Strategies for combinatorial biosynthesis with modular polyketide synthases

Polyketides are assembled by the polyketide synthases (PKS) through a common mechanism, the condensation of small carboxylic acids. However, a large structural variety exists within these molecules, paralleled by their different bioactivities. Structural differences in polyketides mostly stem from variations in the number of elongation cycles, in the extender unit incorporated and the extent of processing occurring during each cycle. A significant fraction of polyketides is made in bacteria by modular PKSs, which direct polyketide synthesis on a protein template, where each module is responsible for selecting, incorporating and processing the appropriate carboxylate unit. Since their discovery in the early nineties, the architecture of modular PKSs and their modus operandi have attracted efforts by several laboratories to reprogram PKSs to produce tailor-made polyketides. The availability of a growing number of modular PKSs of defined sequence, and of well-developed model systems for the in vitro and in vivo analysis of these enzymes, has led to the successful production of many novel polyketides after genetic manipulation of the appropriate PKS. We discuss the different strategies that are followed for the construction of functional "hybrid" systems, with particular emphasis on what can be done in terms of generating chemical diversity, highlighting also the limitations of our current understanding. The prospects of generating novel useful polyketides by genetic engineering are also discussed.     

Evidence for a protein-protein interaction motif on an acyl carrier protein domain from a modular polyketide synthase

During biosynthesis on modular polyketide synthases (PKSs), chain extension intermediates are tethered to acyl carrier protein (ACP) domains through phosphopantetheinyl prosthetic groups. Each ACP must therefore interact with every other domain within the module, and also with a downstream acceptor domain. The nature of these interactions is key to our understanding of the topology and operation of these multienzymes. Sequence analysis and homology modeling implicates a potential helical region (helix II) on the ACPs as a protein-protein interaction motif. Using site-directed mutagenesis, we show that residues along this putative helix lie at the interface between the ACP and the phosphopantetheinyl transferase that catalyzes its activation. Our results accord with previous studies of discrete ACP proteins from fatty acid and aromatic polyketide biosynthesis, suggesting that helix II may also serve as a universal interaction motif in modular PKSs.     

Enhancing the modularity of the modular polyketide synthases: transacylation in modular polyketide synthases catalyzed by malonyl-CoA:ACP transacylase

Selective incorporation of extender units in modular polyketide synthases is primarily controlled by acyl transferase (AT) domains. The AT domains catalyze transacylation of the extender unit from acyl-CoA to the phosphopantetheine arm of an acyl carrier protein (ACP) domain in the same module. New methods that can modulate the extender unit specificity of individual modules with minimal structural or kinetic perturbations in the engineered module are desirable for the efficient biosynthesis of novel natural product analogues. We have demonstrated that transacylation of malonyl groups onto an AT-null form of a mutant modular polyketide synthase by malonyl-CoA:ACP transacylase is an effective strategy for the engineered biosynthesis of site specifically modified polyketides. Using this strategy, 6-deoxyerythronolide B synthase was engineered to exclusively produce 2-desmethyl-6-deoxyerythronolide B. The productivity of the modified system was comparable to that of the wild-type synthase in vitro and in vivo.     

Mechanism and specificity of the terminal thioesterase domain from the erythromycin polyketide synthase

BACKGROUND: Polyketides are important compounds with antibiotic and anticancer activities. Several modular polyketide synthases (PKSs) contain a terminal thioesterase (TE) domain probably responsible for the release and concomitant cyclization of the fully processed polyketide chain. Because the TE domain influences qualitative aspects of product formation by engineered PKSs, its mechanism and specificity are of considerable interest. RESULTS: The TE domain of the 6-deoxyerythronolide B synthase was overexpressed in Escherichia coli. When tested against a set of N-acetyl cysteamine thioesters the TE domain did not act as a cyclase, but showed significant hydrolytic specificity towards substrates that mimic important features of its natural substrate. Also the overall rate of polyketide chain release was strongly enhanced by a covalent connection between the TE domain and the terminal PKS module (by as much as 100-fold compared with separate TE and PKS 'domains'). CONCLUSIONS: The inability of the TE domain alone to catalyze cyclization suggests that macrocycle formation results from the combined action of the TE domain and a PKS module. The chain-length and stereochemical preferences of the TE domain might be relevant in the design and engineered biosynthesis of certain novel polyketides. Our results also suggest that the TE domain might loop back to catalyze the release of polyketide chains from both terminal and pre-terminal modules, which may explain the ability of certain naturally occurring PKSs, such as the picromycin synthase, to generate both 12-membered and 14-membered macrolide antibiotics.     

Enzymatic extender unit generation for in vitro polyketide synthase reactions: structural and functional showcasing of Streptomyces coelicolor MatB

In vitro experiments with modular polyketide synthases (PKSs) are often limited by the availability of polyketide extender units. To determine the polyketide extender units that can be biocatalytically accessed via promiscuous malonyl-CoA ligases, structural and functional studies were conducted on Streptomyces coelicolor MatB. We demonstrate that this adenylate-forming enzyme is capable of producing most CoA-linked polyketide extender units as well as pantetheine- and N-acetylcysteamine-linked analogs useful for in?vitro PKS studies. Two ternary product complex structures, one containing malonyl-CoA and AMP and the other containing (2R)-methylmalonyl-CoA and AMP, were solved to 1.45?? and 1.43?? resolution, respectively. MatB crystallized in the thioester-forming conformation, making extensive interactions with the bound extender unit products. This first structural characterization of an adenylate-forming enzyme that activates diacids reveals the molecular details for how malonate and its derivatives are accepted. The orientation of the α-methyl group of bound (2R)-methylmalonyl-CoA, indicates that it is necessary to epimerize α-substituted extender units formed by MatB before they can be accepted by PKS acyltransferase domains. We demonstrate the in?vitro incorporation of methylmalonyl groups ligated by MatB to CoA, pantetheine, or N-acetylcysteamine into a triketide pyrone by the terminal module of the 6-deoxyerythronolide B synthase. Additionally, a means for quantitatively monitoring certain in?vitro PKS reactions using MatB is presented.     

Acyl-CoA subunit selectivity in the pikromycin polyketide synthase PikAIV: steady-state kinetics and active-site occupancy analysis by FTICR-MS

Polyketide natural products generated by type I modular polyketide synthases (PKSs) are vital components in our drug repertoire. To reprogram these biosynthetic assembly lines, we must first understand the steps that occur within the modular "black boxes." Herein, key steps of acyl-CoA extender unit selection are explored by in?vitro biochemical analysis of the PikAIV PKS model system. Two complementary approaches are employed: a fluorescent-probe assay for steady-state kinetic analysis, and Fourier Transform Ion Cyclotron Resonance-mass spectrometry (FTICR-MS) to monitor active-site occupancy. Findings from five enzyme variants and four model substrates have enabled a model to be proposed involving catalysis based upon acyl-CoA substrate loading followed by differential rates of hydrolysis. These efforts suggest a strategy for future pathway engineering efforts using unnatural extender units with slow rates of hydrolytic off-loading from the acyltransferase domain.     

Type III polyketide synthase beta-ketoacyl-ACP starter unit and ethylmalonyl-CoA extender unit selectivity discovered by Streptomyces coelicolor genome mining

Polyketide synthases (PKSs) are involved in the biosynthesis of many important natural products. In bacteria, type III PKSs typically catalyze iterative decarboxylation and condensation reactions of malonyl-CoA building blocks in the biosynthesis of polyhydroxyaromatic products. Here it is shown that Gcs, a type III PKS encoded by the sco7221 ORF of the bacterium Streptomyces coelicolor, is required for biosynthesis of the germicidin family of 3,6-dialkyl-4-hydroxypyran-2-one natural products. Evidence consistent with Gcs-catalyzed elongation of specific beta-ketoacyl-ACP products of the fatty acid synthase FabH with ethyl- or methylmalonyl-CoA in the biosynthesis of germicidins is presented. Selectivity for beta-ketoacyl-ACP starter units and ethylmalonyl-CoA as an extender unit is unprecedented for type III PKSs, suggesting these enzymes may be capable of utilizing a far wider range of starter and extender units for natural product assembly than believed until now.     

Unprecedented mechanism of chain length determination in fungal aromatic polyketide synthases

Fungal aromatic polyketides show remarkable structural diversity fundamentally derived from variations in chain length and cyclization pattern. Their basic skeletons are synthesized by multifunctional iterative type I polyketide synthases (PKSs). Recently, we have found that the C-terminal thioesterase (TE)-like domain of Aspergillus nidulans WA catalyzes Claisen-type cyclization to form the B-ring of naphthopyrone YWA1. Here we report the unprecedented mechanism of chain length determination by the C-terminal TE-like domain of Colletotrichum lagenarium PKS1, which, in addition to catalyzing Claisen-type cyclization, intercepts the polyketomethylene intermediate from the acyl carrier protein domain during the condensation reaction to produce shorter chain length products. This chain length determination system is novel among PKSs, including bacterial and plant PKSs. The functional diversity of the TE-like domain directly influences the structural diversity of aromatic polyketides in C. lagenarium PKS1.     

Formation of functional heterologous complexes using subunits from the picromycin, erythromycin and oleandomycin polyketide synthases

BACKGROUND: Recently developed tools for the genetic manipulation of modular polyketide synthases (PKSs) have advanced the development of combinatorial biosynthesis technologies for drug discovery. Although many of the current techniques involve engineering individual domains or modules of the PKS, few experiments have addressed the ability to combine entire protein subunits from different modular PKSs to create hybrid polyketide pathways. We investigated this possibility by in vivo assembly of heterologous PKS complexes using natural and altered subunits from related macrolide PKSs. RESULTS: The pikAI and pikAII genes encoding subunits 1 and 2 (modules 1-4) of the picromycin PKS (PikPKS) and the eryAIII gene encoding subunit 3 (modules 5-6) of the 6-deoxyerythronolide B synthase (DEBS) were cloned in two compatible Streptomyces expression vectors. A strain of Streptomyces lividans co-transformed with the two vectors produced the hybrid macrolactone 3-hydroxynarbonolide. Co-expression of the same pik genes with the gene for subunit 3 of the oleandomycin PKS (OlePKS) was also successful. A series of hybrid polyketide pathways was then constructed by combining PikPKS subunits 1 and 2 with modified DEBS3 subunits containing engineered domains in modules 5 or 6. We also report the effect of junction location in a set of DEBS-PikPKS fusions. CONCLUSIONS: We show that natural as well as engineered protein subunits from heterologous modular PKSs can be functionally assembled to create hybrid polyketide pathways. This work represents a new strategy that complements earlier domain engineering approaches for combinatorial biosynthesis in which complete modules or PKS protein subunits, in addition to individual enzymatic domains, are used as building blocks for PKS engineering.     

Exploring the biosynthetic potential of bimodular aromatic polyketide synthases

Polycyclic aromatic polyketides such as actinorhodin and tetracenomycin are synthesized from acetate equivalents by type II polyketide synthases (PKS). Their carbon chain backbones are derived from malonyl-CoA building blocks through the action of a minimal PKS module consisting of a ketosynthase, a chain length factor, an acyl carrier protein (ACP) and a malonyl-CoA/ACP transacylase. In contrast to these acetogenic polyketides, the backbones of a few aromatic polyketide natural products, such as the R1128 antibiotics, are primed by non-acetate building blocks. These polyketides are synthesized by bimodular PKSs comprising of a dedicated initiation module, which includes a ketosynthase, acyl transferase and ACP, as well as a minimal PKS module. Recently we showed that regioselectively modified polyketides could be synthesized through the genetic recombination of initiation modules and minimal PKS modules from different polyketide biosynthetic pathways (Tang et al. PLoS Biol. 2004, 2, 227-238). For example, the actinorhodin and tetracenomycin minimal PKSs could accept and elongate unnatural primer units from the R1128 initiation module. In this report we provide further examples of using heterologous bimodular PKSs for the engineered biosynthesis of new aromatic polyketides. In addition to providing insights into the biosynthetic mechanisms of aromatic PKSs, our findings also highlight considerable potential for crosstalk between amino acid catabolism and aromatic polyketide biosynthesis. For example, exogenously supplied unnatural amino acids are efficiently incorporated into bioactive anthraquinone antibiotics.     

Knowledge-based design of bimodular and trimodular polyketide synthases based on domain and module swaps: a route to simple statin analogues.

BACKGROUND: Polyketides are structurally diverse natural products that have a range of medically useful activities. Nonaromatic bacterial polyketides are synthesised on modular polyketide synthase (PKS) multienzymes, in which each cycle of chain extension requires a different 'module' of enzymatic activities. Attempts to design and construct modular PKSs that synthesise specified novel polyketides provide a particularly stringent test of our understanding of PKS structure and function. RESULTS: We have constructed bimodular and trimodular PKSs based on DEBS1-TE, a derivative of the erythromycin PKS that contains only modules 1 and 2 and a thioesterase (TE), by substituting multiple domains with appropriate counterparts derived from the rapamycin PKS. Hybrid PKSs were obtained that synthesised the predicted target triketide lactones, which are simple analogues of cholesterol-lowering statins. In constructing intermodular fusions, whether between modules in the same or in different proteins, it was found advantageous to preserve intact the acyl carrier protein-ketosynthase (ACP-KS) didomain that spans the junction between successive modules. CONCLUSIONS: Relatively simple considerations govern the construction of functional hybrid PKSs. Fusion sites should be chosen either in the surface-accessible linker regions between enzymatic domains, as previously revealed, or just inside the conserved margins of domains. The interaction of an ACP domain with the adjacent KS domain, whether on the same polyketide or not, is of particular importance, both through conservation of appropriate protein-protein interactions, and through optimising molecular recognition of the altered polyketide chain in the key transfer of the acyl chain from the ACP of one module to the KS of the downstream module.     

The hidden steps of domain skipping: macrolactone ring size determination in the pikromycin modular polyketide synthase

The pikromycin (Pik) polyketide synthase (PKS) from Streptomyces venezuelae comprises four multifunctional polypeptides (PikAI, PikAII, PikAIII, and PikAIV). This PKS can generate 12- and 14-membered ring macrolactones (10-deoxymethynolide and narbonolide, respectively) through the activity of its terminal modules (PikAIII and PikAIV). We performed a series of experiments involving the functional replacement of PikAIV in mutant strains with homodimeric and heterodimeric PikAIV modules to investigate the details of macrolactone ring size determination. The results suggest a new and surprising mechanism by which the penultimate hexaketide chain elongation intermediate is transferred from PikAIII ACP5 to PikAIV ACP6 before release by the terminal thioesterase domain. Elucidation of this chain transfer mechanism provides important new details about alternative macrolactone ring size formation in modular PKSs and contributes to the potential for rational design of structural diversity by combinatorial biosynthesis.     

Mechanistic analysis of acyl transferase domain exchange in polyketide synthase modules

Many polyketides are synthesized by a class of multifunctional enzymes called type I modular polyketide synthases (PKSs). Several reports have described the power of predictively altering polyketide structure by replacing individual PKS domains with homologues from other PKSs. For example, numerous erythromycin analogues have been generated by replacing individual methylmalonyl-specific acyl transferase (AT) domains of the 6-deoxyerythronolide B synthase (DEBS) with malonyl-, ethylmalonyl-, or methoxymalonyl-specific domains. However, the construction of hybrid PKS modules often attenuates product formation both kinetically and distributively. The molecular basis for this mechanistic imperfection is not understood. We have systematically analyzed the impact of replacing an AT domain of DEBS on acyl-AT formation, acyl-CoA:HS-NAc acyl transferase activity, acyl-CoA:ACP acyl transferase activity (nucleophile charging), acyl-SNAc:ketosynthase acyl transferase activity (electrophile charging), and beta-ketoacyl ACP synthase activity (condensation). As usual, domain junctions were located in interdomain regions flanking the AT domain. Kinetic analysis of hybrid modules containing either malonyl transferase or methylmalonyl transferase domains revealed a 15-20-fold decrease in overall turnover numbers of the hybrid modules as compared to the wild-type module. In contrast, both the activity and the specificity of the heterologous AT domains remained unaffected. Moreover, no defects could be detected in the ability of the heterologous AT domains to catalyze acyl-CoA:ACP acyl transfer. Single turnover studies aimed at directly probing the ketosynthase-catalyzed reaction led to two crucial findings. First, wild-type modules catalyzed chain elongation with comparable efficiency regardless of whether methylmalonyl-ACP or malonyl-ACP were the nucleophilic substrates. Second, chain elongation in all hybrid modules tested was seriously attenuated relative to the wild-type module. Our data suggest that, as currently practiced, the most deleterious impact of AT domain swapping is not on the substrate specificity. Rather, it is due to the impaired ability of the KS and ACP domains in the hybrid module to catalyze chain elongation. Consistent with this proposal, limited proteolysis of wild-type and hybrid modules showed major differences in cleavage patterns, especially in the region between the KR and ACP domains.     

Enzymatic formation of an unnatural C(6)-C(5) aromatic polyketide by plant type III polyketide synthases

[reaction: see text] Substrate specificities of plant polyketide synthases (PKSs) were investigated using analogues of malonyl-CoA, the extension unit of the polyketide chain elongation reactions. When incubated with methylmalonyl-CoA and 4-coumaroyl-CoA, plant PKSs (chalcone synthase from Scutellaria baicalensis, stilbene synthase from Arachis hypogaea, and benzalacetone synthase from Rheum palmatum) afforded an unnatural C(6)-C(5) aromatic polyketide, 1-(4-hydroxyphenyl)pent-1-en-3-one, formed by one-step decarboxylative condensation of the two substrates. In contrast, succinyl-CoA was not accepted as a substrate by the enzymes.     

Novel octaketide macrolides related to 6-deoxyerythronolide B provide evidence for iterative operation of the erythromycin polyketide synthase

BACKGROUND: The macrolide antibiotic erythromycin A, like other complex aliphatic polyketides, is synthesised by a bacterial modular polyketide synthase (PKS). Such PKSs, in contrast to other fatty acid and polyketide synthases which work iteratively, contain a separate set or module of enzyme activities for each successive cycle of polyketide chain extension, and the number and type of modules together determine the structure of the polyketide product. Thus, the six extension modules of the erythromycin PKS (DEBS) together catalyse the production of the specific heptaketide 6-deoxyerythronolide B. RESULTS: A mutant strain of the erythromycin producer Saccharopolyspora erythraea, which accumulates the aglycone intermediate erythronolide B, was found unexpectedly to produce two novel octaketides, both 16-membered macrolides. These compounds were detectable in fermentation broths of wild-type S. erythraea, but not in a strain from which the DEBS genes had been specifically deleted. From their structures, both of these octaketides appear to be aberrant products of DEBS in which module 4 has 'stuttered', that is, has catalysed two successive cycles of chain extension. CONCLUSIONS: The isolation of novel DEBS-derived octaketides provides the first evidence that an extension module in a modular PKS has the potential to catalyse iterative rounds of chain elongation like other type I FAS and PKS systems. The factors governing the extent of such 'stuttering' remain to be determined.     

Biosynthesis and attachment of novel bacterial polyketide synthase starter units

The biosynthesis and mode of attachment of a wide range of polyketide synthase (PKS) starter units in bacteria are covered in this review. Natural, unnatural, and engineered starter units associated with type I and type II PKSs are reported. The literature through early 2001 is reviewed, and 240 references cited.     

Structure-based dissociation of a type I polyketide synthase module

Individual modules of modular polyketide synthases (PKSs) such as 6-deoxyerythronolide B synthase (DEBS) consist of conserved, covalently linked domains separated by unconserved intervening linker sequences. To better understand the protein-protein and enzyme-substrate interactions in modular catalysis, we have exploited recent structural insights to prepare stand-alone domains of selected DEBS modules. When combined in vitro, ketosynthase (KS), acyl transferase (AT), and acyl carrier protein (ACP) domains of DEBS module 3 catalyzed methylmalonyl transfer and diketide substrate elongation. When added to a minimal PKS, ketoreductase domains from DEBS modules 1, 2, and 6 showed specificity for the beta-ketoacylthioester substrate, but not for either the ACP domain carrying the polyketide substrate or the KS domain that synthesized the substrate. With insights into catalytic efficiency and specificity of PKS modules, our results provide guidelines for constructing optimal hybrid PKS systems.     

Assembly line termination in cylindrocyclophane biosynthesis: discovery of an editing type II thioesterase domain in a type I polyketide synthase

The termination step is an important source of structural diversity in polyketide biosynthesis. Most type I polyketide synthase (PKS) assembly lines are terminated by a thioesterase (TE) domain located at the C-terminus of the final module, while other PKS assembly lines lack a terminal TE domain and are instead terminated by a separate enzyme in trans. In cylindrocyclophane biosynthesis, the type I modular PKS assembly line is terminated by a freestanding type III PKS (CylI). Unexpectedly, the final module of the type I PKS (CylH) also possesses a C-terminal TE domain. Unlike typical type I PKSs, the CylH TE domain does not influence assembly line termination by CylI in vitro. Instead, this domain phylogenetically resembles a type II TE and possesses activity consistent with an editing function. This finding may shed light on the evolution of unusual PKS termination logic. In addition, the presence of related type II TE domains in many cryptic type I PKS and nonribosomal peptide synthetase (NRPS) assembly lines has implications for pathway annotation, product prediction, and engineering.     

MCAT is not required for in vitro polyketide synthesis in a minimal actinorhodin polyketide synthase from Streptomyces coelicolor

Background: It has been proposed that Streptomyces malonyl CoA:holo acyl carrier protein transacylases (MCATs) provide a link between fatty acid and polyketide biosynthesis. Two recent studies have provided evidence that the presence of MCAT is essential for polyketide synthesis to proceed in reconstituted minimal polyketide synthases (PKSs). In contrast to this, we previously showed that the holo acyl carrier proteins (ACPs) from type II PKSs are capable of catalytic self -malonylation in the presence of malonyl CoA, which suggests that MCAT might not be necessary for polyketide biosynthesis.Results: We reconstituted a homologous actinorhodin (act) type II minimal PKS in vitro, When act holo-ACP is present in limiting concentrations, MCAT is required by the synthase complex in order for polyketide biosynthesis to proceed. When holo-ACP is present in excess, however, efficient polyketide synthesis proceeds without MCAT, The rate of polyketide production increases with holo-ACP concentration, but at low ACP concentration or equimolar ACP:KS:CLF (KS, ketosynthase; CLF, chain length determining factor) concentrations this rate is significantly lower than expected, indicating that free holo-ACP is sequestered by the KS/CLF complex.Conclusions: The rate of polyketide biosynthesis is dictated by the ratio of holo-ACP to KS and CLF, as well as by the total protein concentration, There is no absolute requirement for MCAT in polyketide biosynthesis in vitro, although the role of MCAT during polyketide synthesis in vivo remains an open question. MCAT might be responsible for the rate enhancement of malonyl transfer at very low free holo-ACP concentrations or it could be required to catalyse the transfer of malonyl groups from malonyl CoA to sequestered holo-ACP.     

Substrate specificity of the acyl transferase domains of EpoC from the epothilone polyketide synthase

The production of epothilone mixtures is a direct consequence of the substrate tolerance of the module 3 acyltransferase (AT) domain of the epothilone polyketide synthase (PKS) which utilises both malonyl- and methylmalonyl-CoA extender units. Particular amino acid motifs in the active site of AT domains influence substrate selection for methylmalonyl-CoA (YASH) or malonyl-CoA (HAFH). This motif appears in hybrid form (HASH) in epoAT3 and may represent the molecular basis for the relaxed specificity of the domain. To investigate this possibility the AT domains from modules 2 and 3 of the epothilone PKS were examined in the heterologous DEBS1-TE model PKS. Substitution of AT1 of DEBS1-TE by epoAT2 and epoAT3 both resulted in functional PKSs, although lower yields of total products were observed when compared to DEBS1-TE (2% and 11.5% respectively). As expected, epoAT3 was significantly more promiscuous in keeping with its nature during epothilone biosynthesis. When the mixed motif (HASH) of epoAT3 within the hybrid PKS was mutated to HAFH (indicative of malonyl-CoA selection) it resulted in a non-productive PKS. When this mixed motif was converted to YASH (indicative of methylmalonyl-CoA selection) the selectivity of the hybrid PKS for methylmalonyl-CoA showed no statistically significant increase, and was associated with a loss of productivity.