Instruments and materialsNMR spectra were measured with a Bruker 600 spectrometer (Bruker Biospin AG, Fällanden, Germany) using TMS as an internal standard. Commercial silica gel (200–300 and 300–400 mesh, Qingdao Haiyang Chemical Group Co., Ltd., Qingdao, China) was used for column chromatography (CC). Reversed-phase (RP) HPLC was performed on an Agilent 1260 Infinity LC equipped with an Agilent Zorbax SB-C18 column (150 mm × 4.6 mm, 5 μm). Preparative HPLC was carried out on an Agilent 1260 Infinity LC equipped with an Agilent Eclipse XDB-C18 column (250 mm × 21.2 mm, 7 μm). All solvents used for CC and HPLC were of analytical grade (Shanghai Chemical Reagents Co., Ltd.) and chromatographic grade (Dikma Technologies Inc., CA, USA), respectively. X-ray diffraction study was carried out on a Bruker D8 Venture diffractometer. GC-MS analyses used a TRACE 1300 GC system (Thermo Scientific, Milan, Italy), equipped with a TriPlus RSH autosampler (Thermo Scientific, Switzerland), coupled to a Q Exactive GC Orbitrap mass spectrometer (Thermo Scientific, Bremen, Germany). The MS was controlled by Xcalibur@ software (Version 4.1, Thermo Scientific), and the database was NIST MS Search 2.3. The enzymes needed for the cloning were sourced from TransGen Biotech. Isopropyl-beta-D-thiogalactopyranoside (IPTG) and isoprenol were ordered from BBI Life Sciences Corporation and Sigma-Aldrich. The information on strains, plasmids, and primer sequences has been included in Supplementary Tables 1–3, and the DNA and protein sequences of MicA are provided in Supplementary Table 4.Isolation of compounds 1–5
The plasmids carrying CDF-MKI4 and pET28a-MicA were transformed into E. coli BL21 Gold (DE3). The transformed strains were cultured in lysogeny broth (LB) containing kanamycin (50 mg/L) and streptomycin (50 mg/L). After overnight growth, the cultures were inoculated into 90 × 1 L of fresh LB medium. IPTG (0.5 mM) and isoprenol (1.0 mM) were added to the cultures when OD600 reached 1.5 and further incubated at 28 °C and 250 rpm for an additional 18 h. After centrifuging the cultures at 3500 g for 15 min, the resulting pellet was extracted with methanol (MeOH). Next, the organic extract underwent partitioning with petroleum ether (PE). The resulting PE solution was concentrated under reduced pressure, resulting in a yellow oil residue weighing 1.8 g. A portion of the residue (1.0 g) was subjected to silica gel CC (300–400 mesh), yielding compound 1 (466.1 mg). Another portion of the residue (500.0 mg) was initially fractioned by octadecylsilyl. It was further purified by RP HPLC [Acetonitrile (CH3CN)/H2O (95:5)] subsequently to afford compound 2 (3.4 mg, tR = 28.9 min) and several minor peaks. At the same time, the minor peaks were directly analyzed by GC-MS. The isolation and purification for compounds 3–5 follow a similar procedure as the previous steps, with variations in the plasmids used. For compounds 3 and 4, the plasmids containing CDF-MKI4 and pET28a-MicA(V220A) were introduced into E. coli BL21 Gold (DE3) for production. It is worth mentioning that compound 4 was isolated via silica gel CC from compound 3 and was not found in the PE phase during the first extraction process. The plasmids carrying CDF-MKI4 and pET28a-MicA(L221A) were utilized to produce compound 5.Synthesis of microeunicellene derivatives (1a–1d)Compound 1 (10.9 mg, 0.04 mmol, 1.0 eq.) was dissolved in ether (Et2O) (1.0 mL) and treated with NaHCO3 (6.7 mg, 0.08 mmol, 2.0 eq.) and then stirred in an ice bath. In a separate step, mCPBA (6.9 mg, 0.04 mmol, 1.0 eq.) was dissolved in Et2O (1.0 mL) and added to the solution. Following 1 h of reaction, the solution was partitioned with Et2O. The Et2O solution was concentrated under reduced pressure and subsequently purified through silica gel CC [PE, PE/Ethylacetate (EA) (100:1 → 9:1)] to obtain compound 1a (6.4 mg, 59% yield). Through the addition of 2 equivalents of mCPBA, compounds 1b (3.2 mg, 30% yield) and 1c (6.9 mg, 64% yield) were successfully synthesized. However, when three equivalents of mCPBA were used, compound 1d (8.2 mg, 75% yield) was formed as the predominant product. Detailed structural elucidation and assignment are shown in Supplementary Information.GC-MS analysisThe GC-MS analyses were conducted using a TRACE 1300 GC system (Thermo Scientific, Milan, Italy), which was equipped with a TriPlus RSH autosampler (Thermo Scientific, Switzerland). This system was coupled to a Q Exactive GC Orbitrap mass spectrometer (Thermo Scientific, Bremen, Germany). The compounds and fractions prepared by reversed-phase HPLC (4 samples, Supplementary Figs. 5–9) were dissolved in acetonitrile (0.10 mg/mL) for the GC-MS analysis. These solutions were then filtered using a 0.22 μm nylon 66 syringe filter (Sigma) to obtain the test samples. The GC settings were as follows: a TraceGOLD TG-5SilMS 30 m × 0.25 mm I.D. × 0.25 µm film capillary column (Thermo Scientific, USA) was used. We injected 1 µL of sample material (with equal volumes of acetonitrile as the blank control) into the GC at a 20:1 split ratio. The inlet temperature was set at 250 °C, and a constant helium flow of 1.2 mL/min was used as the carrier gas. The gradient elution was controlled by an oven temperature program, which started at 50 °C (held for 3 min), then increased by 10 °C/min up to 280 °C (held for 5 min). This resulted in a total run-time of 31.0 min. The MS was operated in the electron ionization positive mode (70 ev) at an ion source temperature of 250 °C. The acquisition was carried out in full-scan mode within a mass range of 35–500 m/z at a resolving power of 60,000. Each sample was collected once for data acquisition. The MS data was analyzed using Xcalibur@ software (Version 4.1, Thermo Scientific), and the database used was NIST MS Search 2.3.X-ray diffraction analysis of 1,
1a–1d
The crystallographic data were collected on a Bruker D8 Venture diffractometer equipped with Cu Kα radiation (λ = 1.54178 Å). The crystals were kept at 171.0 K during data collection. The structures were solved with the ShelXT62, structure solution program using Intrinsic Phasing and refined with the ShelXL63, refinement package using Least Squares minimization.Microeunicellene (1) was crystallized directly by leaving a large amount (~400 mg) in the fridge (−20 °C). Crystal data (M = 272.45 g/mol): [Flack parameter: 0.60(4)], orthorhombic, space group P212121 (no. 19), a = 9.4505(3) Å, b = 10.2972(3) Å, c = 35.6882(11) Å, V = 3472.95(18) Å3, Z = 8, T = 100 K, μ (Cu Kα) = 0.419 mm−1, Dcalc = 1.042 g/cm3, 115639 reflections measured (4.952° ≤ 2θ ≤ 160.484°), 7517 unique (Rint = 0.0877, Rsigma = 0.0326) which were used in all calculations. The final R1 was 0.0508 (l > 2σ(l)) and wR2 was 0.1250 (all data). The crystallographic data were deposited at the Cambridge Crystallographic Data Centre with CCDC number 2310340.(6S,7S)-Epoxy-microeunicellene (1a) was crystallized from MeOH at 4 °C, m.p. 44–46 °C. Crystal data (M = 288.45 g/mol): [Flack parameter: 0.00(7)], monoclinic, space group P21 (no. 4), a = 7.9126(2) Å, b = 6.8092(2) Å, c = 16.4871(5) Å, β = 93.189(2)°, V = 886.92(4) Å3, Z = 2, T = 100 K, μ (Cu Kα) = 0.479 mm−1, Dcalc = 1.080 g/cm3, 1951 reflections measured (5.368° ≤ 2θ ≤ 148.862°), 1951 unique (Rsigma = 0.0538) which were used in all calculations. The final R1 was 0.0634 (l > 2σ(l)) and wR2 was 0.1690 (all data). The crystallographic data were deposited at the Cambridge Crystallographic Data Centre with CCDC number 2310341.(6S,7S,11R)-Bis-epoxy-microeunicellene (1b) was crystallized from PE/isopropyl alcohol (IPA) (8:2) at 4 °C, m.p. 127–129 °C. Crystal data (M = 304.45 g/mol): [Flack parameter: 0.09(11)], orthorhombic, space group P212121 (no. 19), a = 8.8696(4) Å, b = 9.6785(4) Å, c = 21.0117(9) Å, V = 1803.74(13) Å3, Z = 4, T = 170 K, μ (Cu Kα) = 0.538 mm−1, Dcalc = 1.121 g/cm3, 36301 reflections measured (8.416° ≤ 2θ ≤ 133.18°), 3194 unique (Rint = 0.0807, Rsigma = 0.0351) which were used in all calculations. The final R1 was 0.0355 (l > 2σ(l)) and wR2 was 0.0909 (all data). The crystallographic data were deposited at the Cambridge Crystallographic Data Centre with CCDC number 2310342.(6S,7S,11S)-Bis-epoxy-microeunicellene (1c) was crystallized from ethyl acetate (EA) at room temperature (rt), m.p. 100–102 °C. Crystal Data (M = 304.45 g/mol): [Flack parameter: −0.10(15)], monoclinic, space group P21 (no. 4), a = 7.9978(4) Å, b = 6.7005(2) Å, c = 16.6441(6) Å, β = 93.924(3)°, V = 889.85(6) Å3, Z = 2, T = 100 K, μ (Cu Kα) = 0.545 mm−1, Dcalc = 1.136 g/cm3, 16279 reflections measured (5.322° ≤ 2θ ≤ 149.414°), 3478 unique (Rint = 0.0562, Rsigma = 0.0424) which were used in all calculations. The final R1 was 0.0413 (l > 2σ(l)) and wR2 was 0.1079 (all data). The crystallographic data were deposited at the Cambridge Crystallographic Data Centre with CCDC number 2310347.(2S,3R,6S,7S,11S)-Tri-epoxy-microeunicellene (1d) was crystallized from PE/dichloromethane (DCM) (1:1) at rt, m.p. 140–142 °C. Crystal Data for (M = 320.45 g/mol): [Flack parameter: −0.02(10)], orthorhombic, space group P212121 (no. 19), a = 33.6260(8) Å, b = 6.7974(2) Å, c = 7.9786(2) Å, V = 1823.66(8) Å3, Z = 4, T = 100 K, μ (Cu Kα) = 0.599 mm−1, Dcalc = 1.167 g/cm3, 24,595 reflections measured (5.256° ≤ 2θ ≤ 149.596°), 3736 unique (Rint = 0.0835, Rsigma = 0.0485) which were used in all calculations. The final R1 was 0.0725 (l > 2σ(l)) and wR2 was 0.2221 (all data). The crystallographic data were deposited at the Cambridge Crystallographic Data Centre with CCDC number 2310348.Synthesis of geranyllinaloyl diphosphate (GLPP)1-Chloro-geranyllinalool. Conditions for the following reaction were based on those described by Cane64. A solution of mixtures of (S)- and (R)-geranyllinalool (CAS No.: 1113-21-9, 435.02 mg, 1.497 mmol) and triphenylphosphine (CAS No.: 603-35-0, 785.30 mg, 2.994 mol, 2.00 eq.) in 5 mL of carbon tetrachloride (CAS No.: 56-23-5) was heated at reflux at 84 °C for 12 h. The solution was cooled and diluted with pentane (50 mL). The triphenylphosphine oxide precipitated from the solution was removed by vacuum filtration through a fritted-glass funnel. Most of the solvent was removed with a rotary evaporator at aspiratory pressure. The last traces of solvent were removed by pumping at a high vacuum for one and a half hours to yield 254.25 mg 1-chloro-geranyllinalool (55%). The resulting pale-yellow oil was used directly in the next step.Geranyllinaloyl diphosphate (GLPP). Woodside’s procedure was employed to prepare GLPP65. Tris (tetrabutylammonium) hydrogen pyrophosphate trihydrate ((Bu4N)3P2O7H) was prepared as follows. Disodium dihydrogen pyrophosphate (CAS No.: 7758-16-9, 3.10 g, 15.349 mmol) was dissolved in 25 mL of 4% (v/v) aqueous ammonium hydroxide. The clear solution was loaded onto a 2 × 30 cm column of Dowex AG 50W-X8 cation exchange resin (100–200 mesh, H+ form), which had been prewashed with deionized water. The free acid was eluted with 150 mL of deionized water, and the eluant was immediately titrated to pH 7.3 with 25% (w/w) aqueous tetrabutylammonium hydroxide (CAS No.: 2052-49-5, ~15 mL). The resulting solution (~190 mL total volume) was dried by freezing the solution in a dry ice isopropanol bath and lyophilizing for ~2 days to yield a hygroscopic white solid (9.80 g, 77%), which was stored over phosphorus pentoxide until required.The mixture of 1-chloro-geranyllinalool (0.823 mmol), (Bu4N)3P2O7H (1.92 g, 2.008 mmol, 2.44 eq.) and acetonitrile (5 mL) was stirred at room temperature for 2 h. The solvent was then removed with a rotary evaporator using a 40 °C water bath to give a yellow oil, which was dissolved in ion-exchange buffer (1:49 iPrOH:25 mM NH4HCO3), and the aqueous phase was washed with Et2O (2 × 4 mL), collected, and loaded onto a 4 × 15 cm column of Dowex AG 50W-X8 cation exchange resin (100–200 mesh, NH4+ form). The column was eluted with 360 mL (two column volumes) of ion-exchange buffer, and the fractions containing GLPP (fractions that stained purple on silica TLC plates with vanillin-sulfuric acid) were lyophilized to give a white solid (to prevent moisture absorption and oxidation, store in −80 °C with nitrogen padding) containing inorganic phosphate and probably some NH4HCO3 (164.99 mg, ~40%). 1H NMR (D2O, 600 MHz) of GLPP: δ 1.55 (s, 3H, CH3), 1.58 (s, 3H, CH3), 1.63 (s, 3H, CH3), 1.69 (s, 3H, CH3), 1.90–2.14 (m, 12H, CH2), 4.44 (m, 2H, CH2), 5.10 (m, 3H, vinyl H), 5.43 (m, 1H).Synthesis of 2Z-geranylgeranyl diphosphate (2Z-GGPP) and GGPPEthyl geranylgeranate. This compound was synthesized according to a published procedure by Rabe66. Diisopropylamine (CAS No.: 108-18-9, 3.52 mL, 25.163 mmol, 1.05 eq.) was dissolved in dry THF (120 mL) and cooled to 0 °C. After adding nBuLi (CAS No.: 109-72-8, 2.7 M in hexane, 9.33 mL, 25.191 mmol, 1.05 eq.), the reaction mixture was stirred for 60 min at 0 °C. The mixture was cooled to −78 °C and triethyl phosphonoacetate (CAS No.: 867-13-0, 4.76 mL, 24.012 mmol, 1.00 eq.) was added and stirred for 2 h at −78 °C, followed by the addition of the (5E,9E)-6,10,14-trimethylpentadeca-5,9,13-trien-2-one (CAS No.: 1117-52-8, 7.14 mL, 23.982 mmol, 1.00 eq.). The reaction mixture was stirred for 2 h at −78 °C, then warmed to room temperature and stirred overnight. The reaction was quenched with 120 mL H2O, and the mixture was extracted three times with Et2O. The combined organic layers were dried with MgSO4 and concentrated under reduced pressure. The residue was purified by CC on silica gel with PE/EA (200:1) to yield 7.82 g (98%) ethyl geranylgeranate as mixtures of 2Z and 2E diastereomers (1:5).2Z- and 2E- geranylgeraniol. The ethyl geranylgeranate (4.60 g, 13.833 mmol, 1.00 eq.) was dissolved in dry Et2O and cooled to 0 °C. After adding DIBAlH (CAS No.: 1191-15-7, 1 M in hexane, 30.52 mL, 30.520 mmol, 2.21 eq.), the reaction mixture was stirred for 2 h. The reaction was quenched by adding 1.2 mL water dropwise, 1.2 mL 15% NaOH, and 3 mL water. The mixture was extracted three times with Et2O. The combined organic layers were dried with MgSO4 and concentrated under reduced pressure. The residue was purified by CC on silica gel with PE/EA (20:1) to give the 2Z- and 2E- geranylgeraniol (691.08 mg and 2.66 g, 17% and 66%). 1H NMR (CDCl3, 600 MHz) of 2Z-geranylgeraniol: δ 1.58 (s, 3H, CH3), 1.59 (s, 3H, CH3), 1.59 (s, 3H, CH3), 1.67 (s, 3H, CH3), 1.74 (s, 3H, CH3), 1.94–2.10 (m, 12H, CH2), 4.08 (d, 1H, J = 7.2 Hz, CH2), 5.09 (m, 3H, vinyl H), 5.42 (brt, 1H, J = 7.1 Hz, vinyl H). 1H NMR (CDCl3, 600 MHz) of 2E-geranylgeraniol: δ 1.59 (s, 3H, CH3), 1.60 (s, 3H, CH3), 1.60 (s, 3H, CH3), 1.68 (s, 3H, CH3), 1.68 (s, 3H, CH3), 1.95–2.14 (m, 12H, CH2), 4.15 (d, 1H, J = 6.9 Hz, CH2), 5.11 (m, 3H, vinyl H), 5.42 (tq, 1H, J = 7.0, 1.4 Hz, vinyl H).2Z-GGPP and GGPP. Diphosphorylations of 2Z- and 2E- geranylgeraniol (401.12 mg and 405.25 mg) were carried out as described above for GLPP to yield 2Z-GGPP and GGPP (159.27 mg and 139.92 mg, 23% and 20%). 1H NMR (D2O, 600 MHz) of 2Z-GGPP: δ 1.56 (s, 3H, CH3), 1.56 (s, 3H, CH3), 1.59 (s, 3H, CH3), 1.63 (s, 3H, CH3), 1.75 (s, 3H, CH3), 1.91–2.15 (m, 12H, CH2), 4.44 (t, 2H, J = 7.2 Hz, CH2), 5.07 (t, 1H, J = 8.6 Hz, vinyl H), 5.10 (t, 1H, J = 8.1 Hz, vinyl H), 5.15 (t, 1H, J = 7.0 Hz, vinyl H), 5.46 (t, 1H, J = 7.0 Hz, vinyl H). 1H NMR (D2O, 600 MHz) of GGPP: δ 1.62 (s, 3H, CH3), 1.63 (s, 3H, CH3), 1.64 (s, 3H, CH3), 1.70 (s, 3H, CH3), 1.74 (s, 3H, CH3), 2.01–2.20 (m, 12H, CH2), 4.48 (t, 2H, J = 6.6 Hz, CH2), 5.20 (t, 2H, J = 7.2 Hz, vinyl H), 5.24 (t, 1H, J = 7.0 Hz, vinyl H), 5.48 (t, 1H, J = 7.2 Hz, vinyl H).Synthesis of 1,1-2H2-GGPP1,1-2H2-2E-geranylgeraniol. The ethyl geranylgeranate (4.56 g, 13.713 mmol, 1.00 eq.) was dissolved in dry THF and cooled to 0 °C. After adding LiAl2H4 (CAS No.: 14128-54-2, 1.15 g, 27.394 mmol, 2.00 eq.), the reaction mixture was stirred for 4 h. The reaction was quenched by adding 1.0 mL water dropwise, 1.0 mL 15% NaOH, and 3.0 mL water. The mixture was extracted three times with Et2O. The combined organic layers were dried with MgSO4 and concentrated under reduced pressure. The residue was purified by CC on silica gel with PE/EA (20:1) to give the 1,1-2H2-2E-geranylgeraniol (2.76 g, 69%). 1H NMR (CDCl3, 600 MHz) of 1,1-2H2-2E-geranylgeraniol: δ 1.59 (s, 3H, CH3), 1.60 (s, 3H, CH3), 1.60 (s, 3H, CH3), 1.68 (s, 3H, CH3), 1.68 (s, 3H, CH3), 1.95–2.14 (m, 12H, CH2), 5.10 (m, 3H, vinyl H), 5.41 (s, 1H, vinyl H).1,1-2H2-GGPP. Diphosphorylations of 1,1-2H2-2E-geranylgeraniol (415.22 mg) were carried out as described above for GLPP to give 1,1-2H2-GGPP (135.80 mg, 19%). 1H NMR (D2O, 600 MHz) of 1,1-2H2-GGPP: δ 1.56 (s, 3H, CH3), 1.56 (s, 3H, CH3), 1.59 (s, 3H, CH3), 1.64 (s, 3H, CH3), 1.70 (s, 3H, CH3), 1.91–2.13 (m, 12H, CH2), 5.10 (m, 3H, vinyl H), 5.41 (s, 1H, vinyl H).In vitro enzymatic assays of MicAThe assays were performed in 50 mM Tris-HCl, pH 8.0, containing 5 mM GGPP, 15 mM MgCl2, and 20 µM TS enzyme in a total volume of 100 µL. The reactions were initiated by the addition of enzyme and incubated for 3 h at 37 °C. The reactions were quenched with 200 µL of ice-cold acetonitrile and then 50 µL of saturated NaCl solution was added to form two layers. The upper organic layer was taken for HPLC analysis. Chromatographic separation was carried out at 35 °C, with a flow rate of 1 mL/min, and an 18 min solvent gradient from 5 to 95% CH3CN in water. The linear gradient program was run as follows: 0–3 min, 5% CH3CN; 3–18 min, 5–95% CH3CN; 18–35 min, 95% CH3CN. Enzyme products were detected by monitoring at 210 nm with a photodiode array detector. Enzyme reactions with 2Z-GGPP, GLPP or 1,1-2H2-GGPP were similarly carried out.Gene cloning and site-directed mutationUsing synthesized pET28a-MicE plasmid, we subcloned the genetic fragment of MicE into the CDF-MKI vector, which was linearized with XhoI, and transformed into E. coli Turbo (Shanghai Weidi Biotechnology) to produce plasmid CDF-MKI-MicE. The plasmid was confirmed by DNA sequencing. Overlapped PCR was used for site-directed mutagenesis of MicA, with the synthesized gene MicA as the template using TransStar FastPfu Fly DNA polymerase. After gel extraction, the PCR products were purified and inserted into a linearized pET28a (+) vector using BamHI and HindIII restriction enzymes. The resulting constructs were then transformed into E. coli Turbo to generate the desired plasmids and further confirmed by DNA sequencing. The plasmids with the correct sequencing were transferred into E. coli BL21 Gold (DE3) for expression, followed by HPLC analysis. Primer sequences used in this study were shown in Table S3.Protein expression and purificationPlasmids carrying MicA and its mutants were separately introduced into E. coli BL21 Gold (DE3). The E. coli strains carrying the respective plasmids were cultured in LB containing kanamycin (50 mg/mL) for antibiotic selection. Each strain was cultured in 4 × 1 L of LB at 37 °C, shaking at 200 rpm until reaching an OD600 of 0.6. Induction of gene expression was achieved by adding IPTG (0.3 mM), followed by further cultivation of the cells at 16 °C for 16 h at 200 rpm. The cells were collected by centrifugation at 3500 g for 10 min at 4 °C. The resulting pellet was then re-suspended in cold lysis buffer (50 mM Tris-HCl, pH 8.0, containing 150 mM NaCl). Cells were disrupted using a High-Pressure Homogenizer (Scientz), followed by centrifugation at 18,000 g for 40 min at 4 °C. The desired proteins were purified through nickel affinity chromatography by applying the supernatant onto a column filled with Ni SepharoseTM 6 Fast Flow (Cytiva), washing with wash buffer (lysis buffer containing 20 mM imidazole), and eluting with elution buffer (lysis buffer containing 500 mM imidazole). The purified protein was promptly desalted using a PD-10 column (Cytiva). The protein purities were determined by analyzing them using SDS-PAGE. Each protein was divided into individual aliquots, which were then quickly frozen in liquid nitrogen and stored at −80 °C until used.Protein structure prediction and docking analysisThe tFold55, RoseTTAFold56, and AlphaFold57 (ColabFold v1.5.3: AlphaFold2 using MMseqs2) were employed to generate de novo predicted models of MicA. To proceed, carefully follow the instructions provided on the respective individual webpage. Afterward, the generated models were uploaded onto the tFold platform to evaluate their quality. The highest-scoring model, as determined by the “Protein Model Quality Assessment” system, was then uploaded to Alphafill (https://alphafill.eu/) to obtain the MicA structure with three Mg2+ ions, which was then docked with the ligand (Supplementary Fig. 34). The coordinates for the tFold de novo model of MicA were deposited into figshare (https://doi.org/10.6084/m9.figshare.26124145.v1). The ligand structure was drawn and subsequently converted into 3D conformation. The receptor and prepared ligand were then submitted to SwissDock and AutoDoc4 for molecular docking analysis. The default settings were utilized for the entire procedure. The relevant websites are provided as follows:the tFold: https://drug.ai.tencent.com/console/cn/tfoldRoseTTAFold: https://colab.research.google.com/github/sokrypton/ColabFold/blob/main/RoseTTAFold.ipynbAlphaFold: https://colab.research.google.com/github/sokrypton/ColabFold/blob/main/AlphaFold2.ipynbSwissDock: http://www.swissdock.ch/dockingDFT calculationsAll the DFT calculations were carried out with the Gaussian 16 package. All geometries were optimized using the B3LYP/6-31+G(d,p) method67,68,69,70. All stationary points were characterized by frequency calculations and reported energies include zero-point energy corrections (unscaled). IRC calculations were performed using the Local Quadratic Approximation (LQA) algorithm71 at this same level of theory, with a maximum of 100 points (maxpoint = 100)72,73. The single-point energies were calculated using the mPW1PW91 method with the 6-31+G(d,p) basis set.Conformational search and optimizationThe conformational generation and optimization of all molecules is done by Schrödinger (version 2020) and Gauss16 package. The molecules are first prepared using LigPrep with default parameter, and then sampled through Prime Macrocycle Sampling, where the sampling intensity is set to through. All generated conformations were further optimized using the B3LYP/6-31+G(d,p)67,68,69,70 method for all atoms in vacuum. Frequency analysis was conducted at the same level of theory to verify the stationary points to be minimum or saddle points. The single-point energies were calculated using mPW1PW91/6-31+G(d,p). We calculated the free energies of different conformations of a molecule and count the Boltzmann conformational distribution of the molecule.Reporting summaryFurther information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Discovery of a terpene synthase synthesizing a nearly non-flexible eunicellane reveals the basis of flexibility
