Two pathways for pyrrole formation in coumermycin A(1) biosynthesis: the central pyrrole moiety is formed from L-threonine

Coumermycin A₁ is an aminocoumarin antibiotic produced by Streptomyces rishiriensis. It contains three pyrrole rings, that is, two terminal 5‐methyl‐pyrrole‐2‐carboxyl moieties and a central 3‐methylpyrrole‐2,4‐dicarboxylic acid moiety. The biosynthesis of the terminal pyrrole moieties has been elucidated previously. However, the biosynthetic precursors of the central pyrrole moiety have remained unknown, and none of the genes or enzymes involved in its formation has been identified. We now show that five genes, contained in a contiguous 4.7 kb region within the coumermycin biosynthetic gene cluster, are required for the biosynthesis of this central pyrrole moiety. Each of these genes was deleted individually, resulting in a strong reduction or an abolishment of coumermycin production. External feeding of the central pyrrole moiety restored coumermycin production. One of these genes shows similarity to L ‐threonine kinase genes. Feeding of [U‐¹³C,¹⁵N]L ‐threonine and ¹³C NMR analysis of the resulting compound unequivocally proved that threonine was incorporated intact into the central pyrrole (19 % enrichment) to provide the heterocyclic nitrogen as well as four of the seven carbons of this moiety. Therefore, this pyrrole is formed via a new, hitherto unknown biosynthetic pathway. A hypothesis for the reaction sequence leading to the central pyrrole moiety of coumermycin A₁ is presented.     

Reevaluation of the violacein biosynthetic pathway and its relationship to indolocarbazole biosynthesis

The biosynthetic pathways for violacein and for indolocarbazoles (rebeccamycin, staurosporine) include a decarboxylative fusion of two tryptophan units. However, in the case of violacein, one of the tryptophans experiences an unusual 1-->2 shift of the indole ring. The violacein biosynthetic gene cluster was previously reported to consist of four genes, vioABCD. Here we studied the violacein pathway through expression of vio genes in Escherichia coli and Streptomyces albus. A pair of genes (vioAB), responsible for the earliest steps in violacein biosynthesis, was functionally equivalent to the homologous pair in the indolocarbazole pathway (rebOD), directing the formation of chromopyrrolic acid. However, chromopyrrolic acid appeared to be a shunt product, not a violacein intermediate. In addition to vioABCD, a fifth gene (vioE) was essential for violacein biosynthesis, specifically for production of the characteristic 1-->2 shift of the indole ring. We also report new findings on the roles played by the VioC and VioD oxygenases, and on the origin of violacein derivatives of the chromoviridans type.     

A gene cluster for prenylated naphthoquinone and prenylated phenazine biosynthesis in Streptomyces cinnamonensis DSM 1042

Streptomyces cinnamonensis DSM 1042 produces two classes of secondary metabolites of mixed isoprenoid/nonisoprenoid origin: the polyketide-isoprenoid compound furanonaphthoquinone I (FNQ I) and several prenylated phenazines, predominantly endophenazine A. We now report the cloning and sequence analysis of a 55 kb gene cluster required for the biosynthesis of these compounds. Several inactivation experiments confirmed the involvement of this gene cluster in the biosynthesis of FNQ I and endophenazine A. The six identified genes for endophenazine biosynthesis showed close similarity to phenazine biosynthetic genes from Pseudomonas. Of the 28 open reading frames identified in the adjacent FNQ I cluster, 13 showed close similarity to genes contained in the cluster for furaquinocin-a structurally similar metabolite from another Streptomyces strain. These genes included a type III polyketide synthase sequence, a momA-like monooxygenase gene, and two cloQ-like prenyltransferase genes designated fnq26 and fnq28. Inactivation experiments confirmed the involvement of fnq26 in FNQ I biosynthesis, whereas no change in secondary-metabolite formation was observed after fnq28 inactivation. The FNQ I cluster contains a contiguous group of five genes, which together encode all the enzymatic functions required for the recycling of S-adenosylhomocysteine (SAH) to S-adenosylmethionine (SAM). Two SAM-dependent methyltransferases are encoded within the cluster. Inactivation experiments showed that fnq9 is responsible for the 7-O-methylation and fnq27 for the 6-C-methylation reaction in FNQ I biosynthesis.     

Use of a halogenase of hormaomycin biosynthesis for formation of new clorobiocin analogues with 5-chloropyrrole moieties

The depsipeptide antibiotic hormaomycin, which is produced by Streptomyces griseoflavus W-384, contains a 5-chloropyrrole moiety. In the producer strain we identified the gene hrmQ that shows sequence similarity to FADH(2)-dependent halogenases. This gene was cloned and heterologously expressed in Streptomyces roseochromogenes var. oscitans DS12.976, which is the producer of the aminocoumarin antibiotic clorobiocin, which contains a 5-methylpyrrole moiety. For the present experiment, we used a mutant of this strain in which the respective pyrrole-5-methyltransferase had been inactivated. Expression of the halogenase hrmQ in this mutant strain led to the formation of two new clorobiocin derivatives that carried a 5-chloropyrrole moiety. These compounds were isolated on a preparative scale, their structures were elucidated by (1)H NMR spectroscopy and mass spectrometry, and their antibacterial activity was determined. The substrate of HrmQ is likely to be a pyrrole-2-carboxyl-S-[acyl carrier protein] thioester. If this assumption is true, this study presents the first experiment in combinatorial biosynthesis that uses a halogenase that acts on an acyl carrier protein-bound substrate.     

Epoxyquinone formation catalyzed by a two-component flavin-dependent monooxygenase involved in biosynthesis of the antibiotic actinorhodin

The biosynthetic gene cluster of the aromatic polyketide antibiotic actinorhodin (ACT) in Streptomyces coelicolor A3(2) carries a pair of genes, actVA-ORF5 and actVB, that encode a two-component flavin-dependent monooxygenase (FMO). Our previous studies have demonstrated that the ActVA-ORF5/ActVB system functions as a quinone-forming C-6 oxygenase in ACT biosynthesis. Furthermore, we found that this enzyme system exhibits an additional oxygenation activity with dihydrokalafungin (DHK), a proposed intermediate in the ACT biosynthetic pathway, and generates two reaction products. These compounds were revealed to be monooxygenated derivatives of kalafungin, which is spontaneously formed through oxidative lactonization of DHK. Their absolute structures were elucidated from their NMR spectroscopic data and by computer modeling and X-ray crystallography as (5S,14R)-epoxykalafungin and (5R,14S)-epoxykalafungin, demonstrating an additional epoxyquinone-forming activity of the ActVA-ORF5/ActVB system in vitro.     

In vivo mutational analysis of the mupirocin gene cluster reveals labile points in the biosynthetic pathway: the "leaky hosepipe" mechanism

A common feature of the mupirocin and other gene clusters of the AT-less polyketide synthase (PKS) family of metabolites is the introduction of carbon branches by a gene cassette that contains a beta-hydroxy-beta-methylglutaryl CoA synthase (HMC) homologue and acyl carrier protein (ACP), ketosynthase (KS) and two crotonase superfamily homologues. In vivo studies of Pseudomonas fluorescens strains in which any of these components have been mutated reveal a common phenotype in which the two major isolable metabolites are the truncated hexaketide mupirocin H and the tetraketide mupiric acid. The structure of the latter has been confirmed by stereoselective synthesis. Mupiric acid is also the major metabolite arising from inactivation of the ketoreductase (KR) domain of module 4 of the modular PKS. A number of other mutations in the tailoring region of the mupirocin gene cluster also result in production of both mupirocin H and mupiric acid. To explain this common phenotype we propose a mechanistic rationale in which both mupirocin H and mupiric acid represent the products of selective and spontaneous release from labile points in the pathway that occur at significant levels when mutations block the pathway either close to or distant from the labile points.     

Functional analysis of the aureothin iterative type I polyketide synthase

The modular-type polyketide synthase (PKS) that is involved in aureothin (aur) biosynthesis represents one of the first examples in which a single PKS module (AurA) is used in an iterative fashion. Here we report on the heterologous expression of an engineered AurAB fusion protein that unequivocally proves the iterative nature of AurA. In addition, point mutations reveal that aur PKS module 4 participates in polyketide biosynthesis despite its aberrant acyltransferase domain.     

Discovery of the Tiancilactone Antibiotics by Genome Mining of Atypical Bacterial Type II Diterpene Synthases

Although genome mining has advanced the identification, discovery, and study of microbial natural products, the discovery of bacterial diterpenoids continues to lag behind. Herein, we report the identification of 66 putative producers of novel bacterial diterpenoids, and the discovery of the tiancilactone (TNL) family of antibiotics, by genome mining of type II diterpene synthases that do not possess the canonical DXDD motif. The TNLs, which are broad‐spectrum antibiotics with moderate activities, are produced by both Streptomyces sp. CB03234 and Streptomyces sp. CB03238 and feature a highly functionalized diterpenoid skeleton that is further decorated with chloroanthranilate and γ‐butyrolactone moieties. Genetic manipulation of the tnl gene cluster resulted in TNL congeners, which provided insights into their biosynthesis and structure–activity relationships. This work highlights the biosynthetic potential that bacteria possess to produce diterpenoids and should inspire continued efforts to discover terpenoid natural products from bacteria.     

Parallel Post‐Polyketide Synthase Modification Mechanism Involved in FD‐891 Biosynthesis in Streptomyces graminofaciens A‐8890

To isolate a key polyketide biosynthetic intermediate for the 16‐membered macrolide FD‐891 ( 1 ), we inactivated two biosynthetic genes coding for post‐polyketide synthase (PKS) modification enzymes: a methyltransferase (GfsG) and a cytochrome P450 (GfsF). Consequently, FD‐892 ( 2 ), which lacks the epoxide moiety at C8–C9, the hydroxy group at C10, and the O‐methyl group at O‐25 of FD‐891, was isolated from the gfsF/gfsG double‐knockout mutant. In addition, 25‐O‐methyl‐FD‐892 ( 3 ) and 25‐O‐demethyl‐FD‐891 ( 4 ) were isolated from the gfsF and gfsG mutants, respectively. We also confirmed that GfsG efficiently catalyzes the methylation of 2 and 4 in vitro. Further, GfsF catalyzed the epoxidation of the double bond at C8‐C9 of 2 and 3 and subsequent hydroxylation at C10, to afford 4 and 1 , respectively. These results suggest that a parallel post‐PKS modification mechanism is involved in FD‐891 biosynthesis.     

Harnessing the Synthetic Capabilities of Glycopeptide Antibiotic Tailoring Enzymes: Characterization of the UK‐68,597 Biosynthetic Cluster

In this study, a draft genome sequence of Actinoplanes sp. ATCC 53533 was assembled, and an 81‐kb biosynthetic cluster for the unusual sulfated glycopeptide UK‐68,597 was identified. Glycopeptide antibiotics are important in the treatment of infections caused by Gram‐positive bacteria. Glycopeptides contain heptapeptide backbones that are modified by many tailoring enzymes, including glycosyltransferases, sulfotransferases, methyltransferases, and halogenases, generating extensive chemical and functional diversity. Several tailoring enzymes in the cluster were examined in vitro for their ability to modify glycopeptides, resulting in the synthesis of novel molecules. Tailoring enzymes were also expressed in the producer of the glycopeptide aglycone A47934, generating additional chemical diversity. This work characterizes the biosynthetic program of UK‐68,597 and demonstrates the capacity to expand glycopeptide chemical diversity by harnessing the unique chemistry of tailoring enzymes.     

The butenolide signaling molecules SRB1 and SRB2 induce lankacidin and lankamycin production in Streptomyces rochei

New signaling molecules that induce lankacidin and lankamycin production in Streptomyces rochei were extracted from the culture filtrate and purified by Sephadex LH20 and silica gel chromatography with the help of bioassay. Chiral HPLC and ESI-MS analyses indicated the presence of two active components--SRB1 and SRB2--and their molecular formulas were established to be C15H24O5 and C16H26O5, respectively. By extensive NMR analysis, SRB1 and SRB2 were determined to be 2-(1'-hydroxy-6'-oxo-8'-methylnonyl)-3-methyl-4-hydroxybut-2-en-1,4-olide and 2-(1'-hydroxy-6'-oxo-8'-methyldecyl)-3-methyl-4-hydroxybut-2-en-1,4-olide, respectively. These structures were finally confirmed by chemical synthesis and the absolute configuration at C-1' was determined to be R in each case. The synthetic 1'R isomers induced production of lankacidin and lankamycin at around 40 nM concentrations. SRB1 and SRB2 are therefore distinct from the well-known 2,3-disubstituted γ-butyrolactone molecules such as A-factor, virginia butanolide, and SCB1 and and belong, like avenolide, recently isolated from Streptomyces avermitilis, to the γ-butenolide family.     

Interrogating the Tailoring Steps of Pactamycin Biosynthesis and Accessing New Pactamycin Analogues

Pactamycin is a bacteria‐derived aminocyclitol antibiotic with a wide‐range of biological activity. Its chemical structure and potent biological activities have made it an interesting lead compound for drug discovery and development. Despite its unusual chemical structure, many aspects of its formation in nature remain elusive. Using a combination of genetic inactivation and metabolic analysis, we investigated the tailoring processes of pactamycin biosynthesis in Streptomyces pactum. The results provide insights into the sequence of events during the tailoring steps of pactamycin biosynthesis and explain the unusual production of various pactamycin analogues by S. pactum mutants. We also identified two new pactamycin analogues that have better selectivity indexes than pactamycin against malarial parasites.     

On the sequence of bond formation in loline alkaloid biosynthesis

Loline alkaloids are saturated pyrrolizidines with an oxygen bridge between carbon atoms C‐2 and C‐7 and an amino group on C‐1. They are bioprotective alkaloids produced by Epichloë and Neotyphodium species, mutualistic fungal endophytes that are symbiotic with cool‐season grasses. The sequence of bond formation in loline alkaloid biosynthesis was determined by synthesizing deuterated forms of potential intermediates and feeding them to cultures of the endophyte Neotyphodium uncinatum. These cultures incorporated deuterium from labeled N‐(3‐amino‐3‐carboxypropyl)proline and exo‐1‐aminopyrrolizidine into N‐formylloline. The first result suggests that N‐(3‐amino‐3‐carboxypropyl)proline is the first committed intermediate in loline biosynthesis, and the second result demonstrates that the pyrrolizidine rings form before the ether bridge. The incorporation of these two compounds into lolines and the lack of incorporation of several related compounds clarify the order of bond formation in loline alkaloid biosynthesis.     

The Role of a Nonribosomal Peptide Synthetase in l‐Lysine Lactamization During Capuramycin Biosynthesis

Capuramycins are one of several known classes of natural products that contain an l ‐Lys‐derived l ‐α‐amino‐?‐caprolactam (l ‐ACL) unit. The α‐amino group of l ‐ACL in a capuramycin is linked to an unsaturated hexuronic acid component through an amide bond that was previously shown to originate by an ATP‐independent enzymatic route. With the aid of a combined in vivo and in vitro approach, a predicted tridomain nonribosomal peptide synthetase CapU is functionally characterized here as the ATP‐dependent amide‐bond‐forming catalyst responsible for the biosynthesis of the remaining amide bond present in l ‐ACL. The results are consistent with the adenylation domain of CapU as the essential catalytic component for l ‐Lys activation and thioesterification of the adjacent thiolation domain. However, in contrast to expectations, lactamization does not require any additional domains or proteins and is likely a nonenzymatic event. The results set the stage for examining whether a similar NRPS‐mediated mechanism is employed in the biosynthesis of other l ‐ACL‐containing natural products and, just as intriguingly, how spontaneous lactamization is avoided in the numerous NRPS‐derived peptides that contain an unmodified l ‐Lys residue.     

Exploring recombinant flavonoid biosynthesis in metabolically engineered Escherichia coli

Flavonoids are important plant-specific secondary metabolites synthesized from 4-coumaroyl coenzyme A (CoA), derived from the general phenylpropanoid pathway, and three malonyl-CoAs. The synthesis involves a plant type III polyketide synthase, chalcone synthase. We report the cloning and coexpression in Escherichia coli of phenylalanine ammonia lyase, cinnamate-4-hydroxylase, 4-coumarate:CoA ligase, and chalcone synthase from the model plant Arabidopsis thaliana. Simultaneous expression of all four genes resulted in a blockage after the first enzymatic step caused by the presence of nonfunctional cinnamate-4-hydroxylase. To overcome this problem we fed exogenous 4-coumaric acid to induced cultures. We observed high-level production of the flavanone naringenin as a result. We were also able to produce phloretin by feeding cultures with 3-(4-hydroxyphenyl)propionic acid. Feeding with ferulic or caffeic acid did not yield the corresponding flavanones. We have also cloned and partially characterized a new tyrosine ammonia lyase from Rhodobacter sphaeroides. Tyrosine ammonia lyase was substituted for phenylalanine ammonia lyase and cinnamate-4-hydroxylase in our E. coli clones and three different growth media were tested. After 48 h induction, high-level production (20.8 mg L(-1)) of naringenin in metabolically engineered E. coli was observed for the first time.