Identification and Characterization of a New Type III Polyketide Synthase from a Marine Yeast, Naganishia uzbekistanensis.

A putative Type III Polyketide synthase (PKSIII) encoding gene was identified from a marine yeast, Naganishia uzbekistanensis strain Mo29 (UBOCC-A-208024) (formerly named as Cryptococcus sp.) isolated from deep-sea hydrothermal vents. This gene is part of a distinct phylogenetic branch compared to all known terrestrial fungal sequences. This new gene encodes a C-terminus extension of 74 amino acids compared to other known PKSIII proteins like Neurospora crassa. Full-length and reduced versions of this PKSIII were successfully cloned and overexpressed in a bacterial host, Escherichia coli BL21 (DE3). Both proteins showed the same activity, suggesting that additional amino acid residues at the C-terminus are probably not required for biochemical functions. We demonstrated by LC-ESI-MS/MS that these two recombinant PKSIII proteins could only produce tri- and tetraketide pyrones and alkylresorcinols using only long fatty acid chain from C8 to C16 acyl-CoAs as starter units, in presence of malonyl-CoA. In addition, we showed that some of these molecules exhibit cytotoxic activities against several cancer cell lines.

Widening the bottleneck: Heterologous expression, purification, and characterization of the Ktedonobacter racemifer minimal type II polyketide synthase in Escherichia coli.

Enzyme assemblies such as type II polyketide synthases (PKSs) produce a wide array of bioactive secondary metabolites. While the molecules produced by type II PKSs have found remarkable clinical success, the biosynthetic prowess of these enzymes has been stymied by 1) the inability to reconstitute the bioactivity of the minimal PKS enzymes in vitro and 2) limited exploration of type II PKSs from diverse phyla. To begin filling this unmet need, we expressed, purified, and characterized the ketosynthase chain length factor (KS-CLF) and acyl carrier protein (ACP) from Ktedonobacter racemifer (Kr). Using E. coli as a heterologous host, we obtained soluble proteins in titers signifying improvements over previous KS-CLF heterologous expression efforts. Characterization of these enzymes reveals that KrACP has self-malonylating activity. Sedimentation velocity analytical ultracentrifugation (SV-AUC) analysis of holo-KrACP and KrKS-CLF indicates that these enzymes do not interact in vitro, suggesting that the acylated state of these proteins might play an important role in facilitating biosynthetically relevant interactions. These results lay important groundwork for optimizing the interaction between KrKS-CLF and KrACP and exploring the biosynthetic potential of other non-actinomycete type II PKSs.

Cloning, expression, and purification of intact polyketide synthase modules.

Polyketides are a structurally and functionally diverse family of bioactive natural products that have proven to be a rich source of pharmaceutical and agrochemical lead compounds. Many polyketides are biosynthesized by large multifunctional megaenzymes termed type I modular polyketide synthases (PKSs). These systems possess a distinctive assembly line-like architecture, comprising a series of linearly arranged, multidomain extension modules, housed in sequence within giant polypeptide chains. Due to their inherently modular structures, PKSs represent attractive targets for reengineering, enabling access to functionally optimized "nonnatural" natural products. In this chapter we describe methods for the molecular cloning, recombinant over-expression, and purification of intact PKS modules and multimodular PKS polypeptides. The usefulness of these methods is demonstrated by applying them to the study of the abyssomicin C PKS, a >1MDa multimodular synthase responsible for the biosynthesis of a polyketide antimicrobial lead compound.

PKS-NRPS Enzymology and Structural Biology: Considerations in Protein Production.

The structural diversity and complexity of marine natural products have made them a rich and productive source of new bioactive molecules for drug development. The identification of these new compounds has led to extensive study of the protein constituents of the biosynthetic pathways from the producing microbes. Essential processes in the dissection of biosynthesis have been the elucidation of catalytic functions and the determination of 3D structures for enzymes of the polyketide synthases and nonribosomal peptide synthetases that carry out individual reactions. The size and complexity of these proteins present numerous difficulties in the process of going from gene to structure. Here, we review the problems that may be encountered at the various steps of this process and discuss some of the solutions devised in our and other labs for the cloning, production, purification, and structure solution of complex proteins using Escherichia coli as a heterologous host.

Ralstonins A and B, Lipopeptides with Chlamydospore-Inducing and Phytotoxic Activities from the Plant Pathogen Ralstonia solanacearum.

Ralstonia solanacearum has an orphan hybrid polyketide synthase-nonribosomal peptide synthetase gene cluster. We herein isolate its products (named ralstonins A and B) from R. solanacearum and elucidate their structures and biological activities. Ralstonins are unusual lipodepsipeptides composed of 11 amino acids (containing unique amino acids such as ß-hydroxytyrosine and dehydroalanine) and a 3-amino-2-hydroxyoctadecanoic acid, and their production is controlled by quorum sensing, a mechanism of bacterial cell-cell communication. Ralstonins exhibited chlamydospore-inducing activity and phytotoxicity.

Strain Prioritization and Genome Mining for Enediyne Natural Products.

The enediyne family of natural products has had a profound impact on modern chemistry, biology, and medicine, and yet only 11 enediynes have been structurally characterized to date. Here we report a genome survey of 3,400 actinomycetes, identifying 81 strains that harbor genes encoding the enediyne polyketide synthase cassettes that could be grouped into 28 distinct clades based on phylogenetic analysis. Genome sequencing of 31 representative strains confirmed that each clade harbors a distinct enediyne biosynthetic gene cluster. A genome neighborhood network allows prediction of new structural features and biosynthetic insights that could be exploited for enediyne discovery. We confirmed one clade as new C-1027 producers, with a significantly higher C-1027 titer than the original producer, and discovered a new family of enediyne natural products, the tiancimycins (TNMs), that exhibit potent cytotoxicity against a broad spectrum of cancer cell lines. Our results demonstrate the feasibility of rapid discovery of new enediynes from a large strain collection. IMPORTANCE: Recent advances in microbial genomics clearly revealed that the biosynthetic potential of soil actinomycetes to produce enediynes is underappreciated. A great challenge is to develop innovative methods to discover new enediynes and produce them in sufficient quantities for chemical, biological, and clinical investigations. This work demonstrated the feasibility of rapid discovery of new enediynes from a large strain collection. The new C-1027 producers, with a significantly higher C-1027 titer than the original producer, will impact the practical supply of this important drug lead. The TNMs, with their extremely potent cytotoxicity against various cancer cells and their rapid and complete cancer cell killing characteristics, in comparison with the payloads used in FDA-approved antibody-drug conjugates (ADCs), are poised to be exploited as payload candidates for the next generation of anticancer ADCs. Follow-up studies on the other identified hits promise the discovery of new enediynes, radically expanding the chemical space for the enediyne family.

Tafuketide, a phylogeny-guided discovery of a new polyketide from Talaromyces funiculosus Salicorn 58.

A phylogeny-guided approach was applied to screen endophytic fungi containing type I polyketide synthase (PKS I) biosynthetic gene sequences and aimed to correlate genotype to chemotype for the discovery of novel bioactive polyketides. Salicorn 58, which was identified as Talaromyces funiculosus based on its internal transcribed spacer (ITS) and ribosomal large-subunit (LSU) DNA sequences, showed significant target bands. A chemical investigation of the culture of Salicorn 58 was allowed for the isolation of a new polyketide, Talafun (1), and a new natural product, N-(2'-hydroxy-3'-octadecenoyl)-9-methyl-4,8-sphingadienin (2), together with six known compounds, including chrodrimanin A (3), chrodrimanin B (4), N-(4-hydroxy-2-methoxyphenyl) acetamide (5), butyl ß-glucose (6), 3ß,15ß-dihydroxyl-(22E, 24R)-ergosta-5,8(14),22-trien-7-dione (7), and (3ß,5a,8a,22E)-5,8-epidioxyergosta-6,22-dien-3-ol (8). Their chemical structures were elucidated by extensive spectroscopic analysis and electro circular dichroism (ECD) spectrum calculations. Antioxidant experiments revealed that compound 5 showed strong ABTS(+) radical scavenging activity with an IC50 value of 11.43 ± 1.61 µM and potent ferric reducing activity (FRAP assay) with FRAP value of 187.52 ± 2.97. Antimicrobial assays revealed that compounds 1 and 4 showed high levels of selectivity toward Escherichia coli with MIC values of 18 ± 0.40 and 43 ± 0.52 µM, respectively. Compounds 2 and 3 exhibited broad-spectrum antimicrobial activity against Staphylococcus aureus, Mycobacterium smegmatis, Micrococcus tetragenus, Mycobacterium phlei, and E. coli, respectively. The results from the current research highlight the advantage of phylogeny-guided pipeline for the screening of new polyketides from endophytic fungi containing PKS I genes.

A highly unusual polyketide synthase directs dawenol polyene biosynthesis in Stigmatella aurantiaca.

Enormous progress in the field of polyketide biosynthesis has led to the establishment of rules for general text book biosynthetic logic and consequently to the assumption that biosynthetic genes can be easily correlated with the corresponding natural products. However, non-textbook examples of polyketide assembly continue to be discovered suggesting the gene to product and product to gene predictions need improvement, especially as they are increasingly used in the post-genomic era. Here, we analyzed the genomic blueprint of a myxobacterial multi-producer of secondary metabolites, Stigmatella aurantiaca DW4/3-1, for its biosynthetic potential by genome-mining. In addition to the five polyketide synthase and/or nonribosomal peptide synthetase gene clusters of known function we identified a further 13 genomic regions exemplifying the enormous genetic potential for the production of additional chemical diversity by this strain. We show by gene inactivation and heterologous expression of the newly identified biosynthetic pathway for dawenol that the biosynthesis of this known polyene does not follow text book biosynthetic logic. Intriguingly, a genomic locus encoding an unusual polyketide synthase exhibiting similarity to gene loci involved in the formation of polyunsaturated fatty acids and secondary lipids was identified.

In vitro reconstitution and analysis of the 6-deoxyerythronolide B synthase.

Notwithstanding an extensive literature on assembly line polyketide synthases such as the 6-deoxyerythronolide B synthase (DEBS), a complete naturally occurring synthase has never been reconstituted in vitro from purified protein components. Here, we describe the fully reconstituted DEBS and quantitatively characterize some of the properties of the assembled system that have never been explored previously. The maximum turnover rate of the complete hexamodular system is 1.1 min(-1), comparable to the turnover rate of a truncated trimodular derivative (2.5 min(-1)) but slower than that of a bimodular derivative (21 min(-1)). In the presence of similar concentrations of methylmalonyl- and ethylmalonyl-CoA substrates, DEBS synthesizes multiple regiospecifically modified analogues, one of which we have analyzed in detail. Our studies lay the foundation for biochemically interrogating and rationally engineering polyketide assembly lines in an unprecedented manner.

Genome mining reveals the evolutionary origin and biosynthetic potential of basidiomycete polyketide synthases.

Numerous polyketides are known from bacteria, plants, and fungi. However, only a few have been isolated from basidiomycetes. Large scale genome sequencing projects now help anticipate the capacity of basidiomycetes to synthesize polyketides. In this study, we identified and annotated 111 type I and three type III polyketide synthase (PKS) genes from 35 sequenced basidiomycete genomes. Phylogenetic analysis of PKS genes suggests that all main types of fungal iterative PKS had already evolved before the Ascomycota and Basidiomycota diverged. A comparison of genomic and metabolomic data shows that the number of polyketide genes exceeds the number of known polyketide structures by far. Exploiting these results to design degenerate PCR primers, we amplified and cloned the complete sequence of armB, a PKS gene from the melleolide producer Armillaria mellea. We expect this study will serve as a guide for future genomic mining projects to discover structurally diverse mushroom-derived polyketides.

Establishing a toolkit for precursor-directed polyketide biosynthesis: exploring substrate promiscuities of acid-CoA ligases.

Polyketides are chemically diverse and medicinally important biochemicals that are biosynthesized from acyl-CoA precursors by polyketide synthases. One of the limitations to combinatorial biosynthesis of polyketides has been the lack of a toolkit that describes the means of delivering novel acyl-CoA precursors necessary for polyketide biosynthesis. Using five acid-CoA ligases obtained from various plants and microorganisms, we biosynthesized an initial library of 79 acyl-CoA thioesters by screening each of the acid-CoA ligases against a library of 123 carboxylic acids. The library of acyl-CoA thioesters includes derivatives of cinnamyl-CoA, 3-phenylpropanoyl-CoA, benzoyl-CoA, phenylacetyl-CoA, malonyl-CoA, saturated and unsaturated aliphatic CoA thioesters, and bicyclic aromatic CoA thioesters. In our search for the biosynthetic routes of novel acyl-CoA precursors, we discovered two previously unreported malonyl-CoA derivatives (3-thiophenemalonyl-CoA and phenylmalonyl-CoA) that cannot be produced by canonical malonyl-CoA synthetases. This report highlights the utility and importance of determining substrate promiscuities beyond conventional substrate pools and describes novel enzymatic routes for the establishment of precursor-directed combinatorial polyketide biosynthesis.

Isolation and characterization of culturable endophytic actinobacteria associated with Artemisia annua L.

Endophytic actinobacteria isolated from Artemisia annua were characterized and evaluated for their bioactivities. A total of 228 isolates representing at least 19 different genera of actinobacteria were obtained and several of them seemed to be novel taxa. An evaluation of antimicrobial activity showed that more isolates possessed activity towards plant pathogens than activity against other pathogenic bacteria or yeasts. High frequencies of PCR amplification were obtained for type I polyketide synthases (PKS-I, 21.1%), type II polyketide synthases (PKS-II, 45.2%) and nonribosomal peptide synthetases (NRPS, 32.5%). The results of herbicidal activity screening indicated that 19 out of 117 samples of fermentation broths completely inhibited the germination of Echinochloa crusgalli. This study indicated that endophytic actinobacteria associated with A. annua are abundant and have potentially beneficial and diverse bioactivities which should be pursued for their biotechnical promise.

Analysis of the ketosynthase-chain length factor heterodimer from the fredericamycin polyketide synthase.

The pentadecaketide fredericamycin has the longest carbon chain backbone among polycyclic aromatic polyketide antibiotics whose biosynthetic genes have been sequenced. This backbone is synthesized by the bimodular fdm polyketide synthase (PKS). Here, we demonstrate that the bimodular fdm PKS as well as its elongation module alone synthesize undecaketides and dodecaketides. Thus, unlike other homologs, the fdm ketosynthase-chain length factor (KS-CLF) heterodimer does not exclusively control the backbone length of its natural product. Using sequence- and structure-based approaches, 48 CLF multiple mutants were engineered and analyzed. Unexpectedly, the I134F mutant was unable to turn over but could initiate and partially elongate the polyketide chain. This unprecedented mutant suggests that the KS-CLF heterodimer harbors an as yet uncharacterized chain termination mechanism. Together, our findings reveal fundamental mechanistic differences between the fdm PKS and its well-studied homologs.

Isolation and characterization of a reducing polyketide synthase gene from the lichen-forming fungus Usnea longissima.

The reducing polyketide synthases found in filamentous fungi are involved in the biosynthesis of many drugs and toxins. Lichens produce bioactive polyketides, but the roles of reducing polyketide synthases in lichens remain to be clearly elucidated. In this study, a reducing polyketide synthase gene (U1PKS3) was isolated and characterized from a cultured mycobiont of Usnea longissima. Complete sequence information regarding U1PKS3 (6,519 bp) was obtained by screening a fosmid genomic library. A U1PKS3 sequence analysis suggested that it contains features of a reducing fungal type I polyketide synthase with ß-ketoacyl synthase (KS), acyltransferase (AT), dehydratase (DH), enoyl reductase (ER), ketoacyl reducatse (KR), and acyl carrier protein (ACP) domains. This domain structure was similar to the structure of ccRadsl, which is known to be involved in resorcylic acid lactone biosynthesis in Chaetomium chiversii. The results of phylogenetic analysis located U1PKS3 in the clade of reducing polyketide synthases. RT-PCR analysis results demonstrated that UIPKS3 had six intervening introns and that UIPKS3 expression was upregulated by glucose, sorbitol, inositol, and mannitol.

Expression, purification and crystallization of a plant type III polyketide synthase that produces diarylheptanoids.

Curcuminoid synthase (CUS) from Oryza sativa is a plant-specific type III polyketide synthase that catalyzes the one-pot formation of bisdemethoxycurcumin by the condensation of two molecules of 4-coumaroyl-CoA and one molecule of malonyl-CoA. Recombinant CUS was expressed in Escherichia coli and crystallized by the sitting-drop vapour-diffusion method. The crystals belonged to space group P2(1), with unit-cell parameters a=72.7, b=97.2, c=126.2 A, alpha=gamma=90.0, beta=103.7 degrees. A diffraction data set was collected in-house to 2.5 A resolution.

Identification and function of a polyketide synthase gene responsible for 1,8-dihydroxynaphthalene-melanin pigment biosynthesis in Ascochyta rabiei.

Ascochyta rabiei produces and accumulates one of the well-known fungal polyketides, 1,8-dihydroxynaphthalene-melanin pigment (DHN-melanin), in asexual and sexual fruiting bodies. Degenerate PCR primers were used to isolate an ArPKS1 of A. rabiei encoding a polypeptide with high similarity to polyketide synthase (PKS) involved in biosynthesis of DHN-melanin in other ascomycetous fungi. Site-directed mutagenesis of ArPKS1 in A. rabiei generated melanin-deficient pycnidial mutants but caused no significant reduction of pathogenicity to chickpea. Pycnidiospores in ArPKS1-mutant pycnidia showed higher sensitivity to UV light exposure compared to pycnidiospores in melanized pycnidia of the wild-type progenitor isolate. Integration of an orthologous PKS1 gene from Bipolaris oryzae into the genome of the mutants complemented the dysfunctional ArPKS1 gene. This study demonstrated that A. rabiei uses a DHN-melanin pathway for pigmentation of pycnidia and this molecule may protect pycnidiospores from UV irradiation.

Complete reconstitution of a highly reducing iterative polyketide synthase.

Highly reducing iterative polyketide synthases are large, multifunctional enzymes that make important metabolites in fungi, such as lovastatin, a cholesterol-lowering drug from Aspergillus terreus. We report efficient expression of the lovastatin nonaketide synthase (LovB) from an engineered strain of Saccharomyces cerevisiae, as well as complete reconstitution of its catalytic function in the presence and absence of cofactors (the reduced form of nicotinamide adenine dinucleotide phosphate and S-adenosylmethionine) and its partner enzyme, the enoyl reductase LovC. Our results demonstrate that LovB retains correct intermediates until completion of synthesis of dihydromonacolin L, but off-loads incorrectly processed compounds as pyrones or hydrolytic products. Experiments replacing LovC with analogous MlcG from compactin biosynthesis demonstrate a gate-keeping function for this partner enzyme. This study represents a key step in the understanding of the functions and structures of this family of enzymes.

Characterization of olivetol synthase, a polyketide synthase putatively involved in cannabinoid biosynthetic pathway.

Alkylresorcinol moieties of cannabinoids are derived from olivetolic acid (OLA), a polyketide metabolite. However, the polyketide synthase (PKS) responsible for OLA biosynthesis has not been identified. In the present study, a cDNA encoding a novel PKS, olivetol synthase (OLS), was cloned from Cannabis sativa. Recombinant OLS did not produce OLA, but synthesized olivetol, the decarboxylated form of OLA, as the major reaction product. Interestingly, it was also confirmed that the crude enzyme extracts from flowers and rapidly expanding leaves, the cannabinoid-producing tissues of C. sativa, also exhibited olivetol-producing activity, suggesting that the native OLS is functionally expressed in these tissues. The possibility that OLS could be involved in OLA biosynthesis was discussed based on its catalytic properties and expression profile.

A novel type III polyketide synthase encoded by a three-intron gene from Polygonum cuspidatum.

A type III polyketide synthase cDNA and the corresponding gene (PcPKS2) were cloned from Polygonum cuspidatum Sieb. et Zucc. Sequencing results showed that the ORF of PcPKS2 was interrupted by three introns, which was an unexpected finding because all type III PKS genes studied so far contained only one intron at a conserved site in flowering plants, except for an Antirrhinum majus chalcone synthase gene. Besides the unusual gene structure, PcPKS2 showed some interesting characteristics: (1) the CHS "gatekeepers" Phe215 and Phe265 are uniquely replaced by Leu and Cys, respectively; (2) recombinant PcPKS2 overexpressed in Escherichia coli efficiently afforded 4-coumaroyltriacetic acid lactone (CTAL) as a major product along with bis-noryangonin (BNY) and p-hydroxybenzalacetone at low pH; however, it effectively yielded p-hydroxybenzalacetone as a dominant product along with CTAL and BNY at high pH. Beside p-hydroxybenzalacetone, CTAL and BNY, a trace amount of naringenin chalcone could be detected in assays at different pH. Furthermore, 4-coumaroyl-CoA and feruloyl-CoA were the only cinnamoyl-CoA derivatives accepted as starter substrates. PcPKS2 did not accept isobutyryl-CoA, isovaleryl-CoA or acetyl-CoA as substrate. DNA gel blot analysis indicated that there are two to four PcPKS2 copies in the P. cuspidatum genome. RNA gel blot analysis revealed that PcPKS2 is highly expressed in the rhizomes and in young leaves, but not in the roots of the plant. PcPKS2 transcripts in leaves were induced by pathogen infection, but not by wounding.