Skip to main content
NCBI home page
iScience logoLink to iScience
. 2019 Nov 25;22:430–440. doi: 10.1016/j.isci.2019.11.037

Divergent Biosynthesis of C-Nucleoside Minimycin and Indigoidine in Bacteria

Liyuan Kong 1,4, Gudan Xu 1,4, Xiaoqin Liu 1,4, Jingwen Wang 1, Zenglin Tang 1, You-Sheng Cai 1, Kun Shen 1, Weixin Tao 1, Yu Zheng 2, Zixin Deng 1, Neil PJ Price 3, Wenqing Chen 1,5,∗
PMCID: PMC6908994  PMID: 31816530

Summary

Minimycin (MIN) is a C-nucleoside antibiotic structurally related to pseudouridine, and indigoidine is a naturally occurring blue pigment produced by diverse bacteria. Although MIN and indigoidine have been known for decades, the logic underlying the divergent biosynthesis of these interesting molecules has been obscure. Here, we report the identification of a minimal 5-gene cluster (min) essential for MIN biosynthesis. We demonstrated that a non-ribosomal peptide synthetase (MinA) governs “the switch” for the divergent biosynthesis of MIN and the cryptic indigoidine. We also demonstrated that MinCN (the N-terminal phosphatase domain of MinC), MinD (uracil phosphoribosyltransferase), and MinT (transporter) function together as the safeguard enzymes, which collaboratively constitute an unusual self-resistance system. Finally, we provided evidence that MinD, utilizing an unprecedented substrate-competition strategy for self-resistance of the producer cell, maintains competition advantage over the active molecule MIN-5′-monophosphate by increasing the UMP pool in vivo. These findings greatly expand our knowledge regarding natural product biosynthesis.

Subject Areas: Biotechnology, Microbial Biotechnology, Systems Biology

Graphical Abstract

graphic file with name fx1.jpg

Open in a new tab

Highlights

  • •

    A minimal 5-gene cluster (min) is essential for minimycin biosynthesis

  • •

    Divergent biosynthesis of minimycin and indigoidine is mediated by an NRPS enzyme

  • •

    A cascade of three safeguard enzymes constitutes the unusual self-resistance system

  • •

    MinD functions as the key safeguard enzyme by increasing the UMP pool in vivo

Biotechnology; Microbial Biotechnology; Systems Biology

Introduction

The C-nucleoside antibiotics (Figure 1) constitute an important sub-group of microbial natural products with unusual structural features and diverse biological activities (Isono, 1988, Maffioli et al., 2017). Their biosynthesis generally follows a succinct logic, with sequential modifications of simple precursors originating from primary metabolism (Hong et al., 2019, Isono, 1988, Palmu et al., 2017, Sosio et al., 2018, Wang et al., 2019). Less typically, minimycin (MIN, also called oxazinomycin) is produced by diverse bacterial strains, either Streptomyces sp. or Pseudomonas sp., and shows prominent antimicrobial activities against both Gram-positive and Gram-negative bacteria (Kusakabe et al., 1972, Tymiak et al., 1984). It has also demonstrated antitumor activity against transplantable tumors (Kusakabe et al., 1972).

Figure 1.

Figure 1

Open in a new tab

Chemical Structures of Representative C-Nucleoside Antibiotics and Indigoidine

Several C-nucleoside antibiotics (MIN-showdomycin, group I; pyrazofurin-formycin, group II; malayamycin-pseudouridimycin, group III) shown are deduced to employ distinct enzymatic logics for the assembly of the C-glycosidic bond. The structure of indigoidine is highlighted in blue to represent its intrinsic color.

MIN is structurally similar to pseudouridine, a modified nucleoside that is found abundantly in tRNA (Li et al., 2016). The MIN molecule has a unique structural feature in which a 1,3-oxazine 2,4-dione ring and a ribosyl sugar are linked via a C-glycoside bond (Sasaki et al., 1972) (Figure 1). Previous metabolic labeling studies have shown that the ribosyl portion of MIN derives directly from D-ribose (Isono and Suhadolnik, 1975, Isono and Suhadolnik, 1977), and C-6, C-5, and C-4 of the oxazine ring arise from the corresponding C-3, C-4, and C-5 of L-glutamate (Isono and Suhadolnik, 1977). Metabolic feeding experiments by Isono et al. showed that the C-2 of MIN is derived from carbon dioxide (Isono and Suhadolnik, 1975). Concomitant chemical synthesis of MIN was also achieved, with the finding that the synthetic form of MIN shares identical biological features with that from microbial source (De Bernardo and Weigele, 1977).

Indigoidine (Figure 1) is a bipyridyl pigment that was first documented as a microbial natural product in the 1960s (Kuhn et al., 1965), and later discovered to be produced by surprisingly diverse microbes (Heumann et al., 1968, Starr et al., 1966). The indigoidine biosynthetic gene cluster was initially reported from Dickeya dadantii (formerly known as Erwinia chrysanthemi), and it was revealed that it has important roles in pathogenicity and self-resistance to oxidative stress (Reverchon et al., 2002). The gene for indigoidine biosynthesis was later identified from diverse Streptomyces strains (Novakova et al., 2010, Pait et al., 2017, Yu et al., 2013). A single module non-ribosomal peptide synthetase (NRPS), which selectively recognizes L-glutamine as the starter substrate, was necessary for indigoidine biosynthesis (Brown et al., 2017, Takahashi et al., 2007). The indigoidine NRPS gene was subsequently engineered as a promising tool for synthetic biology purposes, either for natural product discovery (Olano et al., 2014) or as a reporter system (Muller et al., 2012, Rezuchova et al., 2018, Xie et al., 2017). More recently, Ankanahalli et al. have created a transgenic blue rose by introduction of a bacterial indigoidine biosynthesis gene (idgS) and a phosphopantetheinyl transferase gene (sfp) from surfactin biosynthesis (Nanjaraj Urs et al., 2019).

In the present study, we address the biosynthesis of C-nucleoside MIN and indigoidine in Streptomyces hygroscopicus JCM 4712. We report that a minimal 5-gene cluster is essential for MIN biosynthesis and show that the divergent biosynthesis of MIN and indigoidine is mediated by an NRPS, MinA. Moreover, we reveal that the N-terminal phosphatase domain of MinC (called MinCN), the MinD uracil phosphoribosyltransferase, and the MinT transporter are safeguard enzymes, which collaboratively constitute an unusual self-resistance system, in which MinD likely employs an unprecedented substrate-competition strategy for self-resistance by increasing the UMP pool in vivo. Our deciphering of the C-nucleoside MIN pathway expands current understanding regarding natural product biosynthesis and self-resistance.

Results and Discussion

Identification of MIN Biosynthetic Gene Cluster from S. hygroscopicus JCM 4712

To identify the gene cluster responsible for MIN biosynthesis, the genome of S. hygroscopicus JCM 4712 was sequenced using the Illumina Hiseq 4000 method, which renders 8.8-Mb data (G + C content 70.31%) after assembly of clean reads. MIN contains a C-glycosidic bond that is structurally similar to that of pseudouridine and showdomycin (Palmu et al., 2017) (Figure 1), implying that they should employ similar enzymatic logic for C-glycosidic bond formation. We therefore utilize pseudouridine 5′-phosphate glycosidase YeiN (GenBank: CAQ32570.1) and C-glycosynthase SdmA (GenBank: KKZ73237.1) as query sequences to conduct individual BLASTP analysis, leading to the discovery of two homologs ORF5178 (designated as MinB, 46%/49% identities to YeiN/SdmA) (Table 1) and ORF3816 (46%/46% identities to YeiN/SdmA) from the genome of S. hygroscopicus JCM 4712 (GenBank: MN397911). On the genomic region surrounding minB is a closely linked gene coding for a non-ribosomal peptide synthetase (MinA) (Figure 2A). The orf3816 gene is linked with a kinase gene (GenBank: MN397911), which is identical to the YeiN-YeiC cascade for the pseudouridine metabolic pathway in E. coli (Preumont et al., 2008). These data suggest that the target region (min) covering minA and minB is likely to be involved in MIN biosynthesis.

Table 1.

Deduced Functions of the Open Reading Frames in the min Gene Cluster

Protein aa Protein Function Homolog, Origin Identity, Similarity (%) Accession No.
MinR 211 FadR family transcriptional regulator SAMN05444521_6508, Streptomyces sp. 3124.6 87, 94 SHI26670
MinT 419 MFS transporter SAMN05444521_6509, Streptomyces sp. 3124.6 79, 85 SHI26674
MinA 1379 NRPS(A-Ox-T-TE-Tau) IndC, S. chromofuscus ATCC 49982 74, 82 AFV27434
MinB 317 C-glycosynthase IndA, S. chromofuscus ATCC 49982 88, 93 AFV27435
MinC 613 HAD phosphatase and DUF4243 domain IndB, S. chromofuscus ATCC 49982 77, 84 AFV27436
MinD 240 Uracil phosphoribosyltransferase Orf2, S. chromofuscus ATCC 49982 81, 90 AFV27437
Open in a new tab

Figure 2.

Figure 2

Open in a new tab

Genetic Organization and Investigation of the min Gene Cluster

(A) Genetic organization of the MIN gene cluster; A, adeylation domain; Ox, oxidase domain; T, thiolation domain; TE, thioesterase domain; Tau, tautomerase domain.

(B) Bioassays of the metabolites produced by related recombinants of S. coelicolor M1154. The indicator strain is Bacillus subtilis.

(C) HPLC analysis of the metabolites produced by related recombinants of S. coelicolor M1154. Std, the authentic standard of MIN; pCHW301, the metabolites of the recombinant S. coelicolor M1154 containing pCHW301; ΔminA, the metabolites of the recombinant S. coelicolor M1154 containing pCHW301ΔminA, and other samples are correspondingly assigned; pSET152, the metabolites of the recombinant S. coelicolor M1154 containing pSET152 as negative control. The aliphatic numbers correspond to those in the bioassay plate.

See also Figures S1–S3; Tables 1 and S1–S4.

To determine the identity of the min gene cluster, we directly cloned a ca. 11.2-kb region (likely housing the whole min gene cluster) using a two-step PCR strategy (Figure S1A; Tables S1 and S2). After confirmation (Figure S1B), the resultant plasmid pCHW301 was transferred into Streptomyces coelicolor M1154 (Gomez-Escribano and Bibb, 2014). The positive conjugants (S. coelicolor M1154::pCHW301) were then fermented for metabolite analysis. A bioassay indicated that the samples of M1154::pCHW301 show apparent inhibition against the indicator strain Bacillus subtilis, but the negative control (S. coelicolor M1154::pSET152) lacks related bioactivity (Figure S1C). High-performance liquid chromatography (HPLC) analysis showed that the sample of M1154::pCHW301 contains a new peak, which is absent from that of the negative control (Figure S1D). Further liquid chromatography-mass spectrometry (LC-MS) analysis shows that the LC peak is able to generate a characteristic [M + H]+ ion at m/z 246.0609, with major fragment ions at m/z 155.9695, 210.1091, and 228.0699, fully consistent with the theoretical fragmentation pattern of MIN (Figures S1E–S1G).

To confirm the identity of the target metabolite accumulated by M1154::pCHW301, it was HPLC purified for 1D and 2D NMR analysis. As anticipated, the 1D NMR data of the target metabolite are closely matched to those of MIN (Figures S2A and S2B), and further detailed assignments of the compound as MIN are supported by 1H-1H COSY (Correlation Spectroscopy) and HMBC (Heteronuclear Multiple Bond Correlation) spectra (Figures S2C and S2D). Analysis of the COSY NMR data led to the identification of a single isolated proton spin system corresponding to the ribose moiety (C-5′, C-4′, C-3′, C-2′, and C-1′), from which the relative configuration was determined based on the analysis of coupling constants. The connection between the ribose moiety and (2H)-1,3-oxazine-2,4-(3H)-dione subunit was deduced from HMBC correlations of H-6 with C-1′; H-1′ with C-4 and C-6; and H-2′ with C-5 (Table S3). Accordingly, the structure of the target metabolite was determined to be the same as that of MIN as shown. Taken together, these data demonstrate that the target gene cluster is responsible for the biosynthesis of MIN.

The Minimal 5-Gene (minTABCD) Cluster Is Essential for MIN Biosynthesis

In silico analysis revealed that the target 11.2-kb region (included in pCHW301) containing six genes is deduced to be involved in MIN biosynthesis. MinR shows 87% identity to SAMN05444521_6508 of Streptomyces sp. 3124.6, which is likely a FadR family transcriptional regulator. The second gene, minT, codes for an MFS transporter with high homology (79% identity) to SAMN05444521_6509 of Streptomyces sp. 3124.6 (Table 1), which is proposed to transport the antibiotic out of the producer cell. The minA product displays significant homology (74% identities in total) to IndC, an NRPS protein from Streptomyces chromofuscus ATCC 49982, and the domain architectures of both enzymes are highly matched (Table 1). MinB is shown to be homologous to IndA (pseudouridine-5′-phosphate glycosidase) with 88% identities (Table 1). Moreover, in silico analysis shows that MinC possesses 77% identities to IndB of S. chromofuscus ATCC 49982 (Table 1). Notably, minC encodes a 613-amino acid protein containing two domains, an N-terminal HAD phosphatase domain and the C-terminal DUF4243 domain with unassigned function. Concerning MinD, it shows 81% identity to Orf2 (uracil phosphoribosyltransferase) of S. chromofuscus ATCC 49982 (Table 1).

To precisely pinpoint the minimal gene cluster for MIN biosynthesis, we individually mutated the target genes by an in vitro CRISPR-Cas9 system (Liu et al., 2015). After confirmation (Figures S3A and S3B; Tables S2 and S4), the pCHW301 variants were conjugated into S. coelicolor M1154, and the resultant recombinants were fermented for further metabolite analysis. The antibacterial bioassay indicated that all the samples, with the exception for those of M1154::pCHW301 and M1154::pCHW301ΔminR, lack bioactivities against the Bacillus subtilis indicator strain (Figure 2B). Moreover, reverse-phase HPLC analysis indicated that the sample from M1154::pCHW301ΔminR could also produce the distinctive peak for MIN (Figure 2C), whose identity was further confirmed by LC-MS analysis (Figures S3C–S3F), suggesting that MIN biosynthesis is not under strict regulation by minR in S. coelicolor. These data demonstrate that a minimal 5-gene cluster (minTABCD) is essential for the maintenance of MIN biosynthesis.

Reconstitution of the MinA-Mediated Pathway for Indigoidine Biosynthesis

Bioinformatic analysis showed that MinA is a typical NRPS protein containing multiple A-Ox-T-TE-Tau domains (Figure S4A; Table 1). More surprisingly, MinA homologs had been previously characterized as the indigoidine synthetases (Takahashi et al., 2007) (Figure 3A), and we were therefore curious about how this enzyme is utilized to build MIN. To evaluate the functional role of MinA, the minA sequence was optimized on the basis of E. coli preference (Table S5), and the resultant plasmid pET28a/minA* (*signifies the optimized sequence) was transferred into E. coli for protein overexpression and metabolite analysis. As anticipated, the broth of the E. coli recombinant showed a distinctive blue color (indigo) (Figure 3B), implicating the potential production of indigo dye. To further determine the identity of the blue pigment, a metabolite sample from the E. coli pET28a/minA* was submitted for HPLC assessment, generating a characteristic peak that was absent from the E. coli strain lacking minA* (negative control) (Figures 3B, S4B, and S4C). Further LC-MS analysis indicated that the LC peak produces an [M + H]+ ion at m/z 249.0615, and corresponding fragment ions at m/z 216.7811 and 231.9472, consistent with those of indigoidine (Figures S4D and S4E). Hence, this demonstrated that the single NRPS (MinA)-mediated pathway is sufficient to support indigoidine biosynthesis.

Figure 3.

Figure 3

Open in a new tab

In Vivo and In Vitro Reconstitution of the MinA Mediated Assembly Line

(A) Schematic of the MinA-mediated assembly line for indigoidine biosynthesis.

(B) Engineered production of indigoidine in E. coli. Top, the target metabolite (blue) produced by related strains. Bottom, HPLC analysis of the target metabolite indigoidine produced by related E. coli strains. pET28a/indC, extracted metabolite of E. coli BAP1 containing indC (indigoidine synthetase gene from S. chromofuscus ATCC 49982) as positive control; pET28a/minA*, extracted metabolite of E. coli BAP1 containing minA*; pET28a, extracted metabolite of E. coli BAP1 containing pET28a as negative control.

(C) The extracted enzymatic products of the MinA reactions. Complete, the MinA reaction with all essential factors added; -Mg2+, the MinA reaction without adding exogenous Mg2+; -FMN, the MinA reaction without adding exogenous FMN; -L-Gln, the MinA reaction without L-glutamine added; -ATP, the MinA reaction without ATP added; -O2, the MinA reaction under N2 atmosphere.

(D) HPLC analysis of the related MinA reactions. The characteristic [M + H]+ ion of indigoidine was also indicated on this panel, and the information for related samples corresponds to that in panel C.

See also Figures S4 and S5; Tables S1 and S5.

To reconstitute MinA activity in vitro, we over-expressed and purified the MinA protein from E. coli. The purified MinA shows bright yellow color (Figures S4F and S4G), a typical feature of a flavin-dependent protein, as confirmed by LC-MS analysis (Figures S4H and S4I). Enzymatic assays with the recombinant MinA produced an obvious blue color, implicating the production of indigoidine (Figures 3C and 3D), whose identity is further confirmed by LC-MS analysis (Figure S4J). The in vitro reaction without enzyme added (negative control) was unable to generate the blue-colored product, and the enzyme reactions in the absence of L-glutamine, ATP, or O2 also generate similar negative results (Figure 3D). Furthermore, the purified MinA has only partial activity without an exogenous addition of FMN or Mg2+ (Figures 3D and S4K–S4M) and indicates the preference to relative low temperatures (18°C of all temperatures tested) (Figures S4N–S4P and S4R). Subsequently, we tested the substrate flexibility for MinA against the unnatural isomer D-glutamine with negative result (Figures S4Q and S4R). These combined data substantially established that divergent biosynthesis of MIN and indigoidine is mediated by the NRPS (MinA) pathway, in which the compound 1 is likely converted to indigoidine via a non-enzymatic spontaneous reaction.

The potential pathways mediated by MinA homologs, as shown by in silico analysis, are actually more widely distributed than we initially imagined. In addition, there is also a high degree of diversity at the genetic level (Figure S5). Earlier studies reported that indigoidine production affects the pathogenicity of the plant pathogen Dickeya dadantii (Reverchon et al., 2002). In the present study, we tentatively propose that this bacterium could perform its pathogenic role by the coupled production of a potential MIN-related compound (Figure S5), and it is of great interest for future study to elucidate such pathogenic mechanism at the molecular level.

MinCN Is Responsible for the Final Dephosphorylation Step, and MinD Functions as a Uracil Phosphoribosyltransferase (UPRTase)

In silico analysis showed that MinC is a two-domain enzyme, consisting of an N-terminal phosphatase domain and a C-terminal domain of undefined function (we have not yet determine the characteristic properties of MinCC in the present study) (Figures S6A and S6B), and that in some cases these are separated into two independent proteins (Figure S6A). To determine if MinCN catalyzes the final dephosphorylation step (Figure 4A), we over-expressed and purified the protein from E. coli to near homogeneity (Figure S6C). This was assayed in vitro against the substrate MIN-MP, which is supplied by the Udk (uridine kinase from Bacillus subtilis, GenBank: QBJ70019.1) (Table S6)-catalyzed reaction (Figures S6D–S6H). LC-MS analysis of the in vitro reaction products indicated that the MinCN reaction is capable of generating a characteristic [M + H]+ ion at m/z 246.0601, with major fragment ions at m/z 155.9384, 210.1541, and 228.0651, consistent with those of the authentic MIN standard (Figures 4B and S6I–S6K). This characteristic MIN peak was absent from the reaction without MinCN added (Figures 4B and S6L). Hence, these data support the hypothesis that MinCN is responsible for the final dephosphorylation step during MIN biosynthesis.

Figure 4.

Figure 4

Open in a new tab

Functional Characterization of MinCN/MinD as Phosphatase/UPRTase

(A) Schematic of MinCN-catalyzed reaction. Udk, uridine kinase (GenBank: QBJ70019.1) from Bacillus subtilis; MIN-5′-MP, MIN-5′-monophosphate.

(B) LC-HRMS analysis of the Udk/MinCN reactions. MIN Std, the authentic standard of MIN; +Udk, the Udk catalyzed reaction to form MIN-MP; -Udk, the reaction without Udk added as negative control; (+Udk)+MinCN, the Udk reaction, proceeded for 4 h, with further addition of MinCN for another 1 h; (+Udk)-MinCN, the Udk reaction (proceeded for 4 h) without further addition of MinCN (also incubated for another 1 h). The characteristic [M + H]+ ions of MIN and MIN-MP are indicated as well in this panel for related peaks.

(C) Schematic of MinD-catalyzed reaction. PRPP, phosphoribosyl pyrophosphate.

(D) HPLC traces of the MinD reactions. Uracil Std, the authentic standard of uracil; UMP Std, the authentic standard of UMP; +MinD, the MinD reaction using uracil and PRPP as substrates; -MinD, the reaction without MinD added as negative control. The characteristic [M + H]+ ions of UMP/uracil are also indicated in this panel.

See also Figures S6 and S7; Tables S1 and S6.

Bioinformatic analysis suggested that minD encodes an UPRTase, which are generally involved in the pyrimidine salvage pathway by catalyzing the reaction between uracil and phosphoribosylpyrophate (PRPP) to regenerate UMP (Singh et al., 2015) (Figures 4C and S7A). To investigate the enzymatic role of MinD, it was over-expressed and purified from E. coli (Figure S7B), and the corresponding activity was determined in vitro. As anticipated, HPLC analysis showed that the MinD-catalyzed reaction could generate a peak for UMP, corresponding to an authentic standard, which was absent from the negative control (Figure 4D). Further LC-MS analysis indicated that the target peak gave rise to a characteristic [M + H]+ ion for UMP at m/z 325.0428, and main fragment ions at m/z 212.9202 and 227.0645, completely consistent with those of the UMP authentic standard (Figures S7C–S7E). These results establish that MinD functions as an UPRTase for the synthesis of UMP.

Divergent Biosynthesis of MIN and Indigoidine Is Mediated by an NRPS-Associated Assembly Line

Previous isotopic feeding experiments had showed that L-glutamate is incorporated into the oxazine ring of MIN (Isono and Suhadolnik, 1977, Suhadolnik and Reichenbach, 1981), whereas its ribosyl moiety is from D-ribose (Isono and Suhadolnik, 1975). In the present work, this mechanism is shown to be reasonable. MIN biosynthesis is confirmed to be initiated by MinA, which specifically selects L-glutamine as substrate, and the tethered amino acid is then dehydrogenated under strict and precise stereospecificity/regiospecificity at the C2-C3 positions by the oxidase domain (Ox domain). Subsequently, compound 1 is released by hydrolysis of the TE domain. Compound 1, as a key branch intermediate, is proposed to be converted to indigoidine by a spontaneous oxidative coupling reaction (Figure 5). Simultaneously, this intermediate could be tautomerized by the Tau domain to form 2, which is modified by the MinB C-glycosynthase to produce 3. Compound 3 subsequently undergoes an unusual oxidative deamination and recombination reaction catalyzed by MinCC to generate MIN-MP, which is followed by the final dephosphorylation step to complete MIN biosynthesis (Figure 5). MinCC features a DUF4243 domain, which is also present in a Baeyer-Villiger oxidase AflY of the aflatoxin biosynthetic pathway (Ehrlich et al., 2005). In this respect, MinCC represents a novel type of Baeyer-Villiger oxidase catalyzing an unprecedented and intriguing oxidative deamination and recombination reaction, and relevant studies to investigate its enzymatic logic are now underway in our laboratory.

Figure 5.

Figure 5

Open in a new tab

Proposed Pathway for the Biosynthesis of MIN and Indigoidine

Proposed pathway for the divergent biosynthesis of MIN and indigoidine. See also Figure S5.

Indigoidine is a well-established microbial metabolite known for over half a century. However, it has now been established for the first time that this pigment is actually associated with the co-production of the C-nucleoside MIN. Accordingly, it is possible that indigoidine biosynthesis from an increasing number of potential pathways may be associated with related unknown C-nucleoside molecules, because analogs of the MinA-MinB pair are in several cases concomitantly present in the specific gene clusters (Figure S5). L-glutamate is the common precursor for the aglycon of the C-nucleoside antibiotics, including MIN, formycin, pyrazofurin, and showdomycin (Suhadolnik and Reichenbach, 1981), which until recently have been unexplored at the molecular level (Hong et al., 2019, Palmu et al., 2017, Sosio et al., 2018, Wang et al., 2019). Hence, we propose that further insight into the molecular logics underlying the biosynthesis of C-nucleoside antibiotics will identify diverse and unique enzymes for potential synthetic biology purposes.

MinCN, MinD, and MinT Functioning as the Safeguard Enzymes Constitute a Collaborative Self-Resistance System during MIN Biosynthesis

The assignment of MinD as a UPRTase raises an interesting open question: what is the functional role of this enzyme during MIN biosynthesis? Considering the fact that MIN is also active against Streptomyces, we tentatively propose that it may play a key role in self-resistance during MIN biosynthesis by employing an unprecedented mechanism. We therefore introduced minD into S. coelicolor M1154 to determine whether it mediates a safeguard role. In contrast to this expectation, the recombinant strain (M1154::pIB139minD) had no apparent resistance to MIN (10 μg/mL). Accordingly, we re-examined the MIN gene cluster and noticed that minCN and minT are also potential contributors to the self-resistance system. To test the assumption, we individually introduced these two genes into S. coelicolor M1154. However, neither gene was capable of supporting the growth of the corresponding S. coelicolor M1154 recombinants in the presence of MIN (10 μg/mL, the minimal inhibitory concentration against S. coelicolor M1154-derived negative controls in this study) (Figures 6A and 6B).

Figure 6.

Figure 6

Open in a new tab

Proposed Self-Resistance Mechanism for the Biosynthesis of MIN

(A) Plate-grown experiments for S. coelicolor M1154 recombinants containing related genes. For the first two plates, the recombinant incubated for 36 h (in the related square) corresponds to each other. For plates 2–4, it means the identical plate that was incubated for different time (marker at the bottom of each plate); MIN (-), the MIN-negative plate; MIN (10 μg/mL), the plate containing MIN at a final concentation of 10 μg/mL. The recombinant in the related square corresponds to that described as below.

(B) Representational map for the plate-grown experiments. pIB139, S. coelicolor M1154 containing the empty vector pIB139; pIB139D, S. coelicolor M1154 containing pIB139/minD; pIB139T, S. coelicolor M1154 containing pIB139/minT; pIB139CN, S. coelicolor M1154 containing pIB139/minCN; pIB139DT, S. coelicolor M1154 containing pIB139/minDT; pIB139DCN, S. coelicolor M1154 containing pIB139/minDCN; pIB139TCN, S. coelicolor M1154 containing pIB139/minTCN; pIB139DTCN, S. coelicolor M1154 containing pIB139/minDTCN.

(C) The relative abundance of the UMP concentrations in vivo for the strains at different growth stages (48 and 72 h). pSET152, the sample of S. coelicolor M1154::pSET152 (blue); pCHW301, the sample of S. coelicolor M1154::pCHW301 (red); “RA” denotes relative abundance. Data are represented as mean ± SEM.

(D) Proposed collaborative self-resistance system during MIN biosynthesis. Three enzymes, including MinCN, MinT, and MinD, collaborate to fulfill the mission of self-resistance during MIN biosynthesis, and MinD, acting as the key safeguard enzyme, employs an unprecedented strategy of substrate competition to achieve self-resistance during MIN biosynthesis.

See also Figure S7 and Table S1.

These results suggested that some combination of minCN, minD, and minT may collaborate to fulfill the self-resistance role, and we accordingly introduced these different combinations into S. coelicolor M1154. Notably, several of these combinations are able to support the effective growth of the corresponding S. coelicolor M1154 recombinants in the presence of MIN, whereas the introduced single minCN, minD, or minT could not maintain the apparent growth of the related recombinants. It was also observed that minD has a greater contribution to the self-resistance system, and the combination (containing minCN, minD, and minT) could maintain the optimal growth status for the counterpart recombinant (Figures 6A and 6B). Together, these data demonstrate that minD, in collaboration with minCN and minT, determines the functional role in vivo constituting the self-resistance system during MIN biosynthesis.

Biosynthesis of MIN Employs a Substrate-Competition Strategy for Self-Resistance

A further relevant question is how MIN executes its inhibitory role in vivo? We propose that similar to many other nucleoside analogs (Jordheim et al., 2013), the phosphorylated form (MIN-MP) of MIN functions as the active inhibitory molecule. This is also supported by the discovery of kinases (from S. coelicolor M1154 and Bacillus subtilis) capable of recognizing MIN as substrates (Figures S6C–S6H), and the fact that MinD (UPRTase) may function as a safeguard enzyme by enhancing the in vivo UMP pool of microbial cells. To test this suggestion, we conducted bioassays using Bacillus subtilis as an indicator strain by exogenous addition of uridine and MIN at different concentration ratios. This strain showed resistance to MIN in the co-presence of relative high concentration of uridine (Figure S7F), implying that uridine, which is efficiently metabolized to UMP in vivo, is capable of partly alleviating the inhibition of MIN. Moreover, the in vivo UMP concentrations for the S. coelicolor M1154 recombinants, as determined by LC-high-resolution MS (HRMS), are considerably higher than those of the negative control (S. coelicolor M1154::pSET152) (Figures 6C and S7G), providing further support for the self-resistance model employed in MIN biosynthesis. In contrast, the in vivo MIN-MP (the active form of MIN), as indicated by LC-HRMS, is undetectable due to the relatively much lower concentration (if compared with that of UMP) (Figures S7H and S7I). These data support the hypothesis that the inhibitory role of MIN is attenuated by UMP by an unprecedented substrate-competition mechanism.

Biosynthetic bacteria have evolved several strategies for self-resistance during natural product production, often transporting the metabolite out of the producer cell upon its synthesis, or by modification of natural products or the target site (Wencewicz, 2019). The substrate-competition strategy for self-resistance presented here is, to the best of our knowledge, unique, but it may be broadly exploited for microbial natural products biosynthesis. In this self-resistance system, the active inhibitory molecule MIN-MP is immediately dephosphorylated to form the end product MIN, most of which will be promptly transported out of the producer cells. MinD, therefore, functions as a safeguard enzyme that responds to sub-inhibitory concentrations of MIN-MP by increasing the UMP pool in vivo, thereby achieving the self-resistance for the producer cell (Figure 6D).

Limitations of the Study

In this study, we have uncovered that divergent biosynthesis of MIN and indigoidine is mediated by an NRPS enzyme and have demonstrated that an unprecedented collaborative self-resistance system is employed for MIN biosynthesis. However, the precise mechanism of how MIN executes its inhibitory role will require further study. Further studies will also be required to understand the enzymatic formation of the oxazine ring substituent of MIN.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.

Acknowledgments

We are grateful to Prof. Jixun Zhan (Utah State University, USA) for kindly providing us with related research material. This work was supported by grants National Key R & D Program of China (2018YFA0903203), the National Natural Science Foundation of China (31770041, 31970052), and the foundation of Tianjin Engineering Research Center of Microbial Metabolism and Fermentation Process Control, China (ZXKF20180202).

Author Contributions

L.K., G.X., X.L., J.W., and Z.T. conducted the genetic and biochemical experiments. W.T. helped with the preparation of figures. Y.-S.C. and K.S. analyzed the NMR data. W.C. conceived the project and directed the research. W.C. wrote the manuscript, and Y.Z., Z.D., and N.P.J.P. made a critical reading of the manuscript.

Declaration of Interests

The authors declare no competing interests.

Published: December 20, 2019

Footnotes

Supplemental Information can be found online at https://doi.org/10.1016/j.isci.2019.11.037.

Data and Code Availability

The DNA sequence is deposited in the GenBank database under individual accession numbers MK122964 (for the min gene cluster from S. hygrocsopicus JCM 4712) and MN397911 (for the genes orf3815 and orf3816 from S. hygrocsopicus JCM 4712).

Supplemental Information

Document S1. Transparent Methods, Figures S1–S7, and Tables S1–S6
mmc1.pdf (3.9MB, pdf)

References

  1. Brown A.S., Robins K.J., Ackerley D.F. A sensitive single-enzyme assay system using the non-ribosomal peptide synthetase BpsA for measurement of L-glutamine in biological samples. Sci. Rep. 2017;7:41745. doi: 10.1038/srep41745. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. De Bernardo S., Weigele M. Synthesis of oxazinomycin (minimycin) J. Org. Chem. 1977;42:109–112. doi: 10.1021/jo00421a021. [DOI] [PubMed] [Google Scholar]
  3. Ehrlich K.C., Montalbano B., Boue S.M., Bhatnagar D. An aflatoxin biosynthesis cluster gene encodes a novel oxidase required for conversion of versicolorin a to sterigmatocystin. Appl. Environ. Microbiol. 2005;71:8963–8965. doi: 10.1128/AEM.71.12.8963-8965.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Gomez-Escribano J.P., Bibb M.J. Heterologous expression of natural product biosynthetic gene clusters in Streptomyces coelicolor: from genome mining to manipulation of biosynthetic pathways. J. Ind. Microbiol. Biotechnol. 2014;41:425–431. doi: 10.1007/s10295-013-1348-5. [DOI] [PubMed] [Google Scholar]
  5. Heumann W., Young D., Gottlich C. Leucoindigoidine formation by an Arthrobacter species and its oxidation to indigoidine by other micro-organisms. Biochim. Biophys. Acta. 1968;156:429–431. doi: 10.1016/0304-4165(68)90280-8. [DOI] [PubMed] [Google Scholar]
  6. Hong H., Samborskyy M., Zhou Y., Leadlay P.F. C-nucleoside formation in the biosynthesis of the antifungal malayamycin A. Cell Chem. Biol. 2019;26:493–501.e5. doi: 10.1016/j.chembiol.2018.12.004. [DOI] [PubMed] [Google Scholar]
  7. Isono K. Nucleoside antibiotics: structure, biological activity, and biosynthesis. J. Antibiot. (Tokyo) 1988;41:1711–1739. doi: 10.7164/antibiotics.41.1711. [DOI] [PubMed] [Google Scholar]
  8. Isono K., Suhadolnik R.J. The biosynthesis of the nucleoside antibiotics: minimycin formation by Streptomyces hygroscopicus. Ann. N. Y Acad. Sci. 1975;255:390–401. doi: 10.1111/j.1749-6632.1975.tb29247.x. [DOI] [PubMed] [Google Scholar]
  9. Isono K., Suhadolnik R.J. Biosynthesis of the C-nucleoside, minimycin: asymmetric incorporation of glutamate and acetate into the oxazine ring. J. Antibiot. (Tokyo) 1977;30:272–273. doi: 10.7164/antibiotics.30.272. [DOI] [PubMed] [Google Scholar]
  10. Jordheim L.P., Durantel D., Zoulim F., Dumontet C. Advances in the development of nucleoside and nucleotide analogues for cancer and viral diseases. Nat. Rev. Drug Discov. 2013;12:447–464. doi: 10.1038/nrd4010. [DOI] [PubMed] [Google Scholar]
  11. Kuhn R., Starr M.P., Kuhn D.A., Bauer H., Knackmuss H.J. Indigoidine and other bacterial pigments related to 3,3'-Bipyridyl. Arch. Mikrobiol. 1965;51:71–84. doi: 10.1007/BF00406851. [DOI] [PubMed] [Google Scholar]
  12. Kusakabe Y., Nagatsu J., Shibuya M., Kawaguchi O., Hirose C. Minimycin, a new antibiotic. J. Antibiot. (Tokyo) 1972;25:44–47. doi: 10.7164/antibiotics.25.44. [DOI] [PubMed] [Google Scholar]
  13. Li X.Y., Ma S.Q., Yi C.Q. Pseudouridine: the fifth RNA nucleotide with renewed interests. Curr. Opin. Chem. Biol. 2016;33:108–116. doi: 10.1016/j.cbpa.2016.06.014. [DOI] [PubMed] [Google Scholar]
  14. Liu Y., Tao W., Wen S., Li Z., Yang A., Deng Z., Sun Y. In Vitro CRISPR/Cas9 System for efficient targeted DNA editing. MBio. 2015;6 doi: 10.1128/mBio.01714-15. e01714–01715. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Maffioli S.I., Zhang Y., Degen D., Carzaniga T., Del Gatto G., Serina S., Monciardini P., Mazzetti C., Guglierame P., Candiani G. Antibacterial nucleoside-analog inhibitor of bacterial RNA polymerase. Cell. 2017;169:1240–1248.e23. doi: 10.1016/j.cell.2017.05.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Muller M., Auslander S., Auslander D., Kemmer C., Fussenegger M. A novel reporter system for bacterial and mammalian cells based on the non-ribosomal peptide indigoidine. Metab. Eng. 2012;14:325–335. doi: 10.1016/j.ymben.2012.04.002. [DOI] [PubMed] [Google Scholar]
  17. Nanjaraj Urs A.N., Hu Y., Li P., Yuchi Z., Chen Y., Zhang Y. Cloning and expression of a nonribosomal peptide synthetase to generate blue rose. ACS Synth. Biol. 2019;8:1698–1704. doi: 10.1021/acssynbio.8b00187. [DOI] [PubMed] [Google Scholar]
  18. Novakova R., Odnogova Z., Kutas P., Feckova L., Kormanec J. Identification and characterization of an indigoidine-like gene for a blue pigment biosynthesis in Streptomyces aureofaciens CCM 3239. Folia Microbiol. (Praha) 2010;55:119–125. doi: 10.1007/s12223-010-0018-5. [DOI] [PubMed] [Google Scholar]
  19. Olano C., Garcia I., Gonzalez A., Rodriguez M., Rozas D., Rubio J., Sanchez-Hidalgo M., Brana A.F., Mendez C., Salas J.A. Activation and identification of five clusters for secondary metabolites in Streptomyces albus J1074. Microb. Biotechnol. 2014;7:242–256. doi: 10.1111/1751-7915.12116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Pait I.G.U., Kitani S., Kurniawan Y.N., Asa M., Iwai T., Ikeda H., Nihira T. Identification and characterization of lbpA, an indigoidine biosynthetic gene in the gamma-butyrolactone signaling system of Streptomyces lavendulae FRI-5. J. Biosci. Bioeng. 2017;124:369–375. doi: 10.1016/j.jbiosc.2017.04.020. [DOI] [PubMed] [Google Scholar]
  21. Palmu K., Rosenqvist P., Thapa K., Ilina Y., Siitonen V., Baral B., Makinen J., Belogurov G., Virta P., Niemi J. Discovery of the showdomycin gene cluster from Streptomyces showdoensis ATCC 15227 yields insight into the biosynthetic logic of C-nucleoside antibiotics. ACS Chem. Biol. 2017;12:1472–1477. doi: 10.1021/acschembio.7b00078. [DOI] [PubMed] [Google Scholar]
  22. Preumont A., Snoussi K., Stroobant V., Collet J.F., Van Schaftingen E. Molecular identification of pseudouridine-metabolizing enzymes. J. Biol. Chem. 2008;283:25238–25246. doi: 10.1074/jbc.M804122200. [DOI] [PubMed] [Google Scholar]
  23. Reverchon S., Rouanet C., Expert D., Nasser W. Characterization of indigoidine biosynthetic genes in Erwinia chrysanthemi and role of this blue pigment in pathogenicity. J. Bacteriol. 2002;184:654–665. doi: 10.1128/JB.184.3.654-665.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Rezuchova B., Homerova D., Sevcikova B., Nunez L.E., Novakova R., Feckova L., Skultety L., Cortes J., Kormanec J. An efficient blue-white screening system for markerless deletions and stable integrations in Streptomyces chromosomes based on the blue pigment indigoidine biosynthetic gene bpsA. Appl. Microbiol. Biotechnol. 2018;102:10231–10244. doi: 10.1007/s00253-018-9393-7. [DOI] [PubMed] [Google Scholar]
  25. Sasaki K., Kusakabe Y., Esumi S. The structure of minimycin, a novel carbon-linked nucleoside antibiotic related to -pseudouridine. J. Antibiot. (Tokyo) 1972;25:151–154. doi: 10.7164/antibiotics.25.151. [DOI] [PubMed] [Google Scholar]
  26. Singh V., Brecik M., Mukherjee R., Evans J.C., Svetlikova Z., Blasko J., Surade S., Blackburn J., Warner D.F., Mikusova K. The complex mechanism of antimycobacterial action of 5-fluorouracil. Chem. Biol. 2015;22:63–75. doi: 10.1016/j.chembiol.2014.11.006. [DOI] [PubMed] [Google Scholar]
  27. Sosio M., Gaspari E., Iorio M., Pessina S., Medema M.H., Bernasconi A., Simone M., Maffioli S.I., Ebright R.H., Donadio S. Analysis of the pseudouridimycin biosynthetic pathway provides insights into the formation of C-nucleoside antibiotics. Cell Chem. Biol. 2018;25:540–549.e4. doi: 10.1016/j.chembiol.2018.02.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Starr M.P., Cosens G., Knackmuss H.J. Formation of the blue pigment indigoidine by phytopathogenic erwinia. Appl. Microbiol. 1966;14:870–872. doi: 10.1128/am.14.6.870-872.1966. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Suhadolnik R.J., Reichenbach N.L. Glutamate as the common precursor for the aglycon of the naturally occurring C-nucleoside antibiotics. Biochemistry. 1981;20:7042–7046. doi: 10.1021/bi00527a042. [DOI] [PubMed] [Google Scholar]
  30. Takahashi H., Kumagai T., Kitani K., Mori M., Matoba Y., Sugiyama M. Cloning and characterization of a Streptomyces single module type non-ribosomal peptide synthetase catalyzing a blue pigment synthesis. J. Biol. Chem. 2007;282:9073–9081. doi: 10.1074/jbc.M611319200. [DOI] [PubMed] [Google Scholar]
  31. Tymiak A.A., Culver C.A., Goodman J.F., Seiner V.S., Sykes R.B. Oxazinomycin produced by a Pseudomonas species. J. Antibiot. (Tokyo) 1984;37:416–418. doi: 10.7164/antibiotics.37.416. [DOI] [PubMed] [Google Scholar]
  32. Wang S.A., Ko Y., Zeng J., Geng Y., Ren D., Ogasawara Y., Irani S., Zhang Y., Liu H.W. Identification of the formycin A biosynthetic gene cluster from Streptomyces kaniharaensis illustrates the interplay between biological pyrazolopyrimidine formation and de novo purine biosynthesis. J. Am. Chem. Soc. 2019;141:6127–6131. doi: 10.1021/jacs.9b00241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Wencewicz T.A. Crossroads of antibiotic resistance and biosynthesis. J. Mol. Biol. 2019;431:3370–3399. doi: 10.1016/j.jmb.2019.06.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Xie Z., Zhang Z., Cao Z., Chen M., Li P., Liu W., Qin H., Zhao X., Tao Y., Chen Y. An external substrate-free blue/white screening system in Escherichia coli. Appl. Microbiol. Biotechnol. 2017;101:3811–3820. doi: 10.1007/s00253-017-8252-2. [DOI] [PubMed] [Google Scholar]
  35. Yu D., Xu F., Valiente J., Wang S., Zhan J. An indigoidine biosynthetic gene cluster from Streptomyces chromofuscus ATCC 49982 contains an unusual IndB homologue. J. Ind. Microbiol. Biotechnol. 2013;40:159–168. doi: 10.1007/s10295-012-1207-9. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Document S1. Transparent Methods, Figures S1–S7, and Tables S1–S6
mmc1.pdf (3.9MB, pdf)

Data Availability Statement

The DNA sequence is deposited in the GenBank database under individual accession numbers MK122964 (for the min gene cluster from S. hygrocsopicus JCM 4712) and MN397911 (for the genes orf3815 and orf3816 from S. hygrocsopicus JCM 4712).

Articles from iScience are provided here courtesy of Elsevier

RESOURCES