A novel TetR-type repressor directly modulates precursor supply and utilization for erythromycin biosynthesis

Abstract

Antibiotic biosynthesis in actinomycetes is controlled by various transcription factors, with TetR family regulators (TFRs) serving as important modulators of this process. We previously discovered a TFR, SACE_1906, which hinders erythromycin yield in Saccharopolyspora erythraea. However, the precise mechanism by which SACE_1906 regulates erythromycin biosynthesis remains elusive. Herein, we confirmed that SACE_1906 directly inhibits the transcription of its own and neighboring alcohol dehydrogenase gene SACE_1905, as well as all genes within the erythromycin biosynthetic cluster. It was found that two identical conserved sites (5′-CACCGGTCGGTATA-3′) in the intergenic spacer between SACE_1905 and SACE_1906 (SACE_1905-1906-int) are necessary for SACE_1906 binding. Deletion of SACE_1906 resulted in decreased propionyl-CoA content but increased the intracellular supply of methylmalonyl-CoA. We further proved that SACE_1906 restricts methylmalonyl-CoA production from propionyl-CoA to influence intracellular levels of these two erythromycin biosynthetic precursors through the direct repression of several propionyl-CoA carboxylase genes. Moreover, acetaldehyde has been characterized as an effector molecule of SACE_1906 that weakens its binding affinity for SACE_1905-1906-int. This study reveals a novel mechanism by which SACE_1906 orchestrates erythromycin biosynthesis in response to acetaldehyde, expanding our understanding of small-molecule-mediated regulation for antibiotic biosynthesis in actinomycetes.

1. Introduction

Actinomycetes synthesize a diverse array of secondary products, including clinically important antibiotics, immunosuppressants, and anticancer agents, which exhibit distinct structures and biological activities [1,2]. The biosynthesis of these compounds is subject to precise and intricate modulation by multiple types of regulatory factors, which sense various physiological and environmental signals [3,4]. Over the past few decades, classic engineering strategies for actinomycetes, such as manipulating functional gene expression, optimizing promoter strength, and modifying enzyme properties, have been established by dissecting their gene regulatory systems [[5], [6], [7]]. Therefore, the insights into sophisticated regulatory mechanisms of actinomycetes are important for understanding secondary metabolite overproduction.

Saccharopolyspora erythraea (S. erythraea), a model actinomycete, is utilized to industrially produce an important antibiotic erythromycin [8,9]. The condensation of one propionyl-CoA with six (S)-methylmalonyl-CoA units yields 6-deoxyerythronolide B, which undergoes sequential modifications by hydroxylases, glycosyltransferases, and methyltransferases to ultimately produce erythromycin A (Er-A) [10,11]. In S. erythraea, erythromycin biosynthetic gene (ery) cluster comprises 21 members, which are organized into four major polycistronic units [12]. In recent years, the dissection and characterization of various regulatory factors in S. erythraea have uncovered the molecular regulatory network regarding erythromycin biosynthesis [[13], [14], [15]]. The developmental regulator BldD stimulates erythromycin biosynthesis by interacting with the promoters of the ery cluster [13]. PhoP, an OmpR family regulator, directly activates the expression of bldD and most ery genes to manage erythromycin production [16]. The Lrp family transcription factor SACE_Lrp negatively modulates erythromycin biosynthesis using lysine, arginine, and histidine as effectors [14]. Additionally, some TetR family regulators (TFRs) govern erythromycin biosynthesis through the repression or activation of the ery cluster [15,[17], [18], [19], [20], [21], [22]]. Nevertheless, our current knowledge of the sophisticated mechanisms regarding erythromycin biosynthetic regulation is still insufficient.

TFRs are highly prevalent in actinomycetes and exert indispensable regulatory functions in diverse biological processes, such as antibiotic biosynthesis, morphological differentiation, and efflux pumps [23]. Each TFR typically contains an N-terminal domain for DNA binding and a C-terminal domain responsive to effectors, allowing both specific DNA recognition and small-molecule sensing, which constitutes the structural basis for its allosteric modulation [4,24]. Based on our previous prediction, 97 genes encoding TFRs are evenly distributed across the chromosomes of S. erythraea [19]. Recently, a few TFRs regulating erythromycin biosynthesis in response to effector molecules have been investigated in S. erythraea [15,25]. However, difficulties in identifying new TFRs and their effector ligands compromise efforts to elucidate the underlying mechanisms of ligand-mediated modulation of erythromycin biosynthesis.

Previously, we identified the P_eryAI-interactive TFR SACE_1906 and found that it exhibits a negative correlation with erythromycin yield in S. erythraea A226 [26]. Here, we investigated the precise regulatory function of SACE_1906 in erythromycin formation. SACE_1906 directly suppresses the transcription of itself and SACE_1905, as well as the ery cluster. SACE_1906 also directly inhibits several propionyl-CoA carboxylase genes to limit the carboxylation of propionyl-CoA to methylmalonyl-CoA, influencing their intracellular levels. In addition, acetaldehyde acts as an effector of SACE_1906, attenuating its binding affinity for the SACE_1905 and SACE_1906 intergenic spacer. Collectively, we uncovered a small-molecule-mediated regulatory mechanism for SACE_1906 in erythromycin biosynthesis.

2. Materials and methods

2.1. Strains, plasmids, primers, and growth conditions

The strains and plasmids employed in the study are detailed in Table S1, and the primers employed in the study are presented in Table S2. The DH5α (for gene cloning) and BL21(DE3) (for protein expression) strains of Escherichia coli (E. coli) were routinely cultured in LB medium. S. erythraea and its derivatives were cultivated on R3M solid medium for morphological observation, in TSB liquid medium for DNA extraction and seed preparation, and in R5 liquid medium for erythromycin production. The molecular protocols for these two types of strains were conducted following established procedures [15].

2.2. RNA preparation and reverse transcription-quantitative PCR (RT-qPCR) assay

Total RNA was isolated from A226 and ΔSACE_1906 strains grown for 24 or 48 h using TransZol kit (Transgen). The integrity and amount of RNA samples were separately evaluated through agarose gel electrophoresis and microplate spectrophotometry. cDNA synthesis from RNA was performed utilizing a reverse transcription kit (Vazyme), followed by quantitative PCR on a QuantStudio 6 Flex system (Applied Biosystems). The S. erythraea hrdB gene was employed as an endogenous control for data normalization [22].

2.3. Heterologous production and purification of SACE_1906

PCR amplification of SACE_1906 gene from the A226 genomic DNA was performed with the 1906-28a-F and 1906-28a-R primers. The product was double-digested utilizing NdeI and HindIII restriction enzymes and integrated into the same sites of pET28a, yielding pET28a-1906. This plasmid was then introduced into competent BL21(DE3) cells of E. coli via thermal shock. The N-terminally hexa-histidine-tagged SACE_1906 fusion protein was produced after induction with 0.4 mM IPTG at 16 °C for 22 h, followed by purification via Ni²⁺-NTA affinity chromatography. SACE_1906 purity was evaluated by SDS-PAGE while its concentration was determined using the BCA assay (Thermo Fisher).

2.4. Electrophoretic mobility shift assay (EMSA)

EMSA was implemented based on a previously described method [27]. DNA probes labeled with 5′-FAM were gained from the A226 genomic DNA utilizing the primers presented in Table S2. The PCR products were then mixed with different concentrations of purified SACE_1906 in a binding buffer (10 mM Tris–HCl (pH 8.0), 5 mM MgCl₂, 60 mM KCl, 50 mM EDTA, 10 mM dithiothreitol, and 10 % glycerol) under controlled conditions (30 °C, 20 min). For competitive assays, excessive unlabeled specific probes (30-fold) and excessive nonspecific probe poly-dIdC (30-fold) were used. Following the reaction, the mixtures underwent electrophoretic separation on 6 % native polyacrylamide gels.

2.5. Green fluorescent-based reporting assay

To construct the reporter plasmids pKC-1E and pKC-1906-1E, the SACE_1905 promoter (P₁₉₀₅) and egfp encoding enhanced green fluorescent protein (eGFP) were gained from the A226 genome and pKC-EE [15] with 1E-F1/R1 and 1E-F2/R2 primer pairs, respectively. The two fragments were ligated to HindIII and BamHI sites in pKC1139 [15] to create pKC-1E. The apramycin-resistance gene aac(3)IV promoter (P_aac(3)IV) and SACE_1906 were acquired utilizing 1906-1E-F1/R1 and 1906-1E-F2/R2 primer pairs, respectively. The two products were inserted to EcoRV and EcoRI sites in pKC-1E to produce pKC-1906-1E. These two constructed plasmids were introduced into competent DH5α cells of E. coli for fluorescence detection based on a previous protocol [15]. Furthermore, to avoid interference from SACE_1906 in S. erythraea with the eGFP reporter system, we transferred the plasmids pKC-1E and pKC-1906-1E into ΔSACE_1906 for fluorescence detection, as previously reported [28]. To evaluate the interaction between SACE_1906 and its effector, acetaldehyde or ethanol at different concentrations was added to the reporter system.

2.6. DNase I footprinting assay

To recognize accurate nucleotide sequences for SACE_1906 binding, we conducted DNase I footprinting experiment with the SACE_1905 and SACE_1906 intergenic spacer (SACE_1905-1906-int) as previously described [15]. A dual-labeled (5′-FAM/HEX) 211-bp DNA fragment was gained utilizing the FH-1905-1906-F and FH-1905-1906-R primers. The 20 μL reaction mixture, containing approximately 200 ng of the aforementioned probe and different amounts of purified SACE_1906, was treated under controlled conditions (30 °C, 20 min). The above reaction mixture was then adjusted to a final volume of 45 μL by adding 5 μL of DNase I buffer (10 × ), 2 μL of DNase I (1 U/μg, Promega), and nuclease-free water. After maintaining at 25 °C for 60 s, 5 μL of DNase I Stop buffer (Promega) was supplemented, and the mixture was heat-inactivated at 65 °C for 15 min. The recovered DNA samples underwent capillary electrophoresis analysis on an Applied Biosystems 3730XL system, with data processed in GeneMarker v2.2.

2.7. Detection of intracellular acyl-CoA levels

Intracellular acyl-CoAs in S. erythraea were extracted and quantified as previously reported [15]. After 72 h of cultivation in R5 liquid medium, the mycelia of A226 and ΔSACE_1906 were treated twice using the phosphate-buffered saline solution and resuspended in lysis reagent supplemented with 10 % trichloroacetic acid and 10 mM dithiothreitol. Next, the samples underwent triple freeze-thaw treatment using liquid nitrogen and ice-water baths. The clarified supernatants were loaded onto Sep-Pak C18 solid-phase columns (Waters) for sample adsorption. The eluates were lyophilized utilizing a vacuum freeze-drying equipment (Scientz). Quantitative analysis of several acyl-CoA compounds was executed using HPLC system following the established protocol [15].

2.8. Statistical analysis

Experimental data are reported as means ± standard deviations, with intergroup comparisons assessed using two-tailed Student's t-test. Statistical significance is established at P < 0.05, with asterisks representing significance levels: P < 0.05 (∗), P < 0.01 (∗∗), and P < 0.001 (∗∗∗).

3. Results

3.1. SACE_1906 exhibits direct transcriptional repression to itself and SACE_1905

In our previous study, SACE_1906 negatively modulated erythromycin yield through transcriptional repression of its neighboring divergent gene SACE_1905 [26]. Considering that TFRs typically function as autoregulators [23], we investigated whether SACE_1906 controlled the transcription of its own gene. The RT-qPCR experiment displayed that the SACE_1906 transcript of ΔSACE_1906 increased by 57.8- or 77.0-fold relative to that of A226 at 24 or 48 h (Fig. 1A), indicating that SACE_1906 inhibits its own expression. To investigate regulatory modes of SACE_1906 on SACE_1905 and itself, EMSA was performed on the SACE_1905 and SACE_1906 intergenic spacer (SACE_1905-1906-int) (Fig. S1). The result showed that SACE_1906 specifically bound to SACE_1905-1906-int (Fig. 1B). Further, an eGFP reporter system harboring pKC-1E and pKC-1906-1E plasmids was introduced into E. coli DH5α and S. erythraea ΔSACE_1906 to verify the interaction between SACE_1906 and the SACE_1905 promoter (P₁₉₀₅), respectively (Fig. 1C). Bioluminescence in DH5α/pKC-1906-1E or ΔSACE_1906/pKC-1906-1E was markedly diminished compared with that in DH5α/pKC-1E or ΔSACE_1906/pKC-1E (Fig. 1D and E), revealing that SACE_1906 suppresses the expression of SACE_1905 by binding to its promoter. In addition, no discernible change in morphogenesis was observed between A226 and ΔSACE_1906 (Fig. S2). Thus, our data corroborated that SACE_1906 directly suppresses its own and SACE_1905 expression to modulate erythromycin biosynthesis.

Fig. 1.

SACE_1906-mediated regulation to its own gene and SACE_1905. (A) SACE_1906 transcription of A226 and ΔSACE_1906 cultivated for 24 and 48 h. ΔSACE_1906 was constructed by introducing an in-frame deletion of the 372 bp fragment within the SACE_1906 gene [26]. (B) EMSA with SACE_1906 interacting with SACE_1905-1906-int. The two shift bands (complex 1 and 2) suggest the presence of two SACE_1906-binding sites in the probe. (C) Schematic representation for the eGFP reporting experiment. The experiment contains two distinct plasmids: pKC-1E, which expresses egfp under the control of P₁₉₀₅, and pKC-1906-1E, which expresses egfp under P₁₉₀₅ with SACE_1906 expression driven by P_aac(3)IV. (D) Relative bioluminescence units (RBUs) of DH5α/pKC-1E and DH5α/pKC-1906-1E. (E) RBUs of ΔSACE_1906/pKC-1E and ΔSACE_1906/pKC-1906-1E. The mean from three independent measurements is presented, with error bars denoting the standard deviation (SD). P < 0.01 (∗∗); P < 0.001 (∗∗∗).

3.2. SACE_1906 directly suppresses the transcription of ery cluster

Since SACE_1906 inhibited erythromycin production, we wondered whether it influenced the transcription of genes from the ery cluster. RT-qPCR results displayed that except for eryBIII, the transcripts of eryAI, ermE, eryBI, eryBVI, eryCI, eryBIV, and eryK in ΔSACE_1906 exhibited overall increases over those in A226 after 24 h of growth. When culturing for 48 h, the transcription of these eight genes in ΔSACE_1906 were all enhanced relative to those in A226 (Fig. 2A). These observations indicated that SACE_1906 can suppress all genes in the ery cluster. Further, EMSAs were carried out using purified SACE_1906 protein to determine its regulatory effects on the above genes (Fig. 2B). SACE_1906 specifically bound to the promoters of these genes (Fig. 2C). Based on these findings, we concluded that SACE_1906 directly inhibits the expression of all cluster genes.

Fig. 2.

SACE_1906 has direct inhibitory effects on ery cluster. (A) Transcripts of genes from the ery cluster in A226 and ΔSACE_1906 at 24 and 48 h. (B) Genetic organization of the ery cluster in S. erythraea. Short lines represent the probes harboring the ery promoters. (C) EMSAs with SACE_1906 interacting with the ery promoters. A single shift band suggests the presence of only one SACE_1906-binding site in these probes. The mean from three independent measurements is presented, with error bars denoting the SD. P < 0.05 (∗); P < 0.01 (∗∗); P < 0.001 (∗∗∗); ns, non-significant.

3.3. Interrogation of nucleotide sites for SACE_1906 binding

To investigate the exact binding sequences of SACE_1906, a DNase I footprinting experiment was performed with the high-affinity target SACE_1905-1906-int. After adding 800 nM SACE_1906 protein, two nucleotide regions (5′-CCTTCAGGTATACCGACCGGTGTGGTTAC-3′ and 5′-GCCATGTCAACCCACCGGTCGGTATAG-3′) within SACE_1905-1906-int were well protected (Fig. 3A). Subsequently, utilizing the online MEME tool, we recognized an identical motif (CACCGGTCGGTATA) in these two regions (Fig. 3B). To determine whether the motif was necessary for SACE_1906 binding, EMSAs were conducted with the original sequence (O) containing these two identical motifs, two single-sequence mutated fragments (S1 and S2), and a double-sequence mutated fragment (S3) (Fig. 3C). Differing from probe O, probes S1 and S2 each produced one single retarded band, whereas adding probe S3 alone resulted in no detectable band shift (Fig. 3D). Further, the putative transcriptional start site (TSS) and conserved −10 and −35 promoter elements of SACE_1905 and SACE_1906 were identified through the online BDGP and phiSITE programs, respectively (Fig. 3E). Taken together, these results demonstrated that the fragment site U (located from nucleotides 8 to 21 relative to the SACE_1905 TSS) or site D (located from nucleotides 13 to 26 relative to the SACE_1906 TSS) is crucial for SACE_1906 binding to SACE_1905-1906-int.

Fig. 3.

Characterization of the SACE_1906-binding sites. (A) Accurate sequences of SACE_1906 interacting with SACE_1905-1906-int by DNase I footprinting experiment. (B) Determination of two identical sequences utilizing MEME program (http://meme-suite.org/). The top section indicates the WebLogo server's standard code. (C) Illustration of probes with sequence mutations. As a control, we employed probe O, an 85 bp DNA fragment harboring sites U and D from SACE_1905-1906-int. Probe S1, the sequence of site U in probe O was modified; probe S2, the sequence of site D in probe O was modified; probe S3, the sequences of sites U and D in probe O were modified. (D) EMSAs for SACE_1906 binding to probes O, S1, S2, and S3. (E) Prediction of transcriptional elements of SACE_1905 and SACE_1906 using BDGP and phiSITE tools (http://fruitfly.org/; http://phisite.org/). Underlining, the binding sites of SACE_1906; larger font, transcriptional start site (TSS); blue rectangle, the potential −10 or −35 box of SACE_1905; red rectangle, the potential −10 or −35 box of SACE_1906; black rectangle, start codon.

3.4. SACE_1906 modulates intracellular levels of the two erythromycin precursors

The intracellular supply of precursors for erythromycin biosynthesis is closely related to erythromycin production [29,30]. Considering that SACE_1906 negatively managed erythromycin yield by transcriptionally repressing the ery cluster (Fig. 3), we wondered whether it also affected the contents of propionyl-CoA and methylmalonyl-CoA, the original building blocks essential for erythromycin production (Fig. 4A). HPLC experiments were employed to detect intracellular amounts of several acyl-CoAs following our previous method [15]. As shown in Fig. 4B, ΔSACE_1906 exhibited a 56.0 % reduction in propionyl-CoA content and a 17.7 % elevation in methylmalonyl-CoA content over A226, whereas acetyl-CoA, malonyl-CoA, and succinyl-CoA levels did not differ in response to SACE_1906 deletion. These findings indicated that SACE_1906 regulates intracellular levels of the two erythromycin biosynthetic precursors.

Fig. 4.

SACE_1906 manages intracellular contents of erythromycin biosynthetic substrates. (A) Local anabolic routes of propionyl-CoA and methylmalonyl-CoA in S. erythraea. PCC, propionyl-CoA carboxylase; Er-A, erythromycin A. (B) Intracellular concentrations of several acyl-CoAs in A226 and ΔSACE_1906 at 72 h. The mean from three independent measurements is presented, with error bars denoting the SD. P < 0.05 (∗); ns, non-significant.

3.5. SACE_1906 functions as a direct inhibitor of propionyl-CoA carboxylation

In S. erythraea, propionyl-CoA undergoes carboxylation in a reaction mediated by propionyl-CoA carboxylase, producing methylmalonyl-CoA [12,15]. As SACE_1906 deletion resulted in the decreased propionyl-CoA level and enhanced methylmalonyl-CoA level (Fig. 4B), we speculated that SACE_1906 might regulate methylmalonyl-CoA supply through the above carboxylase path. Therefore, we implemented RT-qPCR to determine the transcriptional levels of all putative propionyl-CoA carboxylase genes between A226 and ΔSACE_1906. The SACE_3398, SACE_3400, and SACE_6509 transcripts of ΔSACE_1906 were enhanced by 2.0-, 2.3-, and 2.1-fold compared with those of A226 at 24 h, while the SACE_3400 and SACE_6509 transcription in ΔSACE_1906 exhibited 1.6- and 2.0-fold increases over those in A226 at 48 h, respectively (Fig. 5A and B). Further, EMSA experiments were conducted to characterize the direct interactions of SACE_1906 with these three target genes. Results displayed that SACE_1906 specifically bound to P₃₃₉₈, P₃₄₀₀, or P₆₅₀₉ but not to the negative control P_3241-3242 or P₄₂₃₇ (Fig. 5C and Fig. S3). In addition, we identified a conserved SACE_1906-binding motif (ttgatCgac, t: T, A/C; g: G, T/A; a: A, T/C/G; c: C, A) among promoter regions of SACE_1906, SACE_1905, the ery cluster, and above three genes utilizing the MEME program (Fig. 5D). Collectively, we concluded that SACE_1906 directly represses transcription of the propionyl-CoA carboxylase genes to modulate the supply of methylmalonyl-CoA.

Fig. 5.

SACE_1906 exhibits direct inhibitory effect on propionyl-CoA carboxylation. (A) Transcripts of propionyl-CoA carboxylase genes in A226 and ΔSACE_1906 at 24 h. (B) Transcripts of propionyl-CoA carboxylase genes in A226 and ΔSACE_1906 at 48 h. (C) EMSAs for SACE_1906 interacting with SACE_3398, SACE_3400, and SACE_6509 promoters. (D) MEME-based recognition of a conserved motif in upstream sequences of target genes (SACE_1906, SACE_1905, eryAI, ermE, eryBI, eryBVI, eryK, SACE_3398, SACE_3400, and SACE_6509). The top section indicates the WebLogo server's standard code. The accurate sites of SACE_1906 interacting with SACE_1905-1906-int are boxed. The genes SACE_3241 and SACE_3242 are co-transcribed [25]. SACE_3241, SACE_3398, or SACE_7039 encodes propionyl-CoA carboxylase; SACE_3242, SACE_3400, SACE_4237, or SACE_6509 encodes acetyl-/propionyl-CoA carboxylase [12,25]. The mean from three independent measurements is presented, with error bars denoting the SD. P < 0.05 (∗); P < 0.01 (∗∗); P < 0.001 (∗∗∗); ns, non-significant.

3.6. Acetaldehyde is an effector of SACE_1906 for the regulation of SACE_1905

Previous studies have shown that the DNA-binding activities of TFRs in actinomycetes can be mediated by small molecules [23,31]. Considering that SACE_1906 transcriptionally regulated SACE_1905, which encodes an alcohol dehydrogenase that catalyzes the conversion of aldehydes to alcohols [26], we first tested the effects of acetaldehyde and ethanol on SACE_1905 transcription in vivo. Results showed that after feeding 10 or 20 mM acetaldehyde for 24 h, the SACE_1905 transcript of A226 was markedly increased (Fig. 6A), whereas the transcription of SACE_1905 in A226 did not differ in response to the individual addition of 10 and 20 mM ethanol for 24 h (Fig. 6B). This hinted that acetaldehyde may act as an effector to influence the modulation of SACE_1906 on SACE_1905 transcription. To this end, EMSA experiments were conducted, and results displayed that acetaldehyde attenuated the binding of SACE_1906 to SACE_1905-1906-int, whereas no change was observed in the interaction between SACE_1906 and the probe upon individual addition of ethanol (Fig. 6C and D). Subsequently, the in vivo E. coli eGFP reporter system was applied to interrogate whether acetaldehyde or ethanol affected the repression of SACE_1906 on SACE_1905 (Fig. 1C). When acetaldehyde was added at concentrations ranging from 2 to 16 μM, bioluminescence in DH5α/pKC-1906-1E was enhanced in a dose-dependent manner, whereas no statistical changes were observed in the bioluminescence intensity of DH5α/pKC-1906-1E with ethanol at the same concentrations (Fig. 6E and F). These indicated that acetaldehyde can dissociate SACE_1906 from the target. Further, we employed the eGFP reporter system in S. erythraea to validate the impact of acetaldehyde on the regulation of SACE_1906 under physiological conditions. Bioluminescence of ΔSACE_1906/pKC-1906-1E was stimulated upon the addition of acetaldehyde ranging from 0.3 to 4 μM, while that in ΔSACE_1906/pKC-1E was unrelated to elevated acetaldehyde levels (Fig. 6G). Accordingly, these observations demonstrated that acetaldehyde can relieve the repression of SACE_1906 on SACE_1905, acting as an effector.

Fig. 6.

Acetaldehyde disrupts the interaction of SACE_1906 with SACE_1905-1906-int. (A) SACE_1905 transcription of A226 with or without acetaldehyde at 24 h. (B) SACE_1905 transcription of A226 with or without ethanol at 24 h. (C) EMSA for the impact of acetaldehyde on SACE_1906 interacting with SACE_1905-1906-int. (D) EMSA for the impact of ethanol on SACE_1906 interacting with SACE_1905-1906-int. (E) RBUs of DH5α/pKC-1E and DH5α/pKC-1906-1E with various concentrations of acetaldehyde. (F) RBUs of DH5α/pKC-1E and DH5α/pKC-1906-1E with various concentrations of ethanol. (G) RBUs of ΔSACE_1906/pKC-1E and ΔSACE_1906/pKC-1906-1E with various concentrations of acetaldehyde. The mean from three independent measurements is presented, with error bars denoting the SD. P < 0.05 (∗); P < 0.01 (∗∗); ns, non-significant.

4. Discussion

TFRs are abundant in actinomycetes and function as repressors or activators of antibiotic biosynthesis via multiple regulatory mechanisms [23,32]. We previously identified the P_eryAI-interactive TFR SACE_1906 and confirmed its inhibitory effect on erythromycin yield of S. erythraea [26]. In this study, we uncovered a novel mechanism by which SACE_1906 modulates erythromycin biosynthesis in response to acetaldehyde (Fig. 7). SACE_1906 exhibits direct transcriptional suppression to its own and adjacent gene SACE_1905, as well as all ery genes. SACE_1906 also directly inhibits several putative propionyl-CoA carboxylase genes, negatively modulating the supply from propionyl-CoA to methylmalonyl-CoA, ultimately affecting intracellular levels of these two precursors. Acetaldehyde is identified as an effector that mediates SACE_1906 regulation during erythromycin biosynthesis.

Fig. 7.

Proposed regulatory model for SACE_1906 during erythromycin formation. PCC, propionyl-CoA carboxylase; Er-A, erythromycin A. Pink blocking lines depict direct suppression at the transcriptional level. Yellow-filled arrowheads depict the reaction direction.

Many TFRs have been demonstrated to modulate their own and adjacent genes at the transcriptional level [28,33,34]. As a TFR, SACE_1906 exhibits autoregulation and functions as a repressor for its neighboring SACE_1905 (Fig. 1). Sequence alignment and phylogenetic analyses revealed that homologous proteins of SACE_1906 are widely present among antibiotic-producing actinomycetes (Fig. S4), implying that they possess conserved regulatory functions in secondary metabolism. RT-qPCR and EMSA results displayed that SACE_1906 transcriptionally inhibited all members of the ery cluster through interacting with their promoters (Fig. 2). Similar regulatory behavior was observed in the mechanistic exploration of AcrT, another TFR in S. erythraea [15]. Additionally, a few TFRs have been shown to indirectly regulate erythromycin biosynthesis through cascaded control networks [[20], [21], [22]]. These findings highlight the diversity and complexity of erythromycin biosynthetic regulation.

The propionyl-CoA carboxylase pathway drives the carboxylation of propionyl-CoA, serving as a pivotal step in methylmalonyl-CoA production in actinomycetes [12,35]. Several TFRs have been demonstrated to exert direct regulatory effects on their respective adjacent acetyl-/propionyl-CoA carboxylase genes, including BkdR in Streptomyces albus, AccR in S. avermitilis, and PccD in S. erythraea [25,36,37]. Here, we demonstrated that SACE_1906 transcriptionally represses the non-adjacent acetyl-/propionyl-CoA carboxylase genes SACE_3398, SACE_3400, and SACE_6509 by interacting with their promoter regions (Fig. 5). During erythromycin biosynthesis, methylmalonyl-CoA is required at sixfold higher levels than propionyl-CoA and acts as the rate-limiting substrate for product formation [38]. Compared with A226, propionyl-CoA and methylmalonyl-CoA amounts in ΔSACE_1906 decreased by 56.0 % and increased by 17.7 %, respectively (Fig. 4). The mismatch in the conversion ratio may be attributed to the greater utilization of methylmalonyl-CoA over propionyl-CoA in response to SACE_1906 deletion. In conclusion, these findings indicated that SACE_1906 inactivation optimizes the availability of these two substrates, thus boosting erythromycin production.

Since antibiotic biosynthesis of actinomycetes typically involves intricate regulatory networks [3,39], we screened potential targets of SACE_1906 in the S. erythraea genome utilizing PREDetector program [40] with the identified binding sites (Fig. 3). When cut-off score greater than 7.0, a total of 88 regulatory genes were predicted (data not shown), among which the TFR SACE_4839 was a possible target of SACE_1906. SACE_4839 and the neighboring nitrogen metabolic regulator SACE_4838 have been shown to cross-regulate the biosynthesis of erythromycin [22]. Our experimental results proved that SACE_1906 directly prevented SACE_4838 but activated SACE_4839 at the transcriptional level (Fig. S5). These suggested that SACE_1906 can form hierarchical regulatory networks to modulate erythromycin biosynthesis, and it will be of interest to address these issues in the future.

In many cases, the ligands of TFRs are identical or related to the substrates of cognate target gene products [23]. In this study, the addition of acetaldehyde markedly stimulated the transcription of SACE_1905 (Fig. 6), suggesting that it may weaken the inhibitory effect of SACE_1906 on SACE_1905. In vitro and in vivo results from EMSA and eGFP reporting experiments demonstrated that acetaldehyde dose-dependently attenuated the interaction of SACE_1906 with its target (Fig. 6), acting as a specific effector of SACE_1906. Many toxic molecules have been found to mediate the regulation of TFRs on DNA targets, among which excessive intracellular resorcinol can release the repression of RolR on the rolRHMD gene cluster to accelerate its own degradation in Corynebacterium glutamicum [41,42]. Considering that SACE_1905 encodes an alcohol dehydrogenase that reduces acetaldehyde to ethanol [26], the response of SACE_1906 to acetaldehyde promotes its conversion, thereby preventing its intracellular accumulation.

5. Conclusions

This study characterizes the regulatory function of S. erythraea SACE_1906 during erythromycin formation. SACE_1906 suppresses the transcription of itself and the adjacent SACE_1905, as well as the ery cluster by binding to the corresponding targets. SACE_1906 also directly inhibits SACE_3398, SACE_3400, and SACE_6509 transcription to limit methylmalonyl-CoA production from propionyl-CoA, influencing intracellular levels of these two erythromycin biosynthetic precursors. Furthermore, acetaldehyde functions as an effector of SACE_1906 to coordinate erythromycin biosynthesis. This investigation enriches our knowledge of the intricate regulatory systems underlying antibiotic biosynthesis, expanding the toolkit of genetic elements for actinomycete synthetic biology.

CRediT authorship contribution statement

Panpan Wu: Writing – review & editing, Writing – original draft, Supervision, Funding acquisition, Formal analysis, Data curation, Conceptualization. Zhongqiu Meng: Investigation, Formal analysis, Data curation. Yuling Shu: Investigation, Formal analysis, Data curation. Ketao Chen: Investigation, Formal analysis, Data curation. Zhenyue Xu: Investigation, Data curation. Xunduan Huang: Writing – review & editing. Buchang Zhang: Supervision, Conceptualization. Jing Liu: Writing – review & editing, Funding acquisition, Formal analysis, Conceptualization. Hang Wu: Writing – review & editing, Supervision, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article: