Engineered Biosynthesis of a Novel Amidated Polyketide, Using the Malonamyl-Specific Initiation Module from the Oxytetracycline Polyketide Synthase

Tetracyclines are aromatic polyketides biosynthesized by bacterial type II polyketide synthases (PKSs). Understanding the biochemistry of tetracycline PKSs is an important step toward the rational and combinatorial manipulation of tetracycline biosynthesis. To this end, we have sequenced the gene cluster of oxytetracycline (oxy and otc genes) PKS genes from Streptomyces rimosus. Sequence analysis revealed a total of 21 genes between the otrA and otrB resistance genes. We hypothesized that an amidotransferase, OxyD, synthesizes the malonamate starter unit that is a universal building block for tetracycline compounds. In vivo reconstitution using strain CH999 revealed that the minimal PKS and OxyD are necessary and sufficient for the biosynthesis of amidated polyketides. A novel alkaloid (WJ35, or compound 2) was synthesized as the major product when the oxy-encoded minimal PKS, the C-9 ketoreductase (OxyJ), and OxyD were coexpressed in CH999. WJ35 is an isoquinolone compound derived from an amidated decaketide backbone and cyclized with novel regioselectivity. The expression of OxyD with a heterologous minimal PKS did not afford similarly amidated polyketides, suggesting that the oxy-encoded minimal PKS possesses novel starter unit specificity.     

Engineered biosynthesis of regioselectively modified aromatic polyketides using bimodular polyketide synthases

Bacterial aromatic polyketides such as tetracycline and doxorubicin are a medicinally important class of natural products produced as secondary metabolites by actinomyces bacteria. Their backbones are derived from malonyl-CoA units by polyketide synthases (PKSs). The nascent polyketide chain is synthesized by the minimal PKS, a module consisting of four dissociated enzymes. Although the biosynthesis of most aromatic polyketide backbones is initiated through decarboxylation of a malonyl building block (which results in an acetate group), some polyketides, such as the estrogen receptor antagonist R1128, are derived from nonacetate primers. Understanding the mechanism of nonacetate priming can lead to biosynthesis of novel polyketides that have improved pharmacological properties. Recent biochemical analysis has shown that nonacetate priming is the result of stepwise activity of two dissociated PKS modules with orthogonal molecular recognition features. In these PKSs, an initiation module that synthesizes a starter unit is present in addition to the minimal PKS module. Here we describe a general method for the engineered biosynthesis of regioselectively modified aromatic polyketides. When coexpressed with the R1128 initiation module, the actinorhodin minimal PKS produced novel hexaketides with propionyl and isobutyryl primer units. Analogous octaketides could be synthesized by combining the tetracenomycin minimal PKS with the R1128 initiation module. Tailoring enzymes such as ketoreductases and cyclases were able to process the unnatural polyketides efficiently. Based upon these findings, hybrid PKSs were engineered to synthesize new anthraquinone antibiotics with predictable functional group modifications. Our results demonstrate that (i) bimodular aromatic PKSs present a general mechanism for priming aromatic polyketide backbones with nonacetate precursors; (ii) the minimal PKS controls polyketide chain length by counting the number of atoms incorporated into the backbone rather than the number of elongation cycles; and (iii) in contrast, auxiliary PKS enzymes such as ketoreductases, aromatases, and cyclases recognize specific functional groups in the backbone rather than overall chain length. Among the anthracyclines engineered in this study were compounds with (i) more superior activity than R1128 against the breast cancer cell line MCF-7 and (ii) inhibitory activity against glucose-6-phosphate translocase, an attractive target for the treatment of Type II diabetes.     

Ketosynthases in the initiation and elongation modules of aromatic polyketide synthases have orthogonal acyl carrier protein specificity

Many bacterial aromatic polyketides are synthesized by type II polyketide synthases (PKSs) which minimally consist of a ketosynthase-chain length factor (KS-CLF) heterodimer, an acyl carrier protein (ACP), and a malonyl-CoA:ACP transacylase (MAT). This minimal PKS initiates polyketide biosynthesis by decarboxylation of malonyl-ACP, which is catalyzed by the KS-CLF complex and leads to incorporation of an acetate starter unit. In non-acetate-primed PKSs, such as the frenolicin (fren) PKS and the R1128 PKS, decarboxylative priming is suppressed in favor of chain initiation with alternative acyl groups. Elucidation of these unusual priming pathways could lead to the engineered biosynthesis of polyketides containing novel starter units. Unique to some non-acetate-primed PKSs is a second catalytic module comprised of a dedicated homodimeric KS, an additional ACP, and a MAT. This initiation module is responsible for starter-unit selection and catalysis of the first chain elongation step. To elucidate the protein-protein recognition features of this dissociated multimodular PKS system, we expressed and purified two priming and two elongation KSs, a set of six ACPs from diverse sources, and a MAT. In the presence of the MAT, each ACP was labeled with malonyl-CoA rapidly. In the presence of a KS-CLF and MAT, all ACPs from minimal PKSs supported polyketide synthesis at comparable rates (k(cat) between 0.17 and 0.37 min(-1)), whereas PKS activity was attenuated by at least 50-fold in the presence of an ACP from an initiation module. In contrast, the opposite specificity pattern was observed with priming KSs: while ACPs from initiation modules were good substrates, ACPs from minimal PKSs were significantly poorer substrates. Our results show that KS-CLF and KSIII recognize orthogonal sets of ACPs, and the additional ACP is indispensable for the incorporation of non-acetate primer units. Sequence alignments of the two classes of ACPs identified a tyrosine residue that is unique to priming ACPs. Site-directed mutagenesis of this amino acid in the initiation and elongation module ACPs of the R1128 PKS confirmed the importance of this residue in modulating interactions between KSs and ACPs. Our study provides new biochemical insights into unusual chain initiation mechanisms of bacterial aromatic PKSs.     

Investigation of early tailoring reactions in the oxytetracycline biosynthetic pathway

Tetracyclines are aromatic polyketides biosynthesized by bacterial type II polyketide synthases. The amidated tetracycline backbone is biosynthesized by the minimal polyketide synthases and an amidotransferase homologue OxyD. Biosynthesis of the key intermediate 6-methylpretetramid requires two early tailoring steps, which are cyclization of the linearly fused tetracyclic scaffold and regioselective C-methylation of the aglycon. Using a heterologous host (CH999)/vector pair, we identified the minimum set of enzymes from the oxytetracycline biosynthetic pathway that is required to afford 6-methylpretetramid in vivo. Only two cyclases (OxyK and OxyN) are necessary to completely cyclize and aromatize the amidated tetracyclic aglycon. Formation of the last ring via C-1/C-18 aldol condensation does not require a dedicated fourth-ring cyclase, in contrast to the biosynthetic mechanism of other tetracyclic aromatic polyketides, such as daunorubicin and tetracenomycin. Acetyl-derived polyketides do not undergo spontaneous fourth-ring cyclization and form only anthracene carboxylic acids as demonstrated both in vivo and in vitro. OxyF was identified to be the C-6 C-methyltransferase that regioselectively methylates pretetramid to yield 6-methylpretetramid. Reconstitution of 6-methylpretetramid in a heterologous host sets the stage for a more systematic investigation of additional tetracycline downstream tailoring enzymes and is a key step toward the engineered biosynthesis of tetracycline analogs.     

The acyltransferase homologue from the initiation module of the R1128 polyketide synthase is an acyl-ACP thioesterase that edits acetyl primer units

Type II polyketide synthases (PKSs) synthesize polyfunctional aromatic polyketides through iterative condensations of malonyl extender units. The biosynthesis of most aromatic polyketides is initiated through an acetate unit derived from decarboxylation of malonyl-acyl carrier protein (ACP). Modification of this primer unit represents a powerful method of generating novel polyketides. We have demonstrated that recombination of the initiation module from the R1128 PKS with heterologous elongation modules afforded regioselectively modified polyketides containing alternative primer units. With the exception of the role of the acyltransferase homologue ZhuC, the catalytic cycle of the initiation module has been well explored. ZhuC, along with the ketosynthase III homologue ZhuH and the ACP(p) ZhuG, is essential for the in vivo biosynthesis of aromatic polyketides derived from non-acetate primer units. Here we have studied the role of ZhuC using PKS proteins reconstituted in vitro. We show that the tetracenomycin (tcm) minimal PKS can be directly primed with non-acetate acyl groups. In the presence of approximately 10 microM hexanoyl-ZhuG or approximately 100 microM hexanoyl-CoA, the tcm minimal PKS synthesized hexanoyl-primed analogues of octaketides SEK4 and SEK4b, as well as acetate-primed decaketides SEK15 and SEK15b at comparable levels. Addition of ZhuC abolished synthesis of the acetate-primed decaketides, resulting in exclusive synthesis of the hexanoyl-primed octaketides. In the absence of alternative acyl donors, ZhuC severely retarded the activity of the tcm minimal PKS. The editing capabilities of ZhuC were directly revealed by demonstrating that ZhuC has 100 times greater specificity for acetyl- and propionyl-ACP as compared to hexanoyl- and octanoyl-ACP. Thus, by purging the acetate primer units that otherwise dominate polyketide chain initiation, ZhuC (and presumably its homologues in other PKSs such as the doxorubicin and frenolicin PKSs) allows alternative primer units to be utilized by the elongation module in vivo. The abilities of other alkylacyl primer units to prime the tcm minimal PKS were also investigated in this report.     

Two polyketide synthase-encoding genes are required for biosynthesis of the polyketide virulence factor, T-toxin, by Cochliobolus heterostrophus

Cochliobolus heterostrophus race T, causal agent of southern corn leaf blight, requires T-toxin (a family of C35 to C49 polyketides) for high virulence on T-cytoplasm maize. Production of T-toxin is controlled by two unlinked loci, Tox1A and Tox1B, carried on 1.2 Mb of DNA not found in race O, a mildly virulent form of the fungus that does not produce T-toxin, or in any other Cochliobolus spp. or closely related fungus. PKS1, a polyketide synthase (PKS)-encoding gene at Tox1A, and DEC1, a decarboxylase-encoding gene at Tox1B, are necessary for T-toxin production. Although there is evidence that additional genes are required for T-toxin production, efforts to clone them have been frustrated because the genes are located in highly repeated, A+T-rich DNA. To overcome this difficulty, ligation specificity-based expression analysis display (LEAD), a comparative amplified fragment length polymorphism/gel fractionation/capillary sequencing procedure, was applied to cDNAs from a near-isogenic pair of race T (Tox1+) and race O (Tox1-) strains. This led to discovery of PKS2, a second PKS-encoding gene that maps at Tox1A and is required for both T-toxin biosynthesis and high virulence to maize. Thus, the carbon chain of each T-toxin family member likely is assembled by action of two PKSs, which produce two polyketides, one of which may act as the starter unit for biosynthesis of the mature T-toxin molecule.     

Characterization of two novel non-reducing polyketide synthase genes from the lichen-forming fungus Hypogymnia physodes

Lichens are known to produce a variety of secondary metabolites including polyketides, which have valuable biological activities. Some polyketides are produced solely by lichens. The biosynthesis of these compounds is primarily governed by iterative type I polyketide synthases. Hypogymnia physodes synthesize polyketides such as physodic, physodalic and hydroxyphysodic acid and atranorin, which are non-reducing polyketides. Two novel non-reducing polyketide synthase (PKS) genes were isolated from a fosmid genomic library of a mycobiont of H. physodes using a 409bp fragment corresponding to part of the reductase (R) domain as a probe. H. physodes PKS1 (Hyopks1) and PKS2 (Hypopks2) contain keto synthase (KS), acyl transferase (AT), acyl carrier protein (ACP), methyl transferase (ME) and R domains. Classification based on phylogeny analysis using the translated KS and AT domains demonstrated that Hypopks1 and Hypopks2 are members of the fungal non-reducing PKSs clade III. This is the first report of non-reducing PKSs containing the R domain-mediated release mechanisms in lichens, which are also rare fungal type I PKS in non-lichenized filamentous fungi.     

The Streptomyces peucetius dpsC Gene Determines the Choice of Starter Unit in Biosynthesis of the Daunorubicin Polyketide

The starter unit used in the biosynthesis of daunorubicin is propionyl coenzyme A (CoA) rather than acetyl-CoA, which is used in the production of most of the bacterial aromatic polyketides studied to date. In the daunorubicin biosynthesis gene cluster of Streptomyces peucetius, directly downstream of the genes encoding the beta-ketoacyl:acyl carrier protein synthase subunits, are two genes, dpsC and dpsD, encoding proteins that are believed to function as the starter unit-specifying enzymes. Recombinant strains containing plasmids carrying dpsC and dpsD, in addition to other daunorubicin polyketide synthase (PKS) genes, incorporate the correct starter unit into polyketides made by these genes, suggesting that, contrary to earlier reports, the enzymes encoded by dpsC and dpsD play a crucial role in starter unit specification. Additionally, the results of a cell-free synthesis of 21-carbon polyketides from propionyl-CoA and malonyl-CoA that used the protein extracts of recombinant strains carrying other daunorubicin PKS genes to which purified DpsC was added suggest that this enzyme has the primary role in starter unit discrimination for daunorubicin biosynthesis.     

The first plant type III polyketide synthase that catalyzes formation of aromatic heptaketide

A cDNA encoding a novel plant type III polyketide synthase (PKS) was cloned from rhubarb (Rheum palmatum). A recombinant enzyme expressed in Escherichia coli accepted acetyl-CoA as a starter, carried out six successive condensations with malonyl-CoA and subsequent cyclization to yield an aromatic heptaketide, aloesone. The enzyme shares 60% amino acid sequence identity with chalcone synthases (CHSs), and maintains almost identical CoA binding site and catalytic residues conserved in the CHS superfamily enzymes. Further, homology modeling predicted that the 43-kDa protein has the same overall fold as CHS. This provides new insights into the catalytic functions of type III PKSs, and suggests further involvement in the biosynthesis of plant polyketides.     

Approaches to stabilization of inter-domain recombination in polyketide synthase gene expression plasmids

Regions of extremely high sequence identity are recurrent in modular polyketide synthase (PKS) genes. Such sequences are potentially detrimental to the stability of PKS expression plasmids used in the combinatorial biosynthesis of polyketide metabolites. We present two different solutions for circumventing intra-plasmid recombination within the megalomicin PKS genes in Streptomyces coelicolor. In one example, a synthetic gene was used in which the codon usage was reengineered without affecting the primary amino acid sequence. The other approach utilized a heterologous subunit complementation strategy to replace one of the problematic regions. Both methods resulted in PKS complexes capable of 6-deoxyerythronolide B analogue biosynthesis in S. coelicolor CH999, permitting reproducible scale-up to at least 5-l stirred-tank fermentation and a comparison of diketide precursor incorporation efficiencies between the erythromycin and megalomicin PKSs. Electronic Publication     

Targeting modular polyketide synthases with iteratively acting acyltransferases from metagenomes of uncultured bacterial consortia

Bacterial type I polyketide synthases (PKSs) produce a wide range of biomedically important secondary metabolites. These enzymes possess a modular structure that can be genetically re-engineered to yield novel drug candidates not found in nature. Recently, we have reported the putative pederin PKS from an uncultured bacterial symbiont of Paederus fuscipes beetles. It belongs to an architecturally unusual PKS group, the members of which contain iteratively acting acyltransferases that are not integrated into the PKS modules but are encoded by isolated genes. As these systems are rare, often contain additional unusual features and are of smaller size than regular PKSs, the development of a method for the targeted isolation of new group members would be of great interest. Here, we present a phylogenetic approach to identify these systems rapidly in highly complex metagenomic DNA samples. To demonstrate its practical value, we located two pederin-type PKS systems putatively involved in the biosynthesis of antitumour polyketides in the metagenomic DNA of beetles, sponges and their uncultivated bacterial symbionts.     

Computational approach for prediction of domain organization and substrate specificity of modular polyketide synthases

Modular polyketide synthases (PKSs) are large multi-enzymatic, multi-domain megasynthases, which are involved in the biosynthesis of a class of pharmaceutically important natural products, namely polyketides. These enzymes harbor a set of repetitive active sites termed modules and the domains present in each module dictate the chemical moiety that would add to a growing polyketide chain. This modular logic of biosynthesis has been exploited with reasonable success to produce several novel compounds by genetic manipulation. However, for harnessing their vast potential of combinatorial biosynthesis, it is essential to develop knowledge based in silico approaches for correlating the sequence and domain organization of PKSs to their polyketide products. In this work, we have carried out extensive sequence analysis of experimentally characterized PKS clusters to develop an automated computational protocol for unambiguous identification of various PKS domains in a polypeptide sequence. A structure based approach has been used to identify the putative active site residues of acyltransferase (AT) domains, which control the specificities for various starter and extender units during polyketide biosynthesis. On the basis of the analysis of the active site residues and molecular modelling of substrates in the active site of representative AT domains, we have identified a crucial residue that is likely to play a major role in discriminating between malonate and methylmalonate during selection of extender groups by this domain. Structural modelling has also explained the experimentally observed chiral preference of AT domain in substrate selection. This computational protocol has been used to predict the domain organization and substrate specificity for PKS clusters from various microbial genomes. The results of our analysis as well as the computational tools for prediction of domain organization and substrate specificity have been organized in the form of a searchable computerized database (PKSDB). PKSDB would serve as a valuable tool for identification of polyketide products biosynthesized by uncharacterized PKS clusters. This database can also provide guidelines for rational design of experiments to engineer novel polyketides.     

Exploiting marine actinomycete biosynthetic pathways for drug discovery

Drug discovery relies on the generation of large numbers of structurally diverse compounds from which a potential candidate can be identified. To this end, actinomycetes have often been exploited because of their ability to biosynthesize an impressive array of novel metabolites particularly polyketides. The genetic organization of polyketide synthases (PKSs) makes them readily amenable to manipulation, and thus re-engineering artificial or hybrid PKSs to produce unnatural natural products is a reality. This review highlights two approaches we have used to generate novel polyketides by manipulating genes responsible for starter unit biosynthesis in the Streptomyces maritimus enterocin type II PKS. Our preliminary investigation into the biosynthesis of neomarinone, a rare marine actinomycete-derived meroterpenoid, is also presented.     

Metagenomic analysis reveals diverse polyketide synthase gene clusters in microorganisms associated with the marine sponge Discodermia dissoluta

Sponge-associated bacteria are thought to produce many novel bioactive compounds, including polyketides. PCR amplification of ketosynthase domains of type I modular polyketide synthases (PKS) from the microbial community of the marine sponge Discodermia dissoluta revealed great diversity and a novel group of sponge-specific PKS ketosynthase domains. Metagenomic libraries totaling more than four gigabases of bacterial genomes associated with this sponge were screened for type I modular PKS gene clusters. More than 90% of the clones in total sponge DNA libraries represented bacterial DNA inserts, and 0.7% harbored PKS genes. The majority of the PKS hybridizing clones carried small PKS clusters of one to three modules, although some clones encoded large multimodular PKSs (more than five modules). The most abundant large modular PKS appeared to be encoded by a bacterial symbiont that made up < 1% of the sponge community. Sequencing of this PKS revealed 14 modules that, if expressed and active, is predicted to produce a multimethyl-branched fatty acid reminiscent of mycobacterial lipid components. Metagenomic libraries made from fractions enriched for unicellular or filamentous bacteria differed significantly, with the latter containing numerous nonribosomal peptide synthetase (NRPS) and mixed NRPS-PKS gene clusters. The filamentous bacterial community of D. dissoluta consists mainly of Entotheonella spp., an unculturable sponge-specific taxon previously implicated in the biosynthesis of bioactive peptides.     

Engineered polyketide biosynthesis and biocatalysis in Escherichia coli

Polyketides are important bioactive natural products biosynthesized by bacteria, fungi, and plants. The enzymes that synthesize polyketides are collectively referred to as polyketide synthases (PKSs). Because many of the natural hosts that produce polyketides are difficult to culture or manipulate, establishing a universal heterologous host that is genetically tractable has become an important goal toward the engineered biosynthesis of polyketides and analogues. Here, we summarize the recent progresses in engineering Escherichia coli as a heterologous host for reconstituting PKSs of different types. Our increased understanding of PKS enzymology and structural biology, combined with new tools in protein engineering, metabolic engineering, and synthetic biology, has firmly established E. coli as a powerful host for producing polyketides.     

Non-modular polyketide synthases in myxobacteria

Myxobacteria are prolific producers of a wide variety of secondary metabolites. The vast majority of these compounds are complex polyketides which are biosynthesised by multimodular polyketide synthases (PKSs). In contrast, few myxobacterial metabolites isolated to date are derived from non-modular PKSs, in particular type III PKSs. This review reports our progress on the characterisation of type III PKSs in myxobacteria. We also summarize current knowledge on bacterial type III PKSs, with a special focus on the evolutionary relationship between plant and bacterial enzymes. The biosynthesis of a quinoline alkaloid in Stigmatella aurantiaca by a non-modular PKS is also discussed.     

Cloning and characterization of Streptomyces galilaeus aclacinomycins polyketide synthase (PKS) cluster

We have cloned and sequenced polyketide synthase (PKS) genes from the aclacinomycin producer Streptomyces galilaeus ATCC 31,615. The sequenced 13.5-kb region contained 13 complete genes. Their organization as well as their protein sequences showed high similarity to those of other type II PKS genes. The continuous region included the genes for the minimal PKS, consisting of ketosynthase I (aknB), ketosynthase II (aknC), and acyl carrier protein (aknD). These were followed by the daunomycin dpsC and dpsD homologues (aknE2 and F, respectively), which are rare in type II PKS clusters. They are associated with the unusual starter unit, propionate, used in the biosynthesis of aklavinone, a common precursor of aclacinomycin and daunomycin. Accordingly, when aclacinomycins minimal PKS genes were substituted for those of nogalamycin in the plasmid carrying genes for auramycinone biosynthesis, aklavinone was produced in the heterologous hosts. In addition to the minimal PKS, the cloned region included the PKS genes for polyketide ketoreductase (aknA), aromatase (aknE1) and oxygenase (aknX), as well as genes putatively encoding an aklanonic acid methyl transferase (aknG) and an aklanonic acid methyl ester cyclase (aknH) for post-polyketide steps were found. Moreover, the region carried genes for an activator (aknI), a glycosyl transferase (aknK) and an epimerase (aknL) taking part in deoxysugar biosynthesis.     

Chimeric 6-methylsalicylic acid synthase with domains of acyl carrier protein and methyltransferase from Pseudallescheria boydii shows novel biosynthetic activity

Polyketides are important secondary metabolites, many of which exhibit potent pharmacological applications. Biosynthesis of polyketides is carried out by a single polyketide synthase (PKS) or multiple PKSs in successive elongations of enzyme-bound intermediates related to fatty acid biosynthesis. The polyketide gene PKS306 from Pseudallescheria boydii NTOU2362 containing domains of ketosynthase (KS), acyltransferase (AT), dehydratase (DH), acyl carrier protein (ACP) and methyltransferase (MT) was cloned in an attempt to produce novel chemical compounds, and this PKS harbouring green fluorescent protein (GFP) was expressed in Saccharomyces cerevisiae. Although fluorescence of GFP and fusion protein analysed by anti-GFP antibody were observed, no novel compound was detected. 6-methylsalicylic acid synthase (6MSAS) was then used as a template and engineered with PKS306 by combinatorial fusion. The chimeric PKS containing domains of KS, AT, DH and ketoreductase (KR) from 6MSAS with ACP and MT from PKS306 demonstrated biosynthesis of a novel compound. The compound was identified with a deduced chemical formula of C₇H₁₀O₃, and the chemical structure was named as 2-hydroxy-2-(propan-2-yl) cyclobutane-1,3-dione. The novel compound synthesized by the chimeric PKS in this study demonstrates the feasibility of combinatorial fusion of PKS genes to produce novel polyketides.     

Synthetic biology of polyketide synthases

Complex reduced polyketides represent the largest class of natural products that have applications in medicine, agriculture, and animal health. This structurally diverse class of compounds shares a common methodology of biosynthesis employing modular enzyme systems called polyketide synthases (PKSs). The modules are composed of enzymatic domains that share sequence and functional similarity across all known PKSs. We have used the nomenclature of synthetic biology to classify the enzymatic domains and modules as parts and devices, respectively, and have generated detailed lists of both. In addition, we describe the chassis (hosts) that are used to assemble, express, and engineer the parts and devices to produce polyketides. We describe a recently developed software tool to design PKS system and provide an example of its use. Finally, we provide perspectives of what needs to be accomplished to fully realize the potential that synthetic biology approaches bring to this class of molecules.