GeneralStrains, plasmids, and oligonucleotides used in this study are listed in Supplementary Tables 2, 3. Molecular biology experiments were conducted following the manufacturer’s instructions. Oligonucleotides for the polymerase chain reaction (PCR) were purchased from Eurofins Genomics (Tokyo, Japan). PCR amplification was performed using PrimeSTARTM HS DNA polymerase (Takara Bio Inc., Shiga, Japan) or KOD FX DNA polymerase (TOYOBO CO., LTD., Osaka, Japan). PCR products were purified using a QIAquick PCR Purification Kit (QIAGEN GmbH, Hilden, Germany). Restriction enzymes were purchased from New England Biolabs (Ipswich, MA, USA). DNA ligation was performed using DNA Ligation Kit Ver. 2.1 (Takara Bio Inc., Shiga, Japan). Commercial spinach Fd, spinach FNR, and NADPH (tetrasodium salt) were purchased from Sigma-Aldrich Co. (St. Louis, MO, USA). Synthetic DNAs (Supplementary Table 4) for protein expression of Fds were obtained from Eurofins Genomics (Tokyo, Japan). The Q5 Site-Directed Mutagenesis Kit was purchased from New England Biolabs (Ipswich, MA, USA). Lysozyme from chicken egg white was purchased from Sigma-Aldrich Co. (St. Louis, MO, USA). TurboNuclease was purchased from Accelagen Inc. (San Diego, CA, USA). Precision Plus Protein standards were purchased from Bio-Rad Laboratories (Hercules, CA, USA).All chemicals were commercially obtained and used without further purification. Unless specifically mentioned, the LC–MS analysis was performed on a Waters ACQUITY UPLC H-Class System equipped with ACQUITY QDa Detector (Waters, Milford, MA, USA) and an AB Sciex API3200 system using ESI probe (AB Sciex, Framingham, MA, USA), under control of Empower 3 or Empower 2 for UPLC and Analyst 1.5.1 for API3200, on a Waters ACQUITY UPLC BEH C18 Column (2.1 mm i.d. × 50 mm, 1.7 µm). Medium-pressure liquid chromatography (MPLC) was performed on a CombiFlash companion personal flash chromatography system (Teledyne ISCO, Lincoln, NE, USA) equipped with a RediSep C18 column (80 g). Preparative HPLC analysis was performed on a Waters 600E pump system using a SenshuPak PEGASIL ODS column (20 mm i.d. × 250 mm or 10 mm i.d. × 250 mm, 5 µm) (Senshu Scientific Co., Ltd, Tokyo, Japan). UV–vis spectrum, optical rotations, and IR spectrum were recorded using a JASCO V-630 BIO spectrophotometer (JASCO International, Tokyo, Japan), a HORIBA SEPA-300 high-sensitive polarimeter (HORIBA, Kyoto, Japan), and a HORIBA FT-720 spectrometer with a DuraSampl IR II ATR instrument under control of FT-IR for Windows version 4.07, respectively. HR-ESI-TOF-MS analysis was performed using a SYNAPT G2 Mass Spectrometer under control of MassLynx V4.1. 1H-NMR (at 500 MHz) and 13C-NMR (at 125 MHz) data were obtained on a JEOL ECA-500 FT-NMR spectrometer under control of Delta ver. 5.0.4 (JEOL, Tokyo, Japan). Chemical shifts were reported in ppm, referencing corresponding solvent signals (δH 1.94 and δC 1.39 for acetonitrile-d3).In-frame deletion of vtlG gene, transformation, and metabolites profiling in S. avermitilis SUKA17To perform in-frame deletion of the vtlG gene in pKU503vtl, a PCR targeting and λ-red recombination-based gene replacement approach was used. Plasmid pKD13::aac(3)IV was a template to PCR-amplify the FRT-flanked aac(3)IV gene cassette, which contained upstream and downstream homologous arms of the vtlG gene, using primer set pKD13-Apr-Fwd and pKD13-Apr-Rev. The PCR products and the plasmid pKU503vtl were transformed into E. coli BW25113/pKD46 to replace the vtlG gene with aac(3)IV gene. The resultant recombinant plasmid was further transformed into E. coli XL1-Blue MRF’ strain by electroporation to eliminate the aac(3)IV gene, giving the pKU503vtl::ΔvtlG. Successful deletion of the vtlG gene was confirmed by PCR. After transformation into E. coli GM2929 hsdS::Tn10 by electroporation, the unmethylated form of pKU503vtl::ΔvtlG plasmid was isolated and subsequently introduced into the S. avermitilis SUKA17 host via polyethene glycol-assisted protoplast transformation58. The pKU492aac(3)IV-sav2794p-vtlR plasmid containing a LuxR-family transcriptional regulator vtlR gene under the control of sav2794 promoter15 was also introduced into the same strain. The generated S. avermitilis SUKA17/pKU503vtl::ΔvtlG/pKU492aac(3)IV-sav2794p-vtlR (designated vtlG disruptant) was subjected to metabolites profiling.Metabolites of the vtlG disruptant were profiled following a similar method to our previous report15. In brief, the vtlG disruptant was pre-cultured in 10 mL of SK2 medium with neomycin (final 0.2 µg/mL) and apramycin (final 0.2 µg/mL) at 28 °C with rotary shaking at 250 rpm for 3 days. Subsequently, 1 mL of the pre-culture was inoculated into a 500 mL cylindrical flask containing 70 mL of 0.3× BPS medium and cultured at 28 °C with rotary shaking at 150 rpm for 5 days. The culture broth was mixed with an equivalent volume of acetone, filtered under reduced pressure to remove mycelia, and evaporated in vacuo to remove acetone. The remaining aqueous layer was extracted twice with an equivalent ethyl acetate (EtOAc) volume, dried in vacuo, and re-dissolved in methanol (MeOH) for UPLC analysis.Isolation of compound 4Large-scale fermentation (7 L) of the vtlG disruptant was conducted to isolate compound 4 for enzymatic reactions. The vtlG disruptant was pre-cultured in 9 test tubes, each containing 10 mL SK2 medium with neomycin (final 0.2 µg/mL) and apramycin (final 0.2 µg/mL) at 28 °C with rotary shaking at 250 rpm for 3 days. Subsequently, 0.7 mL of the pre-culture was inoculated into 100 cylindrical flasks (500 mL) containing 70 mL of 0.3× BPS medium and cultured at 28 °C with rotary shaking at 150 rpm for 8 days. The culture broth was treated with acetone following a similar procedure described above. The remaining aqueous layer was extracted twice with an equivalent volume of EtOAc. The EtOAc extract was evaporated with a small amount of silica gel resin. The extract–resin mixture was then fractionated with silica gel column chromatography (65 i.d. × 110 mm) using a stepwise solvent system of CHCl3:MeOH (100:0, 98:2, 95:5, 90:10, 0:100). The target fraction (CHCl3:MeOH; 95:5) was subsequently fractionated by ODS chromatography (40 i.d. × 60 mm) using a stepwise solvent system of H2O:MeOH (50:50, 40:60, 30:70, 20:80, 0:100). The 30:70 fraction was purified again under the same condition. Finally, the target fractions (H2O:MeOH; 20:80 and 0:100) were purified with preparative HPLC using an isocratic solvent system of 40% acetonitrile (CH3CN) with 0.015% formic acid at 6 mL/min. The target peak was collected to give 0.6 mg of 4 as a yellow powder.Isolation of compound 5Large-scale fermentation (36 L) of the vtlG disruptant was conducted to isolate compound 5 for structure determination. The vtlG disruptant was pre-cultured in a test tube containing 10 mL SK2 medium with neomycin (final 0.2 µg/mL) and apramycin (final 0.2 µg/mL) at 28 °C with rotary shaking at 250 rpm for 2 days. Next, 1 mL pre-culture was inoculated into four cylindrical flasks (500 mL) containing 70 mL of 0.3× BPS medium and cultured at 28 °C with rotary shaking at 150 rpm for 2 days to make the seed culture. Subsequently, 3 mL of the seed culture was inoculated into 500 mL cylindrical flasks containing 70 mL of 0.3× BPS medium and cultured at 28 °C with rotary shaking at 150 rpm for 4 days. The culture broth was centrifuged at 3500 × g for 10 min to separate the supernatant and mycelia. The supernatant was extracted thrice with a half volume of EtOAc. The mycelia were resuspended in 1 L of distilled water, mixed with 2 L of acetone, and stirred for 15 h at room temperature. The mycelia–acetone mixture was filtered to remove mycelia, evaporated to remove acetone, and then extracted thrice with a half volume of EtOAc. After that, all the EtOAc layers were combined and added with sodium sulfate anhydrous to remove water traces. This EtOAc layer was evaporated to yield ~3 g of brownish oily crude extract. Next, the crude extract was re-dissolved with a small amount of MeOH, applied to a Sephadex LH-20 column (40 mm i. d. × 250 mm), and eluted with MeOH to give 20 fractions. The 9th (1 g) and 10th (317 mg) fractions were further separated with MPLC using a linear gradient of 20–100% CH3CN in 100 column volume at 60 mL/min to give 120 fractions. The MPLC fractions 22–27 were purified with preparative HPLC using an isocratic solvent system of 72% CH3CN at a 5 mL/min flow rate. The target peak was collected to give 1 mg of 5 as colourless amorphous.In vitro P450 VtlG assay using 4 and 5Time-dependent conversion of compound 4 as a substrate was conducted at 30 °C in a reaction mixture (600 μL) containing 50 mM Tris-HCl (pH 7.5), 1 μM 4 (dissolved in DMSO), 0.5 μM purified VtlG, 0.1 mg/mL commercial spinach Fd, and 1 unit/mL spinach FNR. Compound 4 was quantified according to a standard curve generated based on the peak area at 350 nm using UPLC. After pre-incubation at 30 °C for 2 min, the reaction was initiated by rapidly adding 100 μM NADPH. At each desired time point (1, 30, 60 min), a 200 μL aliquot of the reaction mixture was taken, quenched with 5 μL acetic acid, and extracted twice with 400 μL EtOAc. An appropriate DMSO volume (20 μL) was added before evaporating EtOAc with nitrogen gas. The remaining DMSO fraction (15 μL) was used for UPLC analysis with a linear gradient of 10–100% CH3CN with 0.1% formic acid in 3 min at 0.5 mL/min.The conversion of compound 5 was also examined at 30 °C in a 200 μL reaction mixture containing 50 mM Tris-HCl (pH 7.5), 1 μM 5 (dissolved in DMSO), 0.5 μM purified VtlG, 0.1 mg/mL commercial spinach Fd, and 1 unit/mL spinach FNR. After pre-incubation at 30 °C for 2 min, the reaction was initiated by adding 1 mM NADPH rapidly. After 30 min incubation, the reaction mixture was quenched with 5 μL acetic acid and extracted twice using 400 μL EtOAc. An appropriate DMSO volume (20 μL) was added before evaporating EtOAc with nitrogen gas. The remaining DMSO fraction (15 μL) was used for UPLC analysis with a linear gradient of 10–100% CH3CN with 0.1% formic acid in 3 min at 0.5 mL/min.Kinetic analysis of P450 VtlG against 4To determine the kinetic parameters of the VtlG-catalysed mono-hydroxylation reaction against 4, the reaction mixture (200 μL) contained 50 mM Tris-HCl (pH 7.5), 0.25–2.5 μM 4 (dissolved in DMSO), 5 nM purified VtlG, 0.1 mg/mL commercial spinach Fd, and 1 unit/mL spinach FNR. After pre-incubation at 30 °C for 2 min, the reaction was started by rapidly adding 1 mM NADPH and incubating for another 2 min. The peak area at 350 nm by UPLC was used to quantify the reaction product of 4. The Km and kcat values were obtained by nonlinear curve fitting using the Michaelis–Menten equation in SigmaPlot 12 software (Systat Software, Inc., San Jose, CA, USA).Preparation of 6 by VtlG reactionLarge-scale VtlG reaction to prepare 6 was conducted at 30 °C in a 1 mL reaction mixture containing 50 mM Tris-HCl (pH 7.5), 5 μM 4 (dissolved in DMSO), 0.5 μM purified VtlG, 0.1 mg/mL spinach Fd, and 1 unit/mL spinach FNR. After pre-incubation at 30 °C for 2 min, the reaction was initiated by adding 10 mM NADPH to an excess amount to inhibit [4 + 2] cycloaddition reaction. After incubation for 5 min, the reaction mixture was extracted twice with 1 mL EtOAc. An appropriate DMSO volume (100 μL) was added before evaporating EtOAc with nitrogen gas. The remaining DMSO fraction (80 μL) containing the hydroxylated compound 6 was prepared as a substrate for the subsequent [4 + 2] cycloaddition assays. Compound 6 was quantified according to a standard curve generated based on the peak area at 350 nm using UPLC.In vitro conversion of 4 and 6 by commercial spinach FdEnzyme assays of commercial spinach Fd were conducted at 30 °C in a 200 μL reaction mixture containing 50 mM Tris-HCl (pH 7.5), 1 μM 4 or 6 as reactant (dissolved in DMSO), and 0.5 mg/mL (47 μM) commercial spinach Fd. After pre-incubation at 30 °C for 2 min, the reaction was initiated by adding the reactant. The negative control was the boiled (95 °C, 15 min) spinach Fd. After the reaction, the mixture was extracted twice with 400 μL EtOAc. Then, an appropriate volume (20 μL) of DMSO was added before evaporating EtOAc with nitrogen gas. The remaining DMSO fraction (15 μL) was used for UPLC analysis with a linear gradient of 10–100% CH3CN with 0.1% formic acid at 0.5 mL/min in 3 min for the reaction product of 6 and 12 min for the reaction product of 4.Kinetic analysis of spinach Fd against 4To determine the kinetic parameters of the spinach Fd-catalysed [4 + 2] cycloaddition reaction against 4, the reaction mixture (200 μL) contained 50 mM Tris-HCl (pH 7.5), 0.25–3 μM 4 (dissolved in DMSO), 24 μM commercial spinach Fd. After pre-incubation at 30 °C for 2 min, the reaction was started by rapid addition of 4 and incubated for another 4 min. Compound 5 was quantified according to a standard curve generated based on the peak area at 280 nm using UPLC. The Km and kcat values were obtained by nonlinear curve fitting using the Michaelis–Menten equation in SigmaPlot 12 software.In vitro conversion of 4 and 6 by recombinant FdsEnzyme assays of each recombinant Fds were conducted at 30 °C in a 200 μL reaction mixture containing 50 mM Tris-HCl (pH 7.5), 1 μM 4 or 6 as reactant (dissolved in DMSO), and 50 μM recombinant Fd. After pre-incubation at 30 °C for 2 min, the reaction was initiated by adding the reactant. After the reaction, the mixture was extracted twice with 400 μL EtOAc. An appropriate DMSO volume (20 μL) was added before evaporating EtOAc with nitrogen gas. The remaining DMSO fraction (15 μL) was used for UPLC analysis with a linear gradient of 10–100% CH3CN with 0.1% formic acid at 0.5 mL/min in 3 min for the reaction product of 6 and 12 min for the reaction product of 4.Kinetic analysis of MirFd against 4To determine the kinetic parameters of the MirFd-catalysed [4 + 2] cycloaddition reaction against 4, the reaction mixture (200 μL) contained 50 mM Tris-HCl (pH 7.5), 0.5–3 μM 4 (dissolved in DMSO), and 10 μM recombinant MirFd. After pre-incubation at 30 °C for 2 min, the reaction was initiated by rapidly adding 4 and incubating for another 2 min. Compound 5 was quantified according to a standard curve generated based on peak area at 280 nm using UPLC. The Km and kcat values were obtained by nonlinear curve fitting using the Michaelis–Menten equation in SigmaPlot 12 software.Preparation of apo-FdsEDTA treatment of spinach Fd and MirFd was performed following a minorly modified method33. Briefly, the purified recombinant Fds (1 mg) were boiled at 95 °C in 100 mM EDTA and 50 mM DTT in 1.5 mL Eppendorf tubes. The tubes were inverted until the proteins became colourless. After centrifugation (4 °C; 8000 × g; 10 min), the supernatant was further centrifuged with Amicon® Ultra-15 (10 kDa cutoff) centrifugal filter to remove iron atoms, sulfide, and excess EDTA and DTT. Absorption spectra were analysed. The decrease of absorption maxima at 463, 420, and 325 mm for spinach Fd and 418 mm for MirFd indicated the Fe–S cluster removal. The concentration of spinach Fd and MirFd apo-forms was determined using Bradford assay.PCR-based site-directed mutagenesis was performed following the Q5 Site-Directed Mutagenesis Kit protocol to obtain apo-form spinach Fd and MirFd mutants, which cannot bind the Fe–S cluster. For spinach Fd, plasmid pET-28b(+)::spiFd was a template while primer set SpiFd_C40A_Fwd and SpiFd_C40A_Rev was used for C40A mutation. For MirFd, pET-28b(+)::mirFd was a template while primer set MirFd_C19A_Fwd and MirFd_C19A_Rev was used for C19A. The resultant mutant plasmids were confirmed by Sanger sequencing and transformed into E. coli BL21(DE3) for protein expression.Preparation of reduced Fds by sodium dithioniteThe aerobically purified recombinant Fds were reduced by adding freshly prepared sodium dithionite in 1.5 mL Eppendorf tubes to a final concentration of 1 mM21,22. The tubes were carefully inverted severally until the proteins became almost colourless. Then, UV absorption spectra analyses were performed. The decrease of absorption peaks at 463, 420, and 325 mm for spinach Fd and 418 mm for MirFd indicated the Fe–S cluster reduction.Preparation of GaFd and [15N]-labelled GaFd from Synechocystis
Recombinant proteins of the native Fd (SynFd) and [15N]-labelled SynFd from Synechocystis sp. PCC 6803 were expressed and purified from E. coli BL21(DE3) cells cultured with Luria–Bertani medium and M9 minimal medium (containing 15NH4Cl as the sole nitrogen source), respectively35. The detailed protein purification is described in the Supplementary Methods. Ga-substitution of the SynFd and [15N]-labelled SynFd were performed35. Briefly, 5 mg SynFd and 15 mg [15N]-labelled SynFd were denatured with 6 M HCl to a final concentration of 1 M. Subsequently, we rinsed the pellets with Milli-Q water and resuspended them in 100 mM Tris-HCl (pH 8.0) buffer, respectively. The above steps were repeated thrice to remove iron atoms completely. The final denatured Fd was resuspended in 100 mM Tris-HCl (pH 8.0) buffer containing 10 mM DTT and 6 M guanidine hydrochloride. Next, the apo-forms of SynFd and [15N]-labelled SynFd were refolded at 4 °C by dilution into 20 mM Tris-HCl (pH 8.0), 2 mM DTT, 2 mM GaCl3, and 2 mM Na2S. After overnight incubation, the GaFd and [15N]-labelled GaFd were loaded onto a HiTrap Q HP anion exchange column, eluted with a linear NaCl gradient (0–1 M NaCl), and concentrated by ultrafiltration. The purity of the GaFd and [15N]-labelled GaFd were confirmed using SDS–PAGE analysis, and the concentration was calculated with a molar extinction coefficient (ε280 = 170.2 mM−1 cm−1).NMR analysis of [15N]-labelled GaFd with ligand 4The 1H–15N HSQC spectra were acquired on Avance III 950 US2 spectrometer equipped with a TCI cryogenic probe (Bruker Biospin, Germany) at 950 MHz following a minorly modified method35. The NMR measurement was performed with 10 μM [15N]-labelled GaFd and 10 μM ligand 4 (dissolved in DMSO) in 20 mM sodium phosphate buffer (pH 6.5) containing 50 mM NaCl at 277 and 298 K. The peaks were assigned with the chemical shifts of BMRB entry 16024 by NMRFAM-SPARKY. After measurement, [4 + 2] cycloaddition of compound 4 in the NMR tube was analysed using UPLC as described above.Computational methodsAll the calculations were carried out with the Gaussian 16 (revision B.01) program package59. The molecular structures and harmonic vibrational frequencies were obtained using the hybrid density functional method based on the M06-2X functional60. We used the SDD61 and 6-311 + G** 62 basis set. The self-consistent reaction field (SCRF) method based on the conductor-like polarisable continuum model (CPCM)63,64 was employed to evaluate the solvent reaction field (water; ε = 78.39). Geometry optimisation and vibrational analysis were performed at the same level. All stationary points were optimised without symmetry assumptions and characterised by normal coordinate analysis at the same level of theory (number of imaginary frequencies, 0 for minima and 1 for TSs). The intrinsic reaction coordinate method65,66 was used to track minimum energy paths from transition structures to the corresponding local minima.In vitro conversion of linear polyenoyltetramic acid 10 by VtlFThe linear polyenoyltetramic acid 10 and authentic standards of compounds 11–13 were prepared56,57. The Δphm7 mutant of Pyrenochaetopsis sp. RK10-F058 was cultured in CYA medium at 28 °C with rotary shaking at 150 rpm for 3 days. The mycelia were collected, washed, and frozen at −80 °C. An in vitro enzyme assay of 10 was performed following a minorly modified method56,57. The mycelial powder was resuspended in DMSO, centrifuged at 4 °C; 20,000 × g; 10 min, and the 10-saturated supernatant was used as a substrate for enzyme assay. Dose-dependent assay of VtlF was conducted in a 100 μL reaction mixture containing 50 mM Tris-HCl (pH 7.5), 10 μL 10-saturated supernatant, and 0, 0.2, 0.4, 0.8 mg/mL recombinant VtlF. After incubation on ice for 5 min, the reaction was quenched with 5 μl ice-cold acetic acid and centrifuged.The UPLC analysis was performed on a Waters ACQUITY UPLC BEH C18 Column (2.1 mm i.d. × 100 mm, 1.7 µm) at 0.25 mL/min flow rate56,57. After the injection of 3 μL reaction product into the column equilibrated with 5% CH3CN with 0.05% formic acid, the column was developed with a linear gradient of 5–60% CH3CN with 0.05% formic acid over 1 min and 60–100% CH3CN with 0.05% formic acid over 4 min. Finally, the column was isocratically eluted with 100% CH3CN with 0.05% formic acid for 6 min.Reporting summaryFurther information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Iron-sulphur protein catalysed [4+2] cycloadditions in natural product biosynthesis
