Background

Four tetracycline antibiotics, chlortetracycline, tetracycline, demeclocycline (6-DCT), and 6-demethyltetracycline, are secondary metabolites of Streptomyces aureofaciens (Zhu et al. 2001). 6-DCT is produced when the 6-methylation step is blocked in chlortetracycline biosynthesis pathway, and is used for the industrial production of semisynthetic tetracyclines (Nakano et al. 2000). There has been a little research into 6-DCT production, whereas many papers are available concerning the gene clusters and ribosomes involved in the biosynthesis of chlortetracycline and other tetracyclines. Chlortetracycline production depends on the expression of proteins involved in its biosynthetic pathway, so identifying and understanding the corresponding proteins are required for the industrial production of 6-DCT biosynthesis (Mikulik et al. 1983; Li et al. 2001). Streptomyces aureofaciens strain A6-9, obtained with the traditional mutagenesis, produces the elevated levels of 6-DCT. This increased formation of 6-DCT may be attributable to changes in the expression of some proteins in the 6-DCT biosynthetic pathway. Modulating the expression of these proteins may significantly affect 6-DCT biosynthesis.

A type II polyketide synthase (PKS) constructs the skeleton of tetracycline antibiotics, and a series of subsequent reactions modify the skeleton to produce four final products (Nakano et al. 2000). The modification reactions (e.g., chlorination, methylation, and oxidation) are important for the diversity of tetracyclines, and 6-DCT is formed when the 6-methylation step in the pathway is lacking. Above all, oxidation reactions catalyzed by monooxygenases are necessary for the formation of 6-DCT (Fig. 1) (Taguchi et al. 2011; Bentley et al. 2002).

Fig. 1
figure 1

Detailed pathway of the precursors involved in the synthesis of 6-DCT in S. aureofaciens. Black boxes show the oxidation steps (CtcJ) in the 6-DCT-forming pathway

Full size image

Although certain features of the secondary metabolism of Streptomyces have been studied in detail, including the synthesis of antibiotics and its differentiation processes, its primary metabolism has not been (Al-Shamma et al. 1983; Hensel et al. 1991). Hydroxyacyl-CoA dehydrogenase is involved in the β-oxidation of fatty acids, which is essential for the whole metabolism of the microorganism (Fujita et al. 2007; Rock and Cronan 1996; DiRusso and Nystrom 1998; Schujman et al. 2003; Lu and Rock 2006). Hydroxyacyl-CoA is oxidized by hydroxyacyl-CoA dehydrogenase to generate acetyl-CoA, NADH, and H+, which are essential for the biosynthesis of 6-DCT (Fig. 2).

Fig. 2
figure 2

Pathway of the β-oxidation of fatty acids, the biosynthetic pathway of demeclocycline (6-DCT) in S. aureofaciens. Black rectangle shows the function of hydroxyacyl-CoA dehydrogenase (CtcH) and the acetyl-CoA generated by the β-oxidation of fatty acids used for the biosynthesis of 6-DCT. Black ellipse shows the function of 6-hydroxylase in the 6-DCT biosynthetic pathway. Abbreviations: PKS, polyketide synthase reaction; 6-CH 3 , 6-methylation reaction; 6-OH, 6-hydroxylation reaction; 7-Cl, 7-chlorination reaction; CoA, coenzyme A; CTC, chlortetracycline; TC, tetracycline; 6-DMT, oxytetracycline

Full size image

In this study, we examined the differences in protein expression between A6-9 and the parental wild-type (WT) strain to identify the enzymes that are important for 6-DCT biosynthesis, using two-dimensional (2D) gel electrophoresis and mass spectrometry. Based on the results, the roles of CtcH (hydroxyacyl-CoA dehydrogenase) and CtcJ (monooxygenase) in 6-DCT biosynthesis were examined by inactivating or overexpressing them individually and together in DT1.

Methods

Microbial strains and plasmids

The strains and plasmids used in this study are listed in Table 1.

Table 1 Strains and plasmids used in this study
Full size table

Media and culture conditions

Escherichia coli DH5α and its derivatives were routinely grown aerobically in LB medium (37 °C, 180 rpm) (Jian et al. 2010; Zhang et al. 2006). ISP4-containing 80 mmol l−1 MgCl2 was used for E.coli ET12567/pUZ8002-Streptomyces conjugative medium (Liu et al. 2015).

The mycelia of S. aureofaciens were stored in 20 % glycerol solution at −80 °C. A 250-ml shaker flask containing 30 ml of seed medium was inoculated with 3 ml of mycelial solution. After incubation for 48 h at 27 °C, a 250-ml fermentation flask containing 30 ml of fermentation medium was inoculated with 3 ml of seed medium, and shaken at 27 °C and 240 rpm for 148 h. All experiments were performed in triplicate. The seed medium and fermentation medium were adjusted to pH 6.5.

Generation of gene-inactivated and gene-overexpressing strains

Genes ctcH and ctcJ were deleted with the non-replicating plasmid pOJ260-neo, which produced deletion mutants substituted with a kanamycin-resistance gene cassette. The transformants (ΔctcH and ΔctcJ) were grown on ISP4 plates-containing kanamycin and nalidixic acid, selected on MS plates-containing kanamycin and apramycin, and verified with DNA sequencing and PCR. All primers used in this study are shown in Additional file 1: Online Resource 1 and the plasmid used for the gene-overexpressing constructs is shown in Additional file 2: Online Resource 2.

Genes ctcH and ctcJ were overexpressed individually or both genes were overexpressed simultaneously with an integrative plasmid, pIB139, containing a strong promoter from the erythromycin-resistance gene and the ϕC31 attp site, which can integrate into the attB site of Streptomyces. A 719-bp DNA fragment containing ctcH was cloned and inserted into pIB139 to generate pIB139-H, and a 763-bp DNA fragment containing ctcJ was inserted into pIB139 to generate pIB139-J. The two fragments were inserted into pIB139 to generate pIB139-H-J. The genotypes of WT and the overexpressing mutants are shown in Fig. 5a. The transformants (ctcH+, ctcJ+, and ctcH–ctcJ+) were isolated on ISP4 plates under apramycin and nalidixic acid selection, cloned on MS plates, and verified with PCR and a transcription analysis.

Determination of 6-DCT

The production of 6-DCT was determined with high-performance liquid chromatography (HPLC), using the linear regression equation: U = 3 × 10−4 A − 10.98, r 2 = 0.9956, where U and A represent the concentration of 6-DCT and the peak area, respectively. Samples (1, 2, 3, 4, or 5 µl) of a standard 6-DCT solution (4000 µg/ml) were injected and eluted after 15 min, and the peak areas were calculated. The linear regression equation was obtained and plotted, with the concentration of 6-DCT on the horizontal axis and the peak area on the vertical axis. As a control, WT DT1 was processed under the same fermentation conditions.

Analytical methods

After fermentation for 7 days, the pH of the culture broth was adjusted to 1.8–2.0 with oxalic acid. The filtration supernatant was assayed with HPLC using an Extend-C18 reverse phase column (4.6 mm × 150 mm, 5 μm; Agilent, USA) at 350 nm with 0.01-M oxalic acid–0.08-M ammonium oxalate solution and 40 % methanol/acetonitrile/water as the mobile phase at a flow rate of 0.4 ml/min. The column temperature was at 25 °C.

2D gel electrophoresis and Mascot search

The mycoproteins were extracted with ultrasonication and their concentrations determined with Coomassie Brilliant Blue. The mycelia were washed in low-salt buffer (2.5 mM KH2PO4, 5 mM NaH2PO4) precooled to 4 °C, and then resuspended in 1 ml of lysis buffer (8 M carbamide, 1 % IPG buffer, 2 % dithiothreitol). Protein sample preparation and 2D gel electrophoresis were performed as previously described (Helmel et al. 2014; Jan et al. 2014). (All chemicals were from Sigma-Aldrich, St. Louis, MO, USA). A Mascot search was performed by Shanghai Bo-Yuan Biological Technology Co., Ltd.

RNA isolation and quantitative real-time reverse transcription-PCR (RT-qPCR)

The transcription levels of ctcH and ctcJ were assayed with RT-qPCR (Bio-Rad CFX96). Sample preparation, RNA isolation, and the PCR were all conducted according to a previous study (Zhu et al. 2013). RT-qPCR was performed with the SYBR Green Two-Step RT-qPCR SuperMix (Takara). The transcription levels were normalized to the expression of 16S rRNA. RT-qPCR was performed in three independent experiments, each in triplicate.

Results

Differences in protein expression by S. aureofaciens strains A6-9 and DT1

A high-6-DCT-yield S. aureofaciens strain A6-9 was generated with the traditional mutagenesis. Compared with the parental strain under the same flask fermentation conditions, A6-9 showed a 20 % increase in 6-DCT production, which might be attributable to changes in protein expression. For this reason, 2D gel electrophoresis and a Mascot search were used to explore the differences in protein expression between the mutant A6-9 and the parental strain DT1.

Seven differentially expressed proteins were identified with mass spectrometry using the PDquest software (Fig. 3; Table 2). Malate dehydrogenase, glutamine synthetase, 6-hydroxylation enzyme, CtcH, and CtcJ were expressed more strongly and ATP synthetase less strongly in A6-9 than in DT1.

Fig. 3
figure 3

Two-dimensional electrophoresis patterns of S. aureofaciens strains A6-9 and DT1. According to the PDquest software, seven protein spots were differentially expressed. a Two-dimensional electrophoresis pattern of strain A6-9. b Two-dimensional electrophoresis pattern of DT1

Full size image
Table 2 Appraisal results of protein points by mass spectrometry (MS)
Full size table

6-DCT biosynthesis requires CtcH and CtcJ

The proteins expressed more strongly in A6-9, according to 2D gel electrophoresis and the Mascot search, might explain, at least in part, the increase in 6-DCT production. CtcH and CtcJ are encoded by ctcH and ctcJ, respectively, which belong to the chlortetracycline biosynthesis gene cluster, and occur near one another on the S. aureofaciens chromosome, but have not yet been investigated. Therefore, their roles in 6-DCT biosynthesis were examined first, and the other enzymes upregulated in A6-9 will be explored in a future study. ctcH and ctcJ were knocked out individually to study their effects on 6-DCT biosynthesis.

The ctcH and ctcJ deletion mutants of S. aureofaciens DT1 were generated by in-frame deletion (Fig. 4a) and confirmed with PCR (Fig. 4b). The mutants and WT were grown in fermentation culture for 7 days and the 6-DCT concentrations detected with HPLC. Less 6-DCT was detected in ΔctcH, and no 6-DCT was detected in ΔctcJ (Fig. 4c). The HPLC analyses of the mutants and WT are shown in Additional file 3: Online Resource 3.

Fig. 4
figure 4

Deletion of ctcH and ctcJ genes. a Schematic representation of the in-frame deletion of ctcH and ctcJ. A 410-bp fragment of ctcH and a 254-bp fragment of ctcJ were replaced with the 1145-bp neo gene by double crossover. b Confirmation of the ctcH deletion and ctcJ deletion mutants with PCR amplification. c 6-DCT production during the fermentation of the parental strain DT1 and the ctcH-deleted, ctcJ-deleted strains

Full size image

Overexpression of ctcH- and/or ctcJ-enhanced 6-DCT production

The results of ctcH or ctcJ deletion suggested that the production of 6-DCT may increase when the genes are overexpressed. To improve 6-DCT production and explore the effect of each gene on it, ctcH and ctcJ were overexpressed individually and together in DT1 (Fig. 5b). As predicted, the ctcH+, ctcJ+, and ctcH–ctcJ+ mutants expressed considerably more 6-DCT than did DT1, with increases of ca. 50, 37, and 60 %, respectively. Compared with the 3164-µg/ml 6-DCT produced by DT1, ctcH+ produced 4715 µg/ml, ctcJ+ produced 4332 µg/ml, and ctcH–ctcJ+ produced 5075 µg/ml (Fig. 5c). The transcription levels of ctcH and ctcJ were also high in the mutants (Fig. 5d), indicating that the overexpression of these genes caused the increase in 6-DCT production. The HPLC analysis of the mutants and WT is shown in Additional file 3: Online Resource 3.

Fig. 5
figure 5

Overexpression of the individual ctcH and ctcJ genes and the simultaneous overexpression of ctcH and ctcJ. a Genotypes of WT and overexpressing mutants. b Confirmation of the ctcH-overexpressing, ctcJ-overexpressing, and ctcH–ctcJ-overexpressing mutants with PCR. The PCR products were 900 bp for pIB139 (empty vector), 1663 bp for ctcJ+, 1619 bp for ctcH+, and 2382 bp for ctcH–ctcJ+. All the PCR products were sequenced to confirm each mutation. c 6-DCT production during the fermentation of the parental strain DT1, and the ctcH-overexpressing, ctcJ-overexpressing, and ctcH–ctcJ-overexpressing mutants. d RT-qPCR analysis of ctcH and ctcJ expression in the three overexpressing mutants and WT. Relative levels of ctcH and ctcJ transcripts were determined after normalization to the internal reference, 16S rRNA. RT-qPCR was performed in three independent experiments, each in triplicate

Full size image

Discussion

In recent years, engineering the primary metabolism has been used to supply more precursors for the production of the secondary metabolites of Streptomyces and rationally engineering the secondary metabolism has shown potential utility in strain improvement, allowing the production of high levels of antibiotics and to reduce the byproducts (Pang et al. 2015; Zhuo et al. 2010; Zhang et al. 2008). Gene overexpression has been an effective way to enhance antibiotic production (Liu et al. 2015; Du et al. 2013; Yuan et al. 2011; Malla et al. 2010). In this study, the overexpression of some key enzymes involved in the 6-DCT biosynthetic pathway in the high-yield strain A6-9 explained the enhanced production of 6-DCT.

ctcH is an endogenous gene that encodes a 3-hydroxylacyl-CoA dehydrogenase, an NADH-dependent enzyme that also has Δ3-cis, Δ2-trans-enoyl-CoA isomerase and (S)/(R)-epimerase activities. It is responsible for the oxidation of 3-hydroxyacyl-CoA, generating NADH, H+, and β-ketoacyl-CoA, which can also generate acetyl-CoA when it is catalyzed by thiolase during the β-oxidation of fatty acids (Fig. 2) (Volodina and Steinbuchel 2014; Fujita et al. 2007; Kunau et al. 1995; Black and DiRusso 1994; Pramanik et al. 1979). Here, the results of modulating ctcH can be explained by the requirement for NADH, H+, and acetyl-CoA in 6-DCT biosynthesis. First, the regulation of ctcH expression has a direct effect on the generation of NADH and H+, which are the major reductants in the 6-DCT biosynthetic pathway, and second, it affects the amount of acetyl-CoA, a major source of the precursors of 6-DCT.

In ΔctcH, 3-hydroxyacyl-CoA could not be oxidized to a keto group, so the mutation of ctcH reduced the generation of NADH, H+, and acetyl-CoA, further reducing 6-DCT production. One cycle of β-oxidation produces one molecule each of NADH, H+, and acetyl-CoA, which are necessary for the 6-DCT biosynthesis (Fig. 2). However, in ctcH+ and ctcH–ctcJ+, the reaction catalyzed by 3-hydroxylacyl-CoA dehydrogenase was enhanced, increasing the amounts of NADH, H+, and acetyl-CoA, and thus increasing 6-DCT production.

ctcJ encodes a monooxygenase involved in antibiotic biosynthesis, and is also involved in the monooxygenation of intermediates in the macrolide biosynthetic pathways (Gaisser et al. 2002). These oxygenases belong to the so-called post-PKS-modifying (tailoring) enzymes, which play crucial roles in the formation of interesting and unique molecular structures. The monooxygenases are representative, tailoring enzymes that are especially important in providing the structural elements essential for the special biological activity of these structures (Taguchi et al. 2011; Faust et al. 2000).

In ΔctcJ, no 6-DCT was formed, and instead, an unknown chemical was produced that was also produced in DT1. These results indicate that ΔctcJ is unable to catalyze the oxidation steps (Fig. 1) required in the 6-DCT biosynthetic pathway. In ctcJ+ and ctcH–ctcJ+, the oxidation steps involving CtcJ in the 6-DCT biosynthesis pathway were enhanced.

Conclusion

This study mainly focused on the differences in protein expression between S. aureofaciens strains A6-9 and DT1 and the effects of CtcH and CtcJ on the production of 6-DCT in S. aureofaciens. Our experiments demonstrate that CtcH and CtcJ play important roles in the formation of 6-DCT. Further exploration of all the enzymes overexpressed in A6-9 will allow us to clarify the 6-DCT biosynthetic pathway. Our findings should be useful in controlling this metabolic node and modifying the metabolic pathway to increase the production of 6-DCT in an industrial strain.