Bacterial strains, plasmids, and chemicals.
Escherichia coli DH5α (Takara, Shiga, Japan) and pBluescript II KS(+) or SK(+) plasmids (Stratagene, La Jolla, CA) were used for sub-cloning experiments and sequencing analyses. The E. coli JW0003-KC strain was purchased from NBRP-E. coli at NIG, Japan and used for complementation studies44,45. Streptomyces albulus CR146, a cryptic plasmid pNO33-curing strain derived from S. albulus NBRC14147, was used as a parent strain for construction of the Δhsd, Δhk, and ΔmetM mutants. Luria Bertani (LB) medium (1.0% Bacto tryptone, 0.5% Bacto yeast extract, and 0.5% NaCl) and M9 medium (0.02% glucose, 6 g/L Na2HPO4, 3 g/L KH2PO4, 0.5 g/L NaCl, 1 g/L NH4Cl, 1 mM MgSO4, 0.1 mM CaCl2, and 10 mg/L thiamine) were generally used to cultivate E. coli strains. SLB medium (10.3% sucrose, 1.0% Bacto tryptone, 0.5% Bacto yeast extract, and 0.5% NaCl) was used for the S. albulus strain pre-culture. The American Type Culture Collection no. 5 (ATCC5) plate, consisting of 0.2% starch, 0.1% Bacto yeast extract, 0.1% Bacto beef extract, and 0.01% FeSO4·7H2O (pH 7.2), was used for spore formation. Antibiotics (200 µg/mL ampicillin, 50 µg/mL kanamycin, 50 µg/mL thiostrepton, 25 µg/mL nalidixic acid, or 25 µg/mL neomycin) were added to media when required. The expression plasmids pET21a, pET28a, pETDuet-1, and pRSFDuet-1 were purchased from Novagen (Darmstadt, Germany). The IMPACT kit plasmid, pTXB1 vector, was purchased from New England Biolabs Japan (Tokyo). The pGM160Δaac(3)I::oriT47 and pLAE00346 plasmids were used to conduct gene disruption. The pLAE006 plasmid48 carrying the ermE* promoter and neomycin resistance gene was used to complement metM in the ΔmetM mutant. All chemicals were purchased from Sigma-Aldrich (St. Louis, MO), Wako Pure Chemical (Osaka, Japan), Kanto Chemicals (Tokyo), and Nacalai Tesque (Kyoto, Japan). DNA restriction enzymes were purchased from Takara Bio Inc. (Shiga, Japan). Primers were purchased from Eurofins Genomics K.K. (Tokyo). All PCR primers used in this study are listed in Supplementary Table 1.
Bioinformatic analysis.
Hsd, HK, SaTS, MetM, MetO, MoeZ, and Mec in S. albulus NBRC14147 were searched for in BLAST, and their Gene IDs (locus_tag) are listed in Supplementary Table 2. The Rfam website (https://rfam.org/)23 was used to analyze the upstream region of metM. MetM homologs in the 550 Streptomyces strains we checked are listed in Dataset S1. The phylogenetic tree of TS and MetM homologs was generated by MEGA11 using the amino acids sequence data listed in Dataset S1. Multiple alignments were generated with the MultAlin web tool (http://multalin.toulouse.inra.fr/multalin/)49 and depicted using ESPript50. The complex model structure of MetM and MetOΔC was generated using ColabFold37.
Gene disruption.
S. albulus mutants lacking hsd, hk, or metM were constructed as follows. The first polymerase chain reactions (PCRs) were performed with the primer pairs upDhsd-Fw/upDhsd-Rv, upDhk-Fw/upDhk-Rv, or upDmetM-Fw/upDmetM-Rv to amplify the upstream region of the hsd, hk, or metM gene using the S. albulus NBRC14147 genome as a template. The second PCR was performed with the primer pairs downDhsd-Fw/ downDhsd-Rv, downDhk-Fw/downDhk-Rv, or downDmetM-Fw/downDmetM-Rv to amplify the downstream region of the hsd, hk, or metM gene. The first and second amplified fragments were digested with Hind III/Spe I and ligated separately into pBluescript II KS(+) digested with Hind III/Spe I for sequence verification. The up- and down-stream region fragments of hsd or metM with the correct sequences were digested with Hind III/Spe I and ligated together into the pGM160Δaac(3)I::oriT digested with Hind III and dephosphorylated by alkaline phosphatase. The resulting plasmids, pΔhsd and pΔmetM, in which the up- and down-stream region fragments were ligated, were used to knock out hsd or metM in S. albulus CR1 (Supplementary Table 3). The up- and down-stream region fragments of hk with the correct sequences were digested with Hind III/Spe I and ligated together into the pBluescript II KS(+) digested with Hind III and dephosphorylated by alkaline phosphatase to yield pBKS-Δhk, in which up- and down-stream region fragments were ligated. To obtain a thiostrepton resistance gene (tsr) fragment, PCR was performed with the primer pairs tsr-Fw/tsr-Rv using pGM160Δaac(3)I::oriT as a template. The amplified tsr fragment was digested with Spe I and ligated into pBluescript II KS(+) digested with Spe I for sequence verification. The tsr fragment with the correct sequence was digested with Spe I and ligated together into pBKS-Δhk digested with Spe I and dephosphorylated by alkaline phosphatase. The resulting plasmid containing an Δhk cassette, in which a tsr fragment was inserted between the up- and down-stream region fragments of hk, was digested with Hind III to yield the Δhk cassette. This cassette was ligated together into pLAE003 digested with Hind III and dephosphorylated by alkaline phosphatase. The resulting plasmid, pLAEΔhk, was used to obtain the hk knock-in mutant (Supplementary Table 3).
Conjugal transfer of the plasmids was performed as described previously using E. coli S17-146. The E. coli donor transformants harboring pΔhsd, pLAEΔhk, or pΔmetM were pre-cultured in LB medium supplemented with 200 µg/mL ampicillin (pΔhsd or pΔmetM) or 25 µg/mL neomycin (pLAEΔhk) at 28°C overnight, transferred into 3 mL of the same fresh medium at 1%, and incubated at 28°C for 3 h. Harvested cells were washed twice and resuspended in 60 µL of LB medium. E. coli cells were mixed with S. albulus CR1 spores in 25% glycerol (1×109/mL, 60 µL), and 60 µL of the mixture was spread on ATCC5 plates supplemented with 40 mM MgCl2. After incubation at 28°C for 21 h, the plates were overlaid with 25 µg/mL nalidixic acid and 50 µg/mL thiostrepton and incubated for an additional 5 d. The resultant transformants were pre-cultured in SLB medium supplemented with 50 µg/mL thiostrepton at 28°C overnight, transferred at 1% into modified SLB medium consisting of 20% sucrose, 1.0% Bacto tryptone, 0.5% Bacto yeast extract, 0.5% NaCl, and 2.0% glycine, and incubated at 28°C for 44 h. Protoplasts were prepared and regenerated on RSA plates as described previously46. To obtain Δhsd or ΔmetM mutant strains, thiostrepton-sensitive colonies were selected from the regenerated colonies and confirmed to be a knockout-mutant by PCR using a primer pair, Hsd-Fw/Hsd-Rv or MetM-Fw/MetM-Rv. To obtain the Δhk mutant strain, thiostrepton-resistant and neomycin-sensitive colonies from the regenerated colonies were selected and confirmed to be a knock-in mutant by PCR using a primer pair, HK-Fw/HK-Rv.
Construction of complemented mutants.
Using S. albulus NBRC 14147 genome DNA as a template, Sats and metM genes were amplified by PCR using two primer pairs: SaTS-Fw/SaTS-Rv and MetM-Fw/ MetM-Rv, respectively. Fragments were digested with BamH I/Hind III and ligated separately into pBluescript II SK(+) digested with BamH I/Hind III to yield pBSK-Sats and pBSK-metM plasmids (Supplementary Table 3). After sequence verification, pBluescript II SK(+) and the constructed plasmids were separately introduced into the E. coli JW0003-KC strain. The resultant strains were named “ΔthrC + empty”, “ΔthrC + Sats”, and “ΔthrC + metM”.
The metM fragment digested with BamH I/Hind III from pBSK-metM was ligated into pLAE006 digested with BamH I/Hind III to yield pLAE006-metM (Supplementary Table 3). An E. coli S17-1 harboring pLAE006 or pLAE006-metM was constructed and used as a donor strain for conjugal transfer to the ΔmetM. The conjugal transfer was examined in the same way as described above but using 25 µg/mL neomycin instead of thiostrepton. The resultant neomycin-resistant strain harboring pLAE006 or pLAE006-metM was named “ΔmetM + empty” or “ΔmetM + metM”.
Growth tests.
Growth tests for the Δhsd, Δhk, ΔmetM, ΔmetM + empty, and ΔmetM + metM strains were examined as follows. Spore solution of the mutants (1.0×106, 107, 108, 109/mL, 5µL) were spotted onto M9 plates (for Δhsd, Δhk, and ΔmetM) or M9 plates containing 25 µg/mL neomycin (for ΔmetM + empty and ΔmetM + metM), and incubated at 28°C for 72 h.
Growth tests for the ΔthrC + empty, ΔthrC + Sats, and ΔthrC + metM strains were conducted as follows. The strains were cultured in LB medium supplemented with 200 µg/mL ampicillin and 25 µg/mL kanamycin at 37°C overnight. Harvested cells were then washed twice and resuspended in M9 medium. After adjusting the cell density (1╳100, 10− 1, 10− 2, 10− 3, or 10− 4) at 600 nm (OD600), washed cell suspensions (5 µL) were spotted onto M9 plates containing 200 µg/mL ampicillin and 25 µg/mL kanamycin and cultured at 37℃ for 4 d.
Plasmid constructions for recombinant protein expression.
To examine the pulldown assay between MetM with a His6-tag at the N-terminus and MetOΔC with a Strep-tag at the N-terminus, we constructed two plasmids, pRSFD-metM and pET21a-ST-metOΔC, as follows (Supplementary Table 3). Using S. albulus NBRC 14147 genome DNA as a template, the metM and metOΔC genes were amplified by PCR using two primer pairs: MetM-Fw2/MetM-Rv and ST-MetO-Fw/SCP-Rv, respectively. The fragments were digested with EcoR I/Hind III (for metM) or Hind III/BamH I (for metOΔC) and cloned separately into a pBluescript II KS(+) digested with EcoR I/Hind III or Hind III/BamH I to yield pBKS-metM and pBKS-ST-metOΔC. After sequence verification, the metM fragment digested with EcoR I/Hind III from pBKS-metM was ligated into pRSFDuet-1 vector digested with EcoR I/Hind III, yielding pRSFD-metM; the metOΔC fragment digested with Nde I/BamH I from pBKS-ST-metOΔC was ligated into pET21a vector digested with Nde I/BamH I, yielding pET21a-ST-metOΔC.
In the enzymatic assay, we used pET28a-mec, pET28a-metO, pET28a-metOΔC, pTXB1-metOΔC, pTXB1-metO-G90, and pETDuet-metM constructed as follows (Supplementary Table 3). Using S. albulus NBRC 14147 genome DNA as a template, the mec, metO, or metOΔC gene was amplified by PCR using one of four primer pairs, Mec-Fw/Mec-Rv, MetO-Fw/MetO-Rv, MetO-Fw/SCP-Rv, or MetO-Fw/SCP-IMPACT-Rv, respectively. The four fragments were digested with Hind III/BamH I and cloned into a pBluescript II KS(+) digested with Hind III/BamH I, respectively, to yield pBKS-mec, pBKS-metO, pBKS-metOΔC, and pBKS-metOΔC-IMPACT. After sequence verification, the mec, metO, or metOΔC fragment digested with Nde I/BamH I from pBKS-mec, pBKS-metO, or pBKS-metOΔC was ligated into pET28a vector digested with Nde I/BamH I to yield pET28a-mec, pET28a-metO, and pET28a-metOΔC. The metOΔC fragment digested with Nde I/Spe I from pBKS-metOΔC-IMPACT was ligated into pTXB1 vector digested with Nde I/Spe I to yield pTXB1-metOΔC. The metM fragment digested with EcoR I/Hind III from pBKS-metM was ligated into pETDuet-1 vector digested with EcoR I/Hind III to yield pETDuet-metM. To prepare the pTXB1-metO-G90 plasmid, an appropriate mutation was introduced into the metO gene on the pTXB1-metOΔC plasmid by inverse PCR using the primer pair G90-IMPACT-Fw/G90-IMPACT-Rv.
Pulldown assay.
For pulldown assay examination between MetM and MetOΔC, three E. coli BL21 (DE3) transformants harboring (i) pRSFD-metM and pET21a, (ii) pRSFDuet-1 and pET21a-ST-metOΔC, or (iii) pRSFD-metM and pET21a-ST-metOΔC were constructed and pre-cultured separately in 3 mL of LB medium supplemented with 50 µg/mL kanamycin and 200 µg/mL ampicillin at 37°C overnight. Each pre-seed culture aliquot (2.0 mL) was inoculated into a 500 mL baffled flask containing 200 mL LB medium, 50 µg/mL kanamycin, and 200 µg/mL ampicillin and grown at 37°C, respectively. After approximately 3 h of growth, isopropyl-β-d-thiogalactopyranoside (IPTG) was added at a final concentration of 0.1 mM, and the temperature was set to 28°C. After overnight culture, each cell was harvested by centrifugation at 3000 × g at 4°C for 10 min, then washed with buffer I (20 mM Tris–HCl, pH 8.0, and 150 mM NaCl), and disrupted by sonication. Cell debris was removed by centrifugation at 20000 × g at 4°C for 10 min. The supernatant was applied to a Strep-Tactin column (QIAGEN K.K., Tokyo) pre-equilibrated with buffer II (100 mM Tris-HCl, pH 8.0). After washing with 3 column volumes (CV) of buffer II three times, the adsorbed proteins were eluted with 3 CV of buffer II supplemented with 2.5 mM d-desthiobiotin and 150 mM NaCl. The recombinant protein expression was verified by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).
Recombinant protein expression.
For recombinant N-His tagged Mec, N-His tagged MetO, N-His tagged MetOΔC, MetOΔC fused with intein and the Chitin binding domain (CBD) at the C-terminus (MetOΔC-int-CBD), MetO-G90 fused with intein and CBD at the C-terminus (MetO-G90-int-CBD), and N-His tagged MetM expression, we constructed six E. coli BL21 (DE3) transformants harboring pET28a-mec, pET28a-metO, pET28a-metOΔC, pTXB1-metOΔC, pTXB1-metO-G90, or pETDuet-metM, respectively. These transformants were pre-cultured respectively in 3 mL of LB medium supplemented with 50 µg/mL kanamycin or 200 µg/mL ampicillin at 37°C overnight. Each pre-seed culture aliquot (0.5 mL) was separately inoculated into a 500 mL baffled flask containing 50 mL LB medium and 50 µg/mL kanamycin or 200 µg/mL ampicillin and grown at 37°C. After approximately 3 h of growth, IPTG was added at a final concentration of 0.1 mM, and the temperature was set to 18°C (for Mec, MetO, MetOΔC, and MetM) or 28°C (for MetOΔC-int-CBD and MetO-G90-int-CBD). After overnight culture, each cell was harvested by centrifugation at 3000 × g at 4°C for 10 min, then washed with buffer III [20 mM HEPES–NaOH and pH 8.0, 150 mM NaCl] for Mec, MetM, MetO, and MetOΔC or buffer IV [20 mM HEPES–NaOH, pH 8.0, 500 mM NaCl, and 0.1 mM ethylenediaminetetraacetic acid (EDTA)] for MetOΔC-int-CBD and MetO-G90-int-CBD, and disrupted by sonication. Cell debris was removed by centrifugation at 20000 × g at 4°C for 10 min. Each supernatant was applied to an Ni-NTA agarose column (QIAGEN K.K., Tokyo) pre-equilibrated with buffer III supplemented with 20 mM imidazole or a Chitin Resin column (New England Biolabs Japan, Tokyo) pre-equilibrated with buffer IV to purify His-tagged proteins (Mec, MetO, SCP, and MetM) or CBD-fused proteins (MetOΔC-int-CBD and MetO-G90-int-CBD). In a Ni-NTA agarose column, after washing with 7.5 CV of buffer III supplemented with 20 mM imidazole, the adsorbed proteins were eluted with 2.5 CV of buffer III supplemented with 500 mM imidazole. In the Chitin Resin column, after washing with 6 CV of buffer IV, flushing with 1 CV of buffer V (buffer IV supplemented with 30 mM DTT) for MetOΔC and MetO-G90, buffer VI (buffer IV supplemented with 50 mM Na2S, pH 8.5-9.0 adjusted with HCl) for MetOΔC-thiocarboxylate, or buffer VII (buffer IV supplemented with 50 mM cysteamine, pH 9.0 adjusted with HCl) for MetO-G90-cysteamine, and capping the columns, 3 CV of buffer V for MetOΔC and MetO-G90, buffer VI for MetOΔC-thiocarboxylate, or buffer VII for MetO-G90-cysteamine was added and incubated overnight at room temperature. Cleaved proteins were eluted with the incubated buffer and 1 CV of buffer V, VI, or VII. The proteins were dialyzed with buffer III or IV and concentrated with a Vivaspin Turbo 4 (MWCO, 3000) concentrator (Sartorius Japan K.K., Tokyo) before the enzyme activity assay. Protein concentrations were measured using a protein assay kit (Bio-Rad, Foster City, CA).
Enzyme activity assay.
The ability of Mec to hydrolyze a C-terminal peptide bond in MetO to MetOΔC and cysteine was examined by incubating 80 µM N-terminal His-tagged MetO or MetOΔC, 2 µM Mec, 10 µM ZnCl2, 100 mM HEPES-NaOH, pH 8.0, and 1 mM TCEP at 30°C for 1 h. After stopping the reaction by adding 1 volume (100 µL) of MeOH and subjecting the mixture to centrifugation at 20000 × g at 4°C for 10 min, supernatant samples supplemented with 0.5 mM tris(2-carboxyethyl)phosphine (TCEP) were analyzed by HPLC-HR-ESI-MS.
Enzymatic homocysteine synthase activity was examined as follows: purified MetM (5 µM) was mixed with 100 µM of substrate candidates (1, 3, 4, or 5), a sulfur source (81 µM MetOΔC-thiocarboxylate containing MetOΔC, 100 µM Na2S adjusted with HCl (pH8.0), or 100 µM Na2S2), and 100 mM HEPES-NaOH (pH 8.0) and incubated at 30°C for 1 h. After stopping the reaction by adding 1 volume (100 µL) of MeOH and subjecting the mixture to centrifugation at 20000 × g at 4°C for 10 min, supernatant samples supplemented with 0.5 mM TCEP were analyzed by HPLC-HR-ESI-MS. Marfey’s method was followed as previously described using Nα-(5-Fluoro-2,4-dinitrophenyl)-l-leucinamide (l-FDLA)51, and the reaction mixture was analyzed by HPLC-HR-ESI-MS.
An enzymatic activity check for MetM using a MetOΔC-thiocarboxylate analog, MetO-G90-cysteamine, was performed as follows: purified MetM (5 µM) was mixed with 100 µM 3, 1.75 mg/mL MetO-G90-cysteamine or Met-G90, and 100 mM HEPES-NaOH (pH 8.0) and incubated at 30°C for 1 h. After stopping the reaction by adding 1 volume (100 µL) of MeOH and subjecting the mixture to centrifugation at 20000 × g at 4°C for 10 min, supernatant samples supplemented with 0.5 mM TCEP were analyzed by HPLC-HR-ESI-MS.
HPLC-HR-ESI-MS analysis.
HPLC-HR-ESI-MS analysis was conducted using maXis plus (Bruker Daltonics, Bremen, Germany). Compounds were analyzed using a positive ion mode. For the detection of 2, a UPLC column, the XBridge® BEH Amide XP column (length, 2.1 × 50 mm; inner diameter (i.d.), 2.5 µm; Nihon Waters K.K., Tokyo), equipped with a guard column, the XBridge® BEH Amide XP VanGuard cartridge (length, 2.1 × 5 mm; i.d., 2.5 µm; Nihon Waters K.K., Tokyo), was used. The column was maintained at 40°C; 5 mM ammonium formate in 90% acetonitrile (solvent A) and 5 mM ammonium formate in 50% acetonitrile (solvent B) were used as mobile phases for the gradient elution of 2 at a 0.3 ml/min flow rate. The separation conditions were as follows: 0% solution B for 1 min, 0–100% solution B over 7 min, 100% solution B for 2 min, 100%-0% solution B over 1 min, and 0% solution B for 5 min.
For the detection of MetO derivatives and l-FDLA derivatives of 2 or cysteine, a UPLC column, the XBridge® Premier BEH C18 column (length, 2.1 × 50 mm; inner diameter (i.d.), 2.5 µm; Nihon Waters K.K., Tokyo), equipped with a guard column, XBridge® Premier BEH C18 VanGuard FIT cartridge (length, 2.1 × 5 mm; i.d., 2.5 µm; Nihon Waters K.K., Tokyo), was used. The column was maintained at 40°C; 0.1% formic acid in H2O (solvent A) and 0.1% formic acid in acetonitrile (solvent B) were used as mobile phases for the gradient elution at a 0.3 ml/min flow rate. The separation conditions were as follows: 10% solution B for 1 min, 10–90% solution B over 7 min, 90% solution B for 2 min, 90%-10% solution B over 1 min, and 10% solution B for 5 min.