Title: Characterization of the cystargolide biosynthetic gene cluster and functional analysis of the methyltransferase CysG
Abstract: Cystargolides are natural products originally isolated from Kitasatospora cystarginea NRRL B16505 as inhibitors of the proteasome. They are composed of a dipeptide backbone linked to a β-lactone warhead. Recently, we identified the cystargolide biosynthetic gene cluster, but systematic genetic analyses had not been carried out because of the lack of a heterologous expression system. Here, we report the discovery of a homologous cystargolide biosynthetic pathway in Streptomyces durhamensis NRRL-B3309 by genome mining. The gene cluster was cloned via transformation-associated recombination and heterologously expressed in Streptomyces coelicolor M512. We demonstrate that it contains all genes necessary for the production of cystargolide A and B. Single gene deletion experiments reveal that only five of the eight genes from the initially proposed gene cluster are essential for cystargolide synthesis. Additional insights into the cystargolide pathway could be obtained from in vitro assays with CysG and chemical complementation of the respective gene knockout. This could be further supported by the in vitro investigation of the CysG homolog BelI from the belactosin biosynthetic gene cluster. Thereby, we confirm that CysG and BelI catalyze a cryptic SAM-dependent transfer of a methyl group that is critical for the construction of the cystargolide and belactosin β-lactone warheads. Cystargolides are natural products originally isolated from Kitasatospora cystarginea NRRL B16505 as inhibitors of the proteasome. They are composed of a dipeptide backbone linked to a β-lactone warhead. Recently, we identified the cystargolide biosynthetic gene cluster, but systematic genetic analyses had not been carried out because of the lack of a heterologous expression system. Here, we report the discovery of a homologous cystargolide biosynthetic pathway in Streptomyces durhamensis NRRL-B3309 by genome mining. The gene cluster was cloned via transformation-associated recombination and heterologously expressed in Streptomyces coelicolor M512. We demonstrate that it contains all genes necessary for the production of cystargolide A and B. Single gene deletion experiments reveal that only five of the eight genes from the initially proposed gene cluster are essential for cystargolide synthesis. Additional insights into the cystargolide pathway could be obtained from in vitro assays with CysG and chemical complementation of the respective gene knockout. This could be further supported by the in vitro investigation of the CysG homolog BelI from the belactosin biosynthetic gene cluster. Thereby, we confirm that CysG and BelI catalyze a cryptic SAM-dependent transfer of a methyl group that is critical for the construction of the cystargolide and belactosin β-lactone warheads. The proteasome is the central component in non-lysosomal protein degradation (1Groll M. Ditzel L. Löwe J. Stock D. Bochtler M. Bartunik H.D. et al.Structure of 20S proteasome from yeast at 2.4 Å resolution.Nature. 1997; 386: 463-471Crossref PubMed Scopus (1976) Google Scholar, 2Ciechanover A. Proteolysis: from the lysosome to ubiquitin and the proteasome.Nat. Rev. Mol. Cell Biol. 2005; 6: 79-87Crossref PubMed Scopus (855) Google Scholar). It plays a vital role in a multitude of cellular processes, such as cell cycle control (3Clurman B.E. Sheaff R.J. Thress K. Groudine M. Roberts J.M. Turnover of cyclin E by the ubiquitin-proteasome pathway is regulated by cdk2 binding and cyclin phosphorylation.Genes. Dev. 1996; 10: 1979-1990Crossref PubMed Google Scholar), apoptosis (4Orlowski R.Z. The role of the ubiquitin-proteasome pathway in apoptosis.Cell Death Differ. 1999; 6: 303-313Crossref PubMed Google Scholar), and immune response (5Rock K.L. Goldberg A.L. Degradation of cell proteins and the generation of MHC class I-presented peptides.Annu. Rev. Immunol. 1999; 17: 739-779Crossref PubMed Scopus (803) Google Scholar). The multi-subunit protein complex is associated with various diseases including neurodegenerative (6Keller J.N. Hanni K.B. Markesbery W.R. Impaired proteasome function in Alzheimer's disease.J. Neurochem. 2000; 75: 436-439Crossref PubMed Scopus (721) Google Scholar, 7Ortega Z. Lucas J.J. Ubiquitin–proteasome system involvement in Huntington's disease.Front. Mol. Neurosci. 2014; 7: 77Crossref PubMed Scopus (81) Google Scholar) or cardiovascular disorders (8Pagan J. Seto T. Pagano M. Cittadini A. Role of the ubiquitin proteasome system in the heart.Circ. Res. 2013; 112: 1046-1058Crossref PubMed Scopus (100) Google Scholar) and different types of cancers (9Almond J. Cohen G. The proteasome: a novel target for cancer chemotherapy.Leukemia. 2002; 16: 433-443Crossref PubMed Google Scholar). Consequently, the proteasome is a validated target for the development of small-molecule pharmaceutics. Most of these are designed to covalently bind the catalytic subunits and inactivate the active site in a reversible or irreversible manner (10Crawford L.J. Walker B. Irvine A.E. Proteasome inhibitors in cancer therapy.J. Cell Commun. Signal. 2011; 5: 101-110Crossref PubMed Scopus (247) Google Scholar). Cystargolides A and B were originally isolated from Kitasatospora cystarginea NRRL B16505 and were found to inhibit the human proteasome and the caseinolytic protease ClpP in the micromolar range (11Gill K.A. Berrué F. Arens J.C. Carr G. Kerr R.G. Cystargolides, 20S proteasome inhibitors isolated from Kitasatospora cystarginea.J. Nat. Prod. 2015; 78: 822-826Crossref PubMed Scopus (20) Google Scholar, 12Niroula D. Hallada L.P. Le Chapelain C. Ganegamage S.K. Dotson D. Rogelj S. et al.Design, synthesis, and evaluation of cystargolide-based beta-lactones as potent proteasome inhibitors.Eur. J. Med. Chem. 2018; 157: 962-977Crossref PubMed Scopus (9) Google Scholar, 13Illigmann A. Vielberg M.T. Lakemeyer M. Wolf F. Dema T. Stange P. et al.Structure of Staphylococcus aureus ClpP bound to the covalent active-site inhibitor cystargolide A.Angew. Chem. Int. Ed. 2023; e202314028PubMed Google Scholar). The cystargolides feature a β-lactone warhead and a dipeptide backbone similar to the well-characterized belactosins (Fig. 1). Belactosins A and C are produced by Streptomyces sp. UCK 14 and, in analogy to the cystargolides, primarily block the function of the chymotrypsin-like β5 subunit of the 20S proteasome (14Asai A. Hasegawa A. Ochiai K. Yamashita Y. Mizukami T. Belactosin A, a novel antituor antibiotic acting on cyclin/CDK mediated cell cycle regulation, produced by Streptomyces sp.J. Antibiot. 2000; 53: 81-83Crossref PubMed Scopus (110) Google Scholar, 15Groll M. Larionov O.V. Huber R. de Meijere A. Inhibitor-binding mode of homobelactosin C to proteasomes: new insights into class I MHC ligand generation.Proc. Natl. Acad. Sci. U. S. A. 2006; 103: 4576-4579Crossref PubMed Scopus (69) Google Scholar). Notably, many β-lactone-containing secondary metabolites are synthesized by non-ribosomal peptide synthetases (NRPSs) or polyketide synthases (PKSs) (16Robinson S.L. Christenson J.K. Wackett L.P. Biosynthesis and chemical diversity of β-lactone natural products.Nat. Prod. Rep. 2019; 36: 458-475Crossref PubMed Google Scholar, 17Kaysser L. Built to bind: biosynthetic strategies for the formation of small-molecule protease inhibitors.Nat. Prod. Rep. 2019; 36: 1654-1686Crossref PubMed Google Scholar). We recently identified the cystargolide biosynthetic gene cluster (BGC) and constructed gene disruption mutants in K. cystarginea NRRL B16505 (18Wolf F. Bauer J.S. Bendel T.M. Kulik A. Kalinowski J. Gross H. et al.Biosynthesis of the β-lactone proteasome inhibitors belactosin and cystargolide.Angew. Chem. Int. Ed. 2017; 56: 6665-6668Crossref PubMed Scopus (29) Google Scholar). This revealed that the cystargolide and belactosin formation does not involve NRPS or PKS machinery but proceeds via single enzyme amino acid ligases. Intriguingly, the route to the carbon scaffold of their β-lactone warheads is reminiscent of leucine biosynthesis. Here, we present a homologous cystargolide BGC from Streptomyces durhamensis NRRL-B3309. The gene cluster was cloned via transformation-associated recombination (TAR) and heterologously expressed in Streptomyces coelicolor M512 allowing a systematic set of gene knockouts to investigate the functions of the encoded proteins. In order to characterize the biosynthetic pathway of the cystargolides, we attempted to heterologously express the BGC we identified in K. cystarginea NRRL B16505 (18Wolf F. Bauer J.S. Bendel T.M. Kulik A. Kalinowski J. Gross H. et al.Biosynthesis of the β-lactone proteasome inhibitors belactosin and cystargolide.Angew. Chem. Int. Ed. 2017; 56: 6665-6668Crossref PubMed Scopus (29) Google Scholar). However, this pathway could not be stably introduced in various Streptomyces strains. Using MultiGeneBLAST, we discovered a homologous putative cystargolide (cys) BGC in S. durhamensis NRRL-B3309 (19Medema M.H. Takano E. Breitling R. Detecting sequence homology at the gene cluster level with MultiGeneBlast.Mol. Biol. Evol. 2013; 30: 1218-1223Crossref PubMed Scopus (244) Google Scholar). This pathway differs only marginally from the one found in K. cystarginea NRRL B16505, with all eight genes showing sequence identities between 63% and 83% to their respective homologs (Fig. 2; Table S1). Organizationally, only the orientation of cysH in S. durhamensis NRRL-B3309 is inverted compared to K. cystarginea NRRL B16505. To confirm the identity of this BGC, the cluster was captured with a pCAP03-derived vector using TAR-based direct cloning technologies (Fig. S3) (20Kouprina N. Larionov V. Selective isolation of genomic loci from complex genomes by transformation-associated recombination cloning in the yeast Saccharomyces cerevisiae.Nat. Protoc. 2008; 3: 371Crossref PubMed Scopus (159) Google Scholar, 21Yamanaka K. Reynolds K.A. Kersten R.D. Ryan K.S. Gonzalez D.J. Nizet V. et al.Direct cloning and refactoring of a silent lipopeptide biosynthetic gene cluster yields the antibiotic taromycin A.Proc. Natl. Acad. Sci. U. S. A. 2014; 111: 1957-1962Crossref PubMed Scopus (353) Google Scholar, 22Tang X. Li J. Millán-Aguiñaga N. Zhang J.J. O'Neill E.C. Ugalde J.A. et al.Identification of thiotetronic acid antibiotic biosynthetic pathways by target-directed genome mining.ACS Chem. Biol. 2015; 10: 2841-2849Crossref PubMed Scopus (187) Google Scholar). Therefore, pCAP03 was linearized and equipped with 299 bp extensions on both ends which contained homologous sequences up- and downstream the putative cys BGC. Co-transformation of Saccharomyces cerevisiae with the capture vector and digested genomic DNA from S. durhamensis resulted in the cloning of a 10,809 bp stretch of DNA that contained the complete putative cys BGC as verified by PCR and restriction analysis. Subsequently, the generated cysDM02-construct was introduced into S. coelicolor M512 via triparental mating and stably integrated into the genome taking advantage of the φC31 integration site. Three individual clones were checked by PCR for the presence of the cys BGC, designated S. coelicolor M512/cysDM02 I-III, and selected for further investigations. Therefore, cells were cultivated in liquid cultures of SM1 medium for 7 days followed by crude extraction with ethyl acetate (EtOAc). A first screening approach revealed the specific activity of extracts from mutants containing the homologous cys pathway in a human proteasome inhibition assay (Fig. S5). Next, the metabolic profiles of the mutants were analyzed by high-performance liquid chromatography coupled with electrospray ionization mass spectrometry (HPLC-ESI-MS). Gratifyingly, the characteristic mass signals of cystargolide A (m/z 357.2 [M + H]+) and cystargolide B (m/z 371.2 [M + H]+) were detected at retention times of 10.2 and 9.5 min, respectively (Fig. 3). Similar ions could not be found in extracts of S. coelicolor M512 without the gene cluster. Moreover, tandem mass spectrometry (MS/MS) revealed characteristic fragmentation patterns for the produced compounds (Fig. S10) identical to cystargolide A and B (18Wolf F. Bauer J.S. Bendel T.M. Kulik A. Kalinowski J. Gross H. et al.Biosynthesis of the β-lactone proteasome inhibitors belactosin and cystargolide.Angew. Chem. Int. Ed. 2017; 56: 6665-6668Crossref PubMed Scopus (29) Google Scholar). This confirms that the isolated BGC from S. durhamensis NRRL-B3309 contains all genes necessary for the synthesis of cystargolides. In order to determine the function of the clustered genes, single-gene knockout variants of the cys BGC were heterologously expressed in S. coelicolor M512. Therefore, the genes were individually replaced with an apramycin resistance cassette using the PCR-targeting system (23Gust B. Challis G.L. Fowler K. Kieser T. Chater K.F. PCR-targeted Streptomyces gene replacement identifies a protein domain needed for biosynthesis of the sesquiterpene soil odor geosmin.Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 1541-1546Crossref PubMed Scopus (1242) Google Scholar). Unmarked in-frame gene deletions were generated by excision of the resistance cassettes, taking advantage of the flanking FLP recombination target (FRT) sites. S. coelicolor M512 derivatives containing the modified gene clusters were cultivated and analyzed by LC-MS. This revealed that cystargolide production was not impacted in the knockout mutants with ΔcysA and ΔcysB (Fig. 3). Interestingly, the gene cysA codes for an isopropylmalate synthase-homolog and is likely to be involved in the formation of 2-isopropylmalate (2-IPM) from acetyl-CoA and α-ketoisovalerate. Isopropylmalate synthases (IPMS) are essential for de novo leucine biosynthesis and are a central part of the bacterial primary metabolism. Therefore, the loss of cysA might be complemented by the house-keeping IPMS homolog from the host organism. Indeed, a BLAST analysis of the genome of S. coelicolor detected a putative IPMS (SCO2528) with a sequence identity of 58% on protein level. CysB is a putative major facilitator superfamily (MFS) permease of the DHA2-family. Such transporters have been associated with the extrusion of a variety of chemicals and can confer resistance and self-resistance to antibiotically active compounds (24Bolhuis H. van Veen H.W. Poolman B. Driessen A.J. Konings W.N. Mechanisms of multidrug transporters.FEMS Microbiol. Rev. 1997; 21: 55-84Crossref PubMed Google Scholar, 25Garrido M.C. Herrero M. Kolter R. Moreno F. The export of the DNA replication inhibitor Microcin B17 provides immunity for the host cell.EMBO J. 1988; 7: 1853-1862Crossref PubMed Scopus (147) Google Scholar). We propose that CysB is involved in the export of cystargolides. However, the detected cystargolide production in the respective ΔcysB mutant suggests that the function of CysB in cystargolide secretion is supported by the activity of other intrinsic transporter systems. Indeed, analysis of the cell culture supernatant and the mycelium showed that cystargolides are present in both, the medium and the cells. This was observed for the intact gene cluster as well as the ΔcysB mutant (Fig. S6). The gene cysH encodes a putative LysR-family regulatory protein. These type of proteins are known as transcriptional activators (26Schell M.A. Molecular biology of the LysR family of transcriptional regulators.Annu. Rev. Microbiol. 1993; 47: 597-626Crossref PubMed Google Scholar). LC-MS analysis of the ΔcysH mutant culture extracts revealed a drastically reduced cystargolide production (Table S6). This strongly indicates that CysH functions as a positive regulator in regards to the cystargolide biosynthesis. In contrast, the knockouts ΔcysC, ΔcysD, ΔcysE, ΔcysF and ΔcysG resulted in the complete abolishment of cystargolide formation (Fig. 3). Of those, three genes, cysC, cysD and cysF, code for putative ATP-dependent carboxylic acid-activating enzymes. CysC and CysF show homology to adenylating enzymes. They generate high-energy acyl-AMP phosphoester intermediates to facilitate e.g. the formation of amide bonds, coenzyme A and acyl-carrier protein (ACP) thioesters or general transesterification reactions. CysD is predicted as a member of the ATP-grasp family which forms reactive acyl monophosphate intermediates that allow similar biochemistry as the adenylates (27Fawaz M.V. Topper M.E. Firestine S.M. The ATP-grasp enzymes.Bioorg. Chem. 2011; 39: 185-191Crossref PubMed Scopus (131) Google Scholar, 28Baulig A. Helmle I. Bader M. Wolf F. Kulik A. Al-Dilaimi A. et al.Biosynthetic reconstitution of deoxysugar phosphoramidate metalloprotease inhibitors using an N–P-bond-forming kinase.Chem. Sci. 2019; 10: 4486-4490Crossref PubMed Google Scholar, 29Quitterer F. List A. Beck P. Bacher A. Groll M. Biosynthesis of the 22nd genetically encoded amino acid pyrrolysine: structure and reaction mechanism of PylC at 1.5A resolution.J. Mol. Biol. 2012; 424: 270-282Crossref PubMed Scopus (16) Google Scholar). We thus conclude that CysC, CysD, and CysF are involved in converting three carboxylic acid groups to form two amides and a cyclic ester in cystargolide biosynthesis. On the other hand, CysG is a putative SAM-dependent methyltransferase whereas CysE is homologous to methylesterases with a predicted alpha/beta-hydrolase topology. Both enzymes have no obvious function in cystargolide biosynthesis, but intriguingly, a similar methyltransferase/methylesterase pair is encoded in the biosynthetic gene cluster of belactosins (Fig. 2; Table S2) (18Wolf F. Bauer J.S. Bendel T.M. Kulik A. Kalinowski J. Gross H. et al.Biosynthesis of the β-lactone proteasome inhibitors belactosin and cystargolide.Angew. Chem. Int. Ed. 2017; 56: 6665-6668Crossref PubMed Scopus (29) Google Scholar, 30Shimo S. Ushimaru R. Engelbrecht A. Harada M. Miyamoto K. Kulik A. et al.Stereodivergent nitrocyclopropane formation during biosynthesis of belactosins and hormaomycins.J. Am. Chem. Soc. 2021; 143: 18413-18418Crossref PubMed Scopus (24) Google Scholar, 31Engelbrecht A. Wolf F. Esch A. Kulik A. Kozhushkov S.I. de Meijere A. et al.Discovery of a cryptic nitro intermediate in the biosynthesis of the 3-(trans-2'-Aminocyclopropyl)alanine moiety of belactosin A.Org. Lett. 2022; 24: 736-740Crossref PubMed Scopus (10) Google Scholar). This finding led us to speculate that CysE and CysG might be involved in the formation of the characteristic β-lactone warhead. Indeed, the complete loss of cystargolide production in the respective gene deletion mutants appear to support this hypothesis (Fig. 3). To further explore the role of CysG in cystargolide biosynthesis we cloned and overexpressed cysG in Escherichia coli Rosetta DE3 pLysS. Using an N-terminal His-tag and Ni-NTA affinity chromatography 10 mg/L pure and soluble recombinant protein was obtained (Fig. S7). Given the absence of a methyl group in the final cystargolide product we wondered if CysG might act on one of the building blocks or their precursors prior to the final compound assembly. We therefore incubated CysG with S-adenosyl methionine (SAM) and 3-isopropylmalate (3-IPM) as a putative acceptor substrate. Strikingly, a single product peak could be detected by LC-MS analysis with m/z 191.08 [M + H]+, matching the expected mono-methylated 3-IPM (Fig. 4A). This compound was absent in the control assays lacking either CysG or SAM. Since 3-IPM offers three possible targets for methyl-group transfer (2-OH, COOH-1, and COOH-4), the product of the CysG-mediated biotransformation was subjected to structural characterization by NMR spectroscopy (Figs. S11–S28 and Tables S7–S9). A comparison of the 1D NMR spectra of 3-IPM with those of its methylated version revealed that all carbon atoms except C-2 and C-7 experienced slight field shifts in the range of 0.4 to 1 ppm in the 13C NMR spectrum and that one additional oxygenated methyl group (δH/C 3.60/51.4) was formed (Tables S7 and S8). In the corresponding 1H-13C-HMBC NMR spectrum, the latter showed long-range correlations either with C-1 or C4, indicating that one of the carboxylic acids was esterified. Due to the biotransformation, C-1 and C-4 shifted and exhibited highly similar 13C NMR shift values (δC 174.03 and 174.22), which prevented initially an unambiguous assignment. Since the application of a higher magnetic field strength (16.4 T instead of 9.4 T) to increase resolution was only marginally successful, an acetylation strategy was applied. Peracetylation of methylated 3-IPM and subsequent NMR analysis unveiled, that the 2-OH group was acetylated as expected, evidenced by cross peaks H3-10/C-9 and H3-10/C-2 in the HMBC spectrum. This modification enabled carbon atoms C-1 and C-4 to be sufficiently magnetically non-equivalent and well resolved (δC 169.2 and 171.6, respectively), which allowed the determination of the exact methylation site. Consequently, the observed long-range coupling between H3-8 and C-1 (Fig. S25) unequivocally led to the conclusion that the original biotransformation product represented indeed 3-IPM-1-methyl-ester. Next, we wanted to confirm that the CysG in vitro product is a genuine intermediate in the cystargolide pathway. Therefore, we fed 3-IPM-1-methyl-ester to cultures of the ΔcysG mutant strain. LC-MS analysis of the respective extracts revealed that the production of cystargolides was restored only in the supplemented cultures (Figs. 3, S9 and S10). Chemical complementation was not observed after feeding with non-methylated 3-IPM, demonstrating the importance of the installed 1-methyl group for the subsequent biosynthetic steps in cystargolide biosynthesis. Our unexpected findings on the role of CysG in cystargolide formation prompted us to examine if the homologous enzyme BelI has a similar function in belactosin biosynthesis. Therefore, we cloned and overexpressed belI in E. coli BL21(DE3) to obtain pure and soluble recombinant protein with an N-terminal His6-SUMO-tag (Fig. S8). Because the predicted substrate of BelI 3-sec butylmalate, was not commercially available we accessed this compound by chemical synthesis using diethyl malate and 2-iodobutane as starting materials (32Khan A.A. Chee S.H. Stocker B.L. Timmer M.S. The synthesis of long-chain α-Alkyl-β-Hydroxy esters using allylic halides in a Fráter–seebach alkylation.Eur. J. Org. Chem. 2012; 5: 995-1002Crossref Scopus (9) Google Scholar, 33de Meijere A. Korotkov V.S. Lygin A.V. Larionov O.V. Sokolov V.V. Graef T. et al.Synthesis and biological activity of simplified belactosin C analogues.Org. Biomol. Chem. 2012; 10: 6363-6374Crossref PubMed Scopus (14) Google Scholar). In the next step, BelI was incubated with SAM and authentic 3-sec butylmalate to provide a single product peak in LC-MS chromatograms (Fig. 4B). The respective molecular ion had a mass-to-charge ratio of m/z 202.69 [M + H]+, matching the expected mono-methylated 3-sec butylmalate. A similar compound was not identified in the control assays lacking either BelI or SAM. These results clearly showed that the biosynthesis of the β-lactone warhead moiety of cystargolides and belactosins proceeds via a cryptic methylated intermediate prior to cyclization. In this study we identified and isolated a cystargolide gene cluster from S. durhamensis NRRL-B3309. The cluster contains eight genes, which are homologs of the original cys BGC from K. cystarginea NRRL B16505. Via heterologous expression, we could unambiguously connect this BGC to the biosynthesis of cystargolides, confirming that it contains all genes necessary for the production of cystargolide A and B. Furthermore, we were able to demonstrate that the genes cysC-cysG are essential for this process. The putative methyltransferase CysG readily methylated 3-IPM in vitro, suggesting that it catalyzes the same reaction in the cystargolide biosynthetic pathway and plays a crucial role in the formation of the β-lactone warhead. The methylation occurs at the C-1 carboxylic acid group of 3-IPM as revealed by NMR spectroscopy, proving that this modification is not involved in the lactonization. Thus we propose that CysG might act as a diverter, committing the primary metabolite 3-IPM to cystargolide biosynthesis and/or masking the 1-carboxylic acid group for the following biotransformations. These findings were supported by the functional characterization of the CysG homolog BelI. BelI catalyzed the methylation of 3-sec butylmalate in a SAM-dependent manner. Intriguingly, CysG and CysE as well as BelI and BelR exhibit conserved protein domains that are found in the methyltransferase/methylesterase pair BioC/BioH of the biotin synthesis pathway in proteobacteria (Figs. S1 and S2) (34Lin S. Hanson R.E. Cronan J.E. Biotin synthesis begins by hijacking the fatty acid synthetic pathway.Nat. Chem. Biol. 2010; 6: 682-688Crossref PubMed Scopus (151) Google Scholar). BioC catalyzes a SAM-dependent methyl esterification of malonyl-ACP allowing its recognition as a starter unit by house-keeping charge-sensitive fatty acid synthetases. After two reiterations of the fatty acid elongation cycle, the generated pimeloyl-ACP methyl ester is hydrolyzed to pimeloyl-ACP and methanol by BioH. Pimeloyl-ACP is then further converted to biotin. One might speculate about a similar role for CysG and CysE in cystargolide biosynthesis. In this reaction trajectory the 3-IPM-1-methyl-ester product of CysG would undergo lactonization to afford a β-lactone methyl ester building block. Prior to amidation, the 1-carboxyl group is presumably liberated by CysE. Investigations to explore this hypothesis and open questions about the formation of the β-lactone warhead of cystargolides and belactosins will be addressed in future studies. Chemicals, micro-, and molecular biological agents were acquired from standard commercial sources. S. coelicolor M512 and its derivatives were grown and maintained on MS agar (2% (w/v) soy flour purchased from Sobo Naturkost, 2% (w/v) mannitol purchased from Roth, 2% (w/v) agar purchased from Becton Dickinson) (35Kieser T. Bibb M. Buttner M. Chater K. Hopwood D. Practical Streptomyces Genetics. The John Innes Foundation, Norwich, UK2000Google Scholar). Liquid cultures were either cultivated in Tryptic Soy Broth (TSB) medium (purchased from Becton Dickinson) or in SM1 Medium (10% (w/v) soy flour, 18% glucose, 1% Na2SO4 and 0.2% CaCO3). E. coli strains were grown in lysogeny broth (LB) medium supplemented with appropriate antibiotics. DNA isolation and manipulations were carried out according to standard methods for E. coli (36Sambrook J. Russell D.W. Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory Press, New York2001Google Scholar) and Streptomyces (35Kieser T. Bibb M. Buttner M. Chater K. Hopwood D. Practical Streptomyces Genetics. The John Innes Foundation, Norwich, UK2000Google Scholar). The 20S human proteasome activity was analyzed using the chymotryptic model substrate LLVY-AMC (Bachem). LLVY-AMC was prediluted in assay buffer to a concentration of 50 μM (20 mM HEPES, pH 7.8, 0.5 mM EDTA/0.035% SDS), and varying dilutions of culture extracts (1:300 and 1:600) were added. The reaction was started by the addition of the human proteasome (Enzo Life Sciences: BML-PW8720-0050) to a final concentration of 0.4 μg/ml at a reaction volume of 100 μl to monitor substrate conversion. The fluorescence was followed over time in a spectrofluorometer (Tecan Spark, Tecan Trading AG) at λex = 380 nm and λem = 460 nm. All measurements were performed in triplicates and two independent experiments were performed. The vector pCAP03-acc(3)IV was used to isolate the cys BGC from S. durhamensis NRRL-B3309 creating cysDM02 following published transformation-assisted recombination protocols (21Yamanaka K. Reynolds K.A. Kersten R.D. Ryan K.S. Gonzalez D.J. Nizet V. et al.Direct cloning and refactoring of a silent lipopeptide biosynthetic gene cluster yields the antibiotic taromycin A.Proc. Natl. Acad. Sci. U. S. A. 2014; 111: 1957-1962Crossref PubMed Scopus (353) Google Scholar). Therefore, homologous hooks of 299 bp (Table S3) were integrated in pCAP03-acc(3)IV by Gibson assembly. The vector was linearized by restriction digest with XhoI and NdeI. For the preparation of spheroplasts S. cerevisiae VL6-48N was cultured in 50 ml YPD medium (20 g/L peptone, 20 g/L glucose, 10 g/L yeast extract) with 1% (w/v) adenine 2% (w/v) glucose until an OD660 of 3.0 was reached. The cells were harvested at 4 °C and 2000g for 5 min and resuspended in 30 ml H2O. After another centrifugation step at 3000g the pellet was resuspended in 20 ml ice-cold 1 M D-sorbit. The cells were incubated on ice for 30 min, centrifuged at 4 °C and 3000g for 5 min and resuspended in 20 ml SPE solution. 125 μl zymolyase and 40 μl β-mercaptoethanol were added and the suspension was incubated at 30 °C and 200 rpm until spheroplast generation was complete. Progress was monitored by measuring the OD660 of 100 μl cells diluted with 900 μl 1M D-sorbit solution. The measurements were compared to the OD660 of 100 μl cells diluted in a 2% (v/v) sodium dodecyl sulfate (SDS) solution. Single gene knockouts in the cys BGC were generated by PCR-targeting using λ-RED recombineering (23Gust B. Challis G.L. Fowler K. Kieser T. Chater K.F. PCR-targeted Streptomyces gene replacement identifies a protein domain needed for biosynthesis of the sesquiterpene soil odor geosmin.Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 1541-1546Crossref PubMed Scopus (1242) Google Scholar). Therefore, an apramycin resistance cassette was amplified and extended from pIJ773 by PCR (Table S5 and Fig. S4). The PCR product was transferred into E. coli BW25113/pKD46 containing fosmid cysDM02 (37Datsenko K.A. Wanner B.L. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products.Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 6640-6645Crossref PubMed Scopus (11654) Google Scholar). After confirmation with restriction analysis, the resistance cassette was removed in E. coli BT340 to avoid polar