile to note that genetic engineering of TFR elements has been shown to increase the yields of valuable antibiotics in industrial settings (13–15). Nonetheless, distinct from research related to TFRs for the biosynthesis of avermectin and erythromycin (13, 15–19), insights into the regulatory role of TFRs in lincomycin biosynthesis have not been reported. In the genome of S. lincolnensis LCGL, a derivative of S. lincolnensis LC-G with artificial synthetic 4×attB^ΦC31 (20), we identified 123 putative TFRs based on BLAST analysis with Pfam PF00440 (TetR_N) and the genome annotation of S. lincolnensis LC-G. Through in vivo and in vitro evidence, we identify a TFR (SLCG_2919) that regulates lincomycin biosynthesis in S. lincolnensis.

RESULTS

SLCG_2919 negatively regulates lincomycin biosynthesis.

To search for potential TFRs associated with lincomycin production, we inactivated a number of TFRs in strain LCGL and successively obtained these desired mutants by thiostrepton resistance screening and PCR confirmation (Fig. 1A and B, with the currently studied SLCG_2919 strain as an example). By shake-flask fermentation and ultraperformance liquid chromatography (UPLC) analysis, we found that the ΔSLCGL_2919 mutant displayed a higher yield of Lin-A than its parental strain LCGL, with ∼25% improvement, and was selected for further investigation.

FIG 1.

SLCG_2919 negatively regulates lincomycin production in S. lincolnensis LCGL. (A) Schematic deletion of SLCG_2919 by homologous recombination in S. lincolnensis LCGL. (B) PCR confirmation of the SLCG_2919 deletion mutant by the primers 2919-P5 and 2919-P6. Lane M, 5,000-bp DNA ladder; lane 1, the positive control, 1,500 bp amplified from pKC1139-Δ2919; lane 2, the negative control, 450 bp amplified from LCGL; lane 3, 1,500 bp amplified from the ΔSLCGL_2919 mutant. (C) Lin-A production of S. lincolnensis LCGL and its derivatives. (D) Lin-A production of S. lincolnensis LA219X and ΔSLA219X_2919. Mean values of three replicates are shown, with the standard deviation indicated by error bars. *, P < 0.05; **, P < 0.01.

According to the available genome sequence of S. lincolnensis LC-G, SLCG_2919 consists of 210 amino acids with a molecular mass of approximately 24 kDa. The locations of SLCG_2919 and its adjacent genes on the chromosome are shown in Fig. 1A. SLCG_2919 is adjoined with genes encoding a putative ATP/GTP binding protein and a phosphotransferase (Fig. 1A), but they are not cotranscriptional (Fig. S1). Furthermore, complementation of the SLCG_2919 gene in the ΔSLCGL_2919 mutant showed approximate recovery of Lin-A production (Fig. 1C), suggesting that SLCG_2919 negatively regulates lincomycin biosynthesis in S. lincolnensis. To further confirm the negatively regulatory role of SLCG_2919 in lincomycin production, pIB139-2919, as well as pIB139, was introduced into strain LCGL. Results showed that the Lin-A yield of LCGL/pIB139-2919 was 15% lower than that of the control LCGL/pIB139 (P < 0.01) (Fig. 1C).

TFRs often play much broader roles, such as in antibiotic production and in morphological differentiation (21, 22), so we compared LCGL and the ΔSLCGL_2919 mutant in cell growth and morphological differentiation. The results showed that the ΔSLCGL_2919 mutant and the parental strain, LCGL, had similar growth rates in yeast-malt-glucose (YMG) culture, as measured by mycelium dry weight, and had similar sporulation rates on MGM agar medium (Fig. S2), indicating that SLCG_2919 was not involved in cell growth and morphological differentiation of S. lincolnensis.

To examine the widespread use of SLCG_2919 as a negative regulator for lincomycin biosynthesis in S. lincolnensis, SLCG_2919 was deleted in the high-yield strain S. lincolnensis LA219X (20). As expected, Lin-A production of the obtained S. lincolnensis ΔSLA219X_2919 mutant (2.89 g/liter) was 15% higher than that of LA129X (2.51 g/liter) when cultured in 30 ml of industrial fermentation medium for 7 days (Fig. 1D).

Determination of transcription units found in the lin cluster.

Peschke et al. (3) had shown eight individual transcription units (lmrA, lmrB, lmrC, lmbC, lmbD, lmbK, lmbW, and lmbA-B1-B2) in the lin cluster of S. lincolnensis. In this study, we confirmed the above eight units and determined the remaining transcription units in the lin cluster, including one monocistronic transcription unit (lmbE) and four multicistronic transcription units (lmbR-S-O-P-Z-N-M-L, lmbJ-IH-G-F, lmbV-T-Q, and lmbU-Y-X) (Fig. 2A and B). These results established a foundation for investigating the effects of regulatory factors on the lin cluster.

FIG 2.

Determination of transcription units in the lin cluster. (A) The validation of the transcription units of the lin cluster by fourteen cross-spaced primers (Table S1). Lane M, 5,000-bp DNA ladder; lanes 1 to 14, the PCR products from cDNA by fourteen cross-spaced primers. (B) Diagram of the transcription units of the lin cluster. Bent arrow, transcription site and direction. The numbers in the circles indicate locations of fourteen cross-spaced primers.

SLCG_2919 directly represses the transcription of structural genes in the lin cluster.

To investigate the regulatory mode of SLCG_2919, we assessed the influence of SLCG_2919 on the expression of structural genes found in the lin cluster by reverse transcription-quantitative PCR (qRT-PCR). The results showed that the transcriptional levels of the structural genes in the ΔSLCGL_2919 mutant, including lmbA, lmbC, lmbE, lmbG, lmbK, lmbR, lmbV, and lmbW, were 1.6-, 3.7-, 3.2-, 2.5-, 2.6-, 22.0-, 4.3-, and 2-fold higher, respectively, than those of LCGL (Fig. 3A). Only the transcriptional level of lmbD was not obviously influenced. These results indicated that SLCG_2919 had a negative effect on lincomycin biosynthesis by repressing the transcription of the structural genes.

FIG 3.

SLCG_2919 directly represses transcription of the lin cluster structural genes. (A) Quantitative transcription levels of lmbA, lmbC, lmbD, lmbE, lmbG, lmbK, lmbR, lmbV, and lmbW in LCGL and the ΔSLCGL_2919 mutant cultured for 24 h in fermentation medium. Mean values of three replicates are shown, with the standard deviation indicated by error bars. *, P < 0.05. (B) EMSAs of His₆-SLCG_2919 with the promoter regions of lmbA, lmbC-lmbD, lmbE, lmbJ-lmbK, lmbR, and lmbV-lmbW. Each lane contained 150 ng of DNA probes. S, unlabeled specific probe (20-fold) was added; N, nonspecific probe poly(dI-dC) (20-fold) was added.

To examine whether SLCG_2919 might directly regulate the transcription of structural genes in lincomycin biosynthesis, we expressed His₆-tagged SLCG_2919 in Escherichia coli BL21(DE3) (Fig. S3) and examined its affinity with the promoter regions of lmbA, lmbC-lmbD, lmbE, lmbJ-lmbK, lmbR, and lmbV-lmbW. As detected by electrophoretic mobility shift assays (EMSAs), mobility shifts were obviously detected upon the addition of different concentrations of His₆-SLCG_2919 (Fig. 3B). When 20-fold unlabeled probes were added into the reaction system, they dramatically competed with labeled probes for binding to His₆-SLCG_2919 (Fig. 3B). As a negative control, a nonspecific DNA, poly(dI-dC), was used to compete with labeled probe, and the shifted band did not disappear, thereby indicating that SLCG_2919 bound specifically to the above promoter regions. These findings revealed that SLCG_2919 directly represses lincomycin biosynthetic structural genes in S. lincolnensis.

SLCG_2919 directly represses transcription of resistance genes in the lin cluster.

In the lin cluster, there are three lincomycin resistance genes, lmrA, lmrB, and lmrC, with ImrA encoding a proton-dependent lincomycin transporter, ImrB encoding a 23S rRNA adenine(2058)-N-MTase, and ImrC encoding a member of the ABC transporter family (3, 23). We found that when SLCG_2919 was inactivated in LCGL, transcriptional levels of the three resistance genes were enhanced by 6-, 1.8-, and 2.5-fold, respectively (Fig. 4A). The results from EMSAs showed that His₆-SLCG_2919 could specifically bind to the promoter regions of lmrA, lmrB, and lmrC (Fig. 4B, C and D), indicating that SLCG_2919 also directly represses the three resistance genes in the lin cluster from S. lincolnensis.

FIG 4.

SLCG_2919 directly represses transcription of resistance genes in the lin cluster. (A) Quantitative transcriptional levels of lmrA, lmrB, and lmrC in LCGL and the ΔSLCGL_2919 mutant cultured for 24 h in fermentation medium. (B to D) EMSAs of His₆-SLCG_2919 with the promoter regions of lmrA (B), lmrB (C), and lmrC (D). Each lane contained 150 ng of DNA probes. S, unlabeled specific probe (20-fold) was added; N, nonspecific probe poly(dI-dC) (20-fold) was added. *, P < 0.05; **, P < 0.01.

SLCG_2919 directly represses transcription of the CSR gene lmbU.

The function of a positive cluster-situated regulator (CSR), LmbU, from the lincomycin biosynthetic pathway was recently described (10). To find a relationship between SLCG_2919 and LmbU, we compared transcriptional levels of lmbU between LCGL and the ΔSLCGL_2919 mutant. Results showed that lmbU was transcriptionally increased by 8-fold in the ΔSLCGL_2919 mutant in comparison to the level in LCGL (Fig. 5A). Since the distance of the intergenic region between lmbU and lmrC is 757 nucleotides (nt), we designed two DNA fragments with no overlapped sequence, P_lmbU1 (372 nt) and P_lmbU2 (385 nt), for EMSAs. The results revealed that His₆-SLCG_2919 bound to either P_lmbU1 or P_lmbU2 (Fig. 5B). Therefore, these data indicated that SLCG_2919 directly represses the expression of the regulatory gene lmbU in S. lincolnensis.

FIG 5.

SLCG_2919 directly represses transcription of the CSR gene lmbU. (A) Quantitative transcriptional level of lmbU in LCGL and the ΔSLCGL_2919 mutant cultured for 24 h in fermentation medium. (B) EMSAs of His₆-SLCG_2919 with the promoter regions of lmbU (P_lmbU1 and P_lmbU2). Each lane contained 150 ng of DNA probes. S, unlabeled specific probe (20-fold) was added; N, nonspecific probe poly(dI-dC) (20-fold) was added. **, P < 0.01.

SLCG_2919 directly represses transcription of its adjacent gene, SLCG_2920.

It was previously found that TFRs generally have significant effects on their adjacent genes (11). To investigate whether SLCG_2919 directly binds to its adjacent genes and to its own promoter regions, His₆-SLCG_2919 was individually used to bind with the promoter regions of SLCG_2918, SLCG_2919, and SLCG_2920. As shown in Fig. 6A, a gel-based band shift was detected at 50 nM His₆-SLCG_2919, with the promoter region of SLCG_2920. However, SLCG_2919 could not bind to the promoter region of SLCG_2918 or to its own promoter region, even when it reached the concentration of 2 μM (data not shown).

FIG 6.

SLCG_2919 directly represses transcription of its adjacent gene. (A) EMSA of His₆-SLCG_2919 with the promoter region of SLCG_2920. Each lane contained 150 ng of DNA probes. S, unlabeled specific probe (20-fold) was added; N, nonspecific probe poly(dI-dC) (20-fold) was added. (B) Quantitative transcriptional level of SLCG_2920 in LCGL and the ΔSLCGL_2919 mutant cultured for 24 h in fermentation medium.

To explore whether SLCG_2919 also regulates the transcription of its adjacent gene SLCG_2920, encoding a putative ATP/GTP binding protein, qRT-PCR was performed with RNA isolated from S. lincolnensis LCGL and from the ΔSLCGL_2919 mutant cultured for 24 h in fermentation medium. The transcriptional level of SLCG_2920 in the ΔSLCGL_2919 mutant exhibited a 2-fold increase compared with that in LCGL (Fig. 6B). Taken together, these results demonstrated that SLCG_2919 directly represses the transcription of SLCG_2920.

Overexpression of SLCG_2920 increases Lin-A production by improving lincomycin resistance in S. lincolnensis LCGL.

Since an ATP/GTP binding protein is likely coupled with a transmembrane protein that may be involved in antibiotic resistance (24, 25), we investigated lincomycin resistance by overexpressing SLCG_2920 in LCGL. As shown in Fig. 7A, when lincomycin concentrations reached 350 μg/ml, LCGL/pIB139-2920 could grow normally, while the growth of LCGL/pIB139 was inhibited, suggesting that overexpression of SLCG_2920 effectively improved the resistance against lincomycin in S. lincolnensis LCGL. Meanwhile, LCGL/pIB139-2920 showed 14% improvement in the production of Lin-A compared with that of LCGL/pIB139 (P < 0.05) (Fig. 7B), indicating that SLCG_2920 positively affects Lin-A production through improving lincomycin resistance.

FIG 7.

Overexpression of SLCG_2920 increases Lin-A resistance and Lin-A production in S. lincolnensis LCGL. (A) Resistance assay of LCGL/pIB139 and LCGL/pIB139-2920 strains against Lin-A. (B) Lin-A production of S. lincolnensis LCGL/pIB139 and LCGL/pIB139-2920. *, P < 0.05.

AT-rich sequence in the promoter region of SLCG_2920 is the precise site of SLCG_2919 binding.

To precisely determine the SLCG_2919 binding site in the promoter region of SLCG_2920, a DNase I footprinting assay was performed using 6-carboxyfluorescein (FAM)-labeled DNA fragments. We found that there was only one SLCG_2919 binding site, a 24-nt sequence (ACCGAGT-AAATTATTTA-ACTCGGT) that included identical 7-nt inverted repeats separated by an internal 10-nt AT-rich sequence (Fig. 8A and B). To confirm the importance of the 24-nt sequence for SLCG_2919 binding, EMSAs were performed, with SLCG_2919 binding to the original DNA fragment P1 and to mutated DNA fragment P1m without the above 24-nt sequence. In contrast to P1, P1m did not show the band shift (Fig. 8B). The binding sequence of SLCG_2919 is located in the upstream region of SLCG_2920, extending from nucleotide −17 to +6 with respect to the first nucleotide of the predicted transcriptional start site of SLCG_2920 (Fig. 8C).

FIG 8.

Analysis of the precise SLCG_2919 binding site. (A) Determination of SLCG_2919 binding site in the promoter region of SLCG_2920 by DNase I footprinting assay. Top fluorogram shows control reaction without protein. Protection regions were acquired with increasing concentrations (0.5 μM and 0.75 μM) of His₆-SLCG_2919 protein. (B) EMSAs of SLCG_2919 binding to P1 and P1m (lacking 24-nt sequence). Underlining indicates inverted repeats. (C) Nucleotide sequences of SLCG_2920 promoter region and SLCG_2919 binding site. Bigger black font, SLCG_2920 transcription start site (TSS); black dotted boxes, putative −10 and −35 regions and start codon; underlining, SLCG_2919 binding site. (D) The different base substitution mutagenesis of the binding site (24 nt). (E) EMSAs of SLCG_2919 binding to the fragment P1 (wild type [WT]) and mutated fragments P2, P3, P4, and P5.

TFRs generally form homodimeric structures and bind to consensus or apparent palindromic DNA sequences (11, 26). To distinguish the importance of the inverted repeats from the AT-rich sequence in the promoter region of SLCG_2920, we constructed DNA fragments with different base substitutions in the 24-nt sequence, with DNA fragment P2 mutated in the left inverted repeat, DNA fragment P3 in the AT-rich sequence, DNA fragment P4 in the right inverted repeat, and DNA fragment P5 in both inverted repeats (Fig. 8D). Unexpectedly, DNA fragments P2, P4, and P5 showed the same band shift as DNA fragment P1 at a 0.1 μM concentration of SLCG_2919, while DNA fragment P3 did not exhibit the binding activity (Fig. 8E). These findings demonstrated that the internal AT-rich (AAATTATTTA) sequence, but not the flanking inverted repeats, was indispensable for SLCG_2919 binding.

DISCUSSION

As a large and important family of one-component signal transduction systems, TFRs are widely associated with metabolism, antibiotic production, quorum sensing, and multidrug resistance (15, 27). In the present study, we discovered that SLCG_2919 can regulate lincomycin biosynthesis by repressing the transcription of lincomycin biosynthetic structural genes, resistance genes, and a regulatory gene in the lin cluster, as well as by repressing its adjacent gene, SLCG_2920, which encodes an ATP/GTP binding protein. We are not aware of any other TFR besides SLCG_2919 in S. lincolnensis.

Based on genome context analysis, Ahn et al. (28) classified TFRs into three groups. The first group is divergently oriented to one of its neighboring genes, which can be used to predict a regulatory relationship between the two genes. The second group is likely to be cotranscribed with its upstream or downstream neighboring gene when separated by 35 bp or less, usually known to be autoregulatory or to regulate the expression of its cotranscribed genes. The third group lacks a defined relationship with its adjacent genes. To date, little is known about the third group of TFRs. In our study, we have demonstrated that SLCG_2919 is not cotranscribed with its neighboring genes (Fig. S1), so it belongs to the third group, which directly represses transcription of its adjacent gene, SLCG_2920. This report provides the evidence for further understanding of the third group of TFRs.

SLCG_2919 is located at approximately 3.5 Mb in the chromosome of S. lincolnensis LC-G and is not closely positioned with the lin cluster (GenBank accession no. CP022744; nt 290041 to 322634, 0.3 Mb). In addition, SLCG_2919 was here proved to bind to all promoter regions of the lin cluster, implying that SLCG_2919 remotely exhibited a directly regulatory role in lincomycin biosynthesis. Although SLCG_2919 bound specifically to the promoter region of lmbC-lmbD, only lmbC was differentially transcribed upon SLCG_2919 inactivation. We suppose that an unknown regulator and SLCG_2919 might collaboratively or competitively modulate the transcription of lmbD. Recently, the CSR LmbU was shown to directly activate expression of the lmbA and lmbW genes and indirectly stimulate the expression of lmbC and lmbJ, while indirectly repressing the expression of lmbK and lmbU (10). These findings implied a more complex transcriptional regulatory mechanism of lincomycin biosynthesis than previously expected.

Members of the TetR family of regulators can repress genes whose products are involved in multidrug resistance (12). SLCG_2920, a putative ATP/GTP binding protein, is likely related to an ABC transporter that may be involved in antibiotic resistance and production (24, 25). In this study, overexpression of the SLCG_2920 gene effectively increased the resistance against lincomycin and the yield of Lin-A in S. lincolnensis. Therefore, we speculate that SLCG_2920 might be involved in exporting lincomycin along with a transmembrane protein and thus reduce the feedback inhibition on lincomycin production inside the cell. A similar phenomenon was also detected in Streptomyces chattanoogensis, in which multiple transporters were identified to be involved in the export of natamycin (29).

Besides the CSR LmbU (10), SLCG_2919 was another regulator participating in lincomycin biosynthesis. SLCG_2919 not only regulated genes in the lin cluster, but also regulated its adjacent gene, SLCG_2920, suggesting that SLCG_2919 might control Lin-A biosynthesis in multiple ways. By BLASTP analysis, we found that SLCG_2919 homologs were broadly distributed in typical antibiotic-producing Streptomyces species, such as SCO4194 from Streptomyces coelicolor A3(2) (74.9% identity), SAV_4017 from Streptomyces avermitillis MA-4680 (79.1% identity), SGR_3980 from Streptomyces griseus IFO13350 (70.6% identity), etc. (Fig. S4A). More importantly, the cassette of SLCG_2919 and its adjoining gene, SLCG_2920, which encodes an ATP/GTP binding protein, mostly exists in those Streptomyces species (Fig. S4B). So far, these TFRs have not yet been functionally investigated, potentially demonstrating a novel transcriptional regulatory paradigm for antibiotic biosynthesis in Streptomyces species.

Many TFRs generally form homodimeric structures and bind to consensus or apparent palindromic DNA sequences (26). However, we found an unusual motif in the SLCG_2919 binding sequence, in which an AT-rich sequence (AAATTATTTA), rather than a common palindromic DNA sequence, was the most important for TFR binding. In addition, PREDetector analysis (http://www.montefiore.ulg.ac.be/~hiard/PreDetector/PreDetector.php) of the AT-rich sequence (AAATTATTTA) in the genome of S. lincolnensis LC-G revealed that only partial SLCG_2919 binding sites were predicted within the lin cluster genes’ promoter regions (lmbJ-lmbK, lmbV-lmbW, lmrA, lmrB, lmrC, and lmbU) (Fig. S5). We suspect that there exist different binding models in other lin cluster promoter regions (lmbA, lmbC-lmbD, lmbE, and lmbR) for SLCG_2919 regulation, which needs to be further investigated.

Based on the present data, we propose a model for the SLCG_2919-mediated transcriptional regulatory network (Fig. 9). SLCG_2919 exerts its negative regulatory effect on lincomycin production in at least three ways, as follows: (i) directly repressing transcription of lincomycin biosynthetic structural genes, (ii) directly repressing expression of resistance genes inside or outside the lin cluster, and (iii) directly controlling the CSR gene lmbU.

FIG 9.

Proposed model of the SLCG_2919-mediated regulatory network in S. lincolnensis. SLCG_2919 exerts regulatory effects on lincomycin biosynthesis by interacting with the promoter regions of structural genes, resistance genes, or the CSR gene. Solid arrows, activation; bars, repression; solid lines, direct control; dashed lines, indirect control.

MATERIALS AND METHODS

Strains, plasmids, and growth conditions.

All strains and plasmids used in this study are listed in Table 1. E. coli was cultured in Luria Bertani (LB) medium at 37°C, supplemented with appropriate antibiotics as required, with shaking at 220 rpm (30). Liquid TSBY medium (3% tryptone soya broth, 0.5% yeast extract, and 10.3% sucrose, with/without apramycin or thiostrepton) was used for DNA extraction (31). S. lincolnensis LCGL and its derivatives were grown on MGM (2% soluble starch, 0.5% soybean flour, 0.1% KNO₃, 0.05% NaCl, 0.05% MgSO₄, 0.05% K₂HPO₄, 0.001% FeSO₄, and 2% agar, with/without apramycin or thiostrepton) for sporulation. Spores were isolated and stored in 20% glycerol at −80°C. Liquid SM medium (0.4% yeast extract, 0.4% tryptone soya broth, 1% glucose, 0.05 g/liter MgSO₄, 0.2 g/liter KH₂PO₄, and 0.4 g/liter K₂HPO₄) was used for S. lincolnensis protoplast preparation (32).

TABLE 1.

Bacterial strains and plasmids used in this study

Strain or plasmid	Description	Reference or source
Strains
E. coli
DH5α	F recA lacZM15	30
BL21(DE3)	F− ompT hsdSB(rB−mB−) dcm gal λ(DE3)	Novagen
S. lincolnensis
LC-G	CGMCC7.209, a lincomycin producer	Xinyu Pharmaceutical Co., Ltd.
LCGL	LC-G derivative with artificial integrated attBΦC31 site	20
ΔSLCGL_2919	LCGL derivative with SLCG_2919 deleted	This study
ΔSLCGL_2919/pIB139	ΔSLCGL_2919 strain carrying pIB139	This study
ΔSLCGL_2919/pIB139-2919	ΔSLCGL_2919 strain carrying pIB139-SLCG_2919	This study
LCGL/pIB139	LCGL carrying pIB139	This study
LCGL/pIB139-2919	LCGL carrying pIB139-SLCG_2919	This study
LCGL/pIB139-2920	LCGL carrying pIB139-SLCG_2920	This study
LA219	A lincomycin high-yield strain	Xinyu Pharmaceutical Co., Ltd.
LA219X	LA219 derivative with artificial integrated attBΦC31 site	20
ΔSLA219X_2919	LA219X derivative with SLCG_2919 deleted	This study
Plasmids
pUCTSR	pUC18 derivative containing a 1.36-kb fragment of a thiostrepton resistance gene in BamHI/SmaI sites	33
pUCTSRΔ2919	pUCTSR derivative containing two 1.5-kb fragments, the upstream and downstream regions of SLCG_2919	This study
pKC1139	ori (pSG5), aac(3)-IV, lacZ	34
pKC1139-Δ2919	pKC1139 derivative for SLCG_2919 deletion	This study
pIB139	ΦC31 attP-int locus, acc(3)-IV, oriT, PermE* promoter	13
pIB139-2919	pIB139 carrying an extra SLCG_2919 for gene complementation	This study
pIB139-2920	pIB139 carrying an extra SLCG_2920 for gene complementation	This study
pET28a	T7 promoter, His tag, kan	Novagen
pET28a-2919	pET28a-derived plasmid carrying SLCG_2919	This study

Gene inactivation, complementation, and overexpression.

The plasmid pKC1139-Δ2919, with an internal 450-bp deletion of SLCG_2919, was constructed in two steps. First, with the genomic DNA of LCGL as the template, two 1.7-kb fragments flanking SLCG_2919 were respectively amplified by PCR using two primer pairs, 2919-P1/2919-P2 and 2919-P3/2919-P4 (Table S1), cleaved by HindIII/XbaI and KpnI/EcoRI, and ligated into the corresponding sites of pUCTSR (33), yielding pUCTSRΔ2919. Second, the 4.8-kb DNA fragment was cleaved with EcoRI/HindIII from pUCTSRΔ2919 and ligated into the same site of pKC1139 (34), generating pKC1139-Δ2919. By polyethylene glycol (PEG)-mediated protoplast transformation, pKC1139-Δ2919 was transformed into S. lincolnensis LCGL. The transformants resistant to thiostrepton but sensitive to apramycin were selected as the ΔSLCGL_2919 double-crossover strain and confirmed by PCR amplification, using the primers 2919-P5 and 2919-P6 (Table S1).

For the complementation of SLCG_2919 in the ΔSLCGL_2919 mutant, SLCG_2919 was amplified from the genomic DNA of S. lincolnensis by PCR, using the primers 2919-P7 and 2919-P8 (Table S1). The PCR product was digested with NdeI/XbaI and inserted into the corresponding sites of pIB139, generating pIB139-2919. Then, pIB139-2919, as well as pIB139, was successively transformed into the ΔSLCGL_2919 mutant, and the ΔSLCGL_2919/pIB139-2919 complementation strain, as well as the ΔSLCGL_2919/pIB139 control strain, was obtained by apramycin resistance screening and further confirmed by PCR amplification with the primers apr-P1 and apr-P2 (Table S1). Furthermore, pIB139 and pIB139-2919 were also transformed into LCGL, generating the control strain LCGL/pIB139 and overexpressed strain LCGL/pIB139-2919, respectively.

For overexpression of SLCG_2920 in LCGL, a 2,031-bp DNA fragment containing a full-length SLCG_2920 was amplified with the primers 2920-F and 2920-R (Table S1). Then, the NdeI/XbaI-digested fragment was inserted into the corresponding sites of pIB139, and the constructed pIB139-2920 was transformed into LCGL to obtain the overexpressed strain LCGL/pIB139-2920.

In accordance with above procedures, we constructed the ΔSLA219X_2919 mutant using a high-yield strain, S. lincolnensis LA219X.

Fermentation and UPLC analysis of Lin-A production.

S. lincolnensis LCGL and its derivatives were grown on MGM for sporulation (with appropriate antibiotics for recombinant strains). A 1-ml aliquot of spore suspension (∼1 × 10⁷ CFU/ml) was inoculated into a 250-ml flask containing 30 ml of the seed medium [2% soluble starch, 1% glucose, 1% soybean flour, 3% cream corn, 0.15% (NH₄)₂SO₄, and 0.4% CaCO₃ for culture, with/without apramycin] at 30°C with shaking at 240 rpm for 2 days. A 2-ml aliquot of seed culture was transferred into 30 ml fermentation medium [10% glucose, 2% soybean flour, 0.15% cream corn, 0.8% NaNO₃, 0.5% NaCl, 0.6% (NH₄)₂SO₄, 0.03% K₂HPO₄, and 0.8% CaCO₃, with/without apramycin]. All fermentation cultures were incubated at 30°C and 240 rpm for 7 days. After fermentation, 200 μl supernatant of fermentation broth was mixed with 800 μl ethanol and centrifuged at 12,000 rpm for 10 min to remove the residue. Subsequently, lincomycin samples extracted from those liquid fermentation cultures were quantified by a Waters H13CHA 394G UPLC system on an Extend-C₁₈ column (5 μm, 150 × 4.6 mm; Agilent), which was equilibrated with 60% methyl alcohol and 40% 5 mM ammonium acetate (pH 9.0). An isocratic program was carried out at a flow rate of 0.4 ml/min. The products were monitored at 214 nm (Fig. S6).

Determination of transcription units of the lin cluster.

Total RNA was isolated from S. lincolnensis LCGL after 24 h of growth in fermentation liquid medium using an RNA extraction/purification kit (SBS), and the RNA concentration was determined using a microplate reader (BioTek). Isolated RNA (500 ng) was treated with DNase I (MBI Fermentas), and reverse transcription was performed using a cDNA synthesis kit (MBI Fermentas). Fourteen primer pairs were designed to amplify intergenic regions in the lin cluster (Table S1). The desired fragments were amplified using cDNA of LCGL as a template with the above-mentioned primer pairs.

In order to verify whether SLCG_2918, SLCG_2919, and SLCG_2920 were cotranscribed, PCR was performed by using cDNA of LCGL as a template with 2918-2919-P1/2918-2919-P2 and 2919-2920-P1/2919-2920-P2 primers (Table S1).

Transcriptional analysis by qRT-PCR.

The relative transcriptional levels of structural genes (lmbA, lmbC, lmbD, lmbE, lmbG, lmbK, lmbR, lmbV, and lmbW), resistance genes (lmrA, lmrB, and lmrC), a regulatory gene (lmbU), and SLCG_2920 were determined by qRT-PCR analyses. Specific primers were designed as listed in Table S1. In accordance with the above procedures, cDNAs were achieved from RNA samples (1 μg) after the DNase I treatment and reverse transcription. Then cDNAs were diluted 10-fold as the templates for qRT-PCR. The reaction volume of 20 μl is composed of 0.5 μl per primer (0.25 mM), 10 μl Maxima SYBR green/Rox qPCR master mix (MBI Fermentas), 2 μl diluted cDNA template (120 ng), and 7 μl RNase-free water. qRT-PCR was performed on the Applied Biosystems QuantStudio 6 Flex system with Maxima SYBR green/ROX qPCR master mix (MBI Fermentas). The reaction protocol consisted of 95°C for 10 min, followed by 40 cycles of 95°C for 19 s and 60°C for 35 s. The melting curve was inserted, ramping from 65°C to 95°C (increment 0.5°C/5 s), to verify specificity of primer amplification based on the presence of a single and sharp peak. The rpoD gene in S. lincolnensis was used as the internal control, and relative transcription was quantified using a comparative cycle threshold method (35).

Expression and purification of SLCG_2919.

The DNA fragment containing intact SLCG_2919 was successively obtained by PCR using the primers 2919-P9/2919-P10, listed in Table S1. The PCR product was digested with NdeI/HindIII and inserted into the corresponding site of pET-28a (Novagen), then introduced into E. coli BL21(DE3). The transformed cells were grown in 50 ml LB liquid medium at 37°C until the optical density at 600 nm (OD₆₀₀) reached 0.4 to 0.6 and then induced with isopropyl-β-d-thiogalactopyranoside (IPTG) at a final concentration of 0.2 mM at 16°C for 20 h. The cells were harvested by centrifugation (8,000 rpm, 10 min, 4°C), resuspended in lysis buffer (20 mM Tris and 500 mM NaCl, pH 8.0) and disrupted by ultrasonication at 4°C. The lysate was centrifuged (12,000 rpm, 30 min, 4°C), and the His₆-tagged SLCG_2919 protein present in the supernatant was recovered using a Ni²⁺-nitrilotriacetic acid (NTA) spin column (Bio-Rad). SLCG_2919 was eluted from the column with eluting buffer (20 mM Tris, 300 mM imidazole, and 500 mM NaCl, pH 8.0). The purified protein was desalted by a molecular sieve column (AKTA primePlus) with buffer A (20 mM Tris and 200 mM NaCl, pH 8.0). The concentration of purified protein was quantified by bicinchoninic acid (BCA) assays, and the purity was judged by SDS-PAGE analysis.

Electrophoretic mobility shift assays.

The EMSAs were performed as reported by Hellman and Fried (36). The promoter regions of structural genes (lmbA, lmbC-lmbD, lmbE, lmbJ-lmbK, lmbR, and lmbV-lmbW), resistance genes (lmrA, lmrB, and lmrC), the regulatory gene (lmbU), SLCG_2918, SLCG_2919, and SLCG_2920 were amplified by PCR with their respective primers (Table S1). The binding reaction system consisted of 10 mM Tris (pH 7.5), 5 mM MgCl₂, 50 mM EDTA, 60 mM KCl, 10 mM dithiothreitol (DTT), 10% glycerol, 150 ng labeled probes, and 0.01 to 2 μM purified His₆-tagged SLCG_2919 in a total volume of 20 μl. After incubation of the mixture at 30°C for 15 min, the samples were separated on 6% native PAGE gels in ice-cold 1× Tris-acetate-EDTA (TAE) buffer at 50 mA for about 40 min.

Analysis of SLCG_2919 binding site in the promoter region of SLCG_2920 by DNase I footprinting.

A DNase I footprinting assay was carried out as described by Zianni et al. (37). To determine the binding site of SLCG_2919 in the promoter region of its adjacent gene, SLCG_2920, a 176-bp 5′-FAM-labeled fragment was amplified by PCR using primers FAM-2919-P1/FAM-2919-P2. The footprinting reaction mixture contained 250 ng labeled DNA fragment, 0.25 to 1 μM His₆-SLCG_2919, and binding buffer (10 mM Tris [pH 7.5], 5 mM MgCl₂, 50 mM EDTA, 60 mM KCl, 10 mM DTT, and 10% glycerol) in a total volume of 50 μl. After incubation of the mixture at 25°C for 20 min, 5.5 μl DNase I buffer and 2 μl DNase I (1 U/μg; Promega) were added. The mixture was incubated at 25°C for 60 s and terminated by addition of DNase I stop solution and heating for 10 min at 65°C. DNA samples were analyzed with a 3730XL DNA genetic analyzer (Applied Biosystems) after purification, and data analyses were performed using the GeneMarker software program v2.2.

Lincomycin resistance test.

To determine the resistance of LCGL/pIB139 and LCGL/pIB139-2920 strains against lincomycin, 10 μl spore suspension of LCGL/pIB139-2920, as well as LCGL/pIB139, was inoculated into 5 ml liquid TSBY with various concentrations of Lin-A (0 μg/ml, 300 μg/ml, 350 μg/ml, 400 μg/ml, 450 μg/ml, and 500 μg/ml) and cultured at 30°C for 48 h to compare growth.

Statistical analysis.

All data in this study were obtained from biological triplicates and shown as means ± standard deviation of the mean (38). Significance analysis was performed with an unpaired two-tailed Student's t test, with * indicating a P value of <0.05, ** indicating a P value of <0.01, and *** indicating a P value of <0.001.

Ethical standards.

This article does not contain any studies with human participants or animals performed by any of the authors.