Precursor-Directed Biosynthesis and Biological Testing of omega-Alicyclic- and neo-Branched Tunicamycin N-Acyl Variants

Tunicamycins (TUNs) are Streptomyces-derived natural products, widely used to block protein N-glycosylation in eukaryotes or cell wall biosynthesis in bacteria. Modified or synthetic TUN analogues that uncouple these activities have considerable potential as novel mode-of-action antibacterial agents. Chemically modified TUNs reported previously with attenuated activity on yeast have pinpointed eukaryotic-specific chemophores in the uridyl group and the N-acyl chain length and terminal branching pattern. A small molecule screen of fatty acid biosynthetic primers identified several novel alicyclic- and neo-branched TUN N-acyl variants, with primer incorporation at the terminal omega-acyl position. TUNs with unique 5- and 6-carbon ω-cycloalkane and ω-cycloalkene acyl chains are produced under fermentation and in yields comparable with the native TUN. The purification, structural assignments, and the comparable antimicrobial properties of 15 of these compounds are reported, greatly extending the structural diversity of this class of compounds for potential medicinal and agricultural applications.

Fatty acid desaturase 2 (FADS2) but not FADS1 desaturates branched chain and odd chain saturated fatty acids

Branched chain fatty acids (BCFA) and linear chain/normal odd chain fatty acids (n-OCFA) are major fatty acids in human skin lipids, especially sebaceous gland (SG) wax esters. Skin lipids contain variable amounts of monounsaturated BCFA and n-OCFA, in some reports exceeding over 20% of total fatty acids. Fatty acid desaturase 2 (FADS2) codes for a multifunctional enzyme that catalyzes Δ4-, Δ6- and Δ8-desaturation towards ten unsaturated fatty acids but only one saturate, palmitic acid, converting it to 16:1n-10; FADS2 is not active towards 14:0 or 18:0. Here we test the hypothesis that FADS2 also operates on BCFA and n-OCFA. MCF-7 cancer cells stably expressing FADS1 or FADS2 along with empty vector control cells were incubated with anteiso-15:0, iso-16:0, iso-17:0, anteiso-17:0, iso-18:0, or n-17:0. BCFA were Δ6-desaturated by FADS2 as follows: iso-16:0 → iso-6Z-16:1, iso-17:0 → iso-6Z-17:1, anteiso-17:0 → anteiso-6Z-17:1 and iso-18:0 → iso-6Z-18:1. anteiso-15:0 was not desaturated in either FADS1 or FADS2 cells. n-17:0 was converted to both n-6Z-17:1 by FADS2 Δ6-desaturation and n-9Z-17:1 by SCD Δ9-desaturation. We thus establish novel FADS2-coded enzymatic activity towards BCFA and n-OCFA, expanding the number of known FADS2 saturated fatty acid substrates from one to six. Because of the importance of FADS2 in human skin, our results imply that dysfunction in activity of sebaceous FADS2 may play a role in skin abnormalities associated with skin lipids.

Engineering Streptomyces coelicolor for production of monomethyl branched chain fatty acids

Branched chain fatty acids (BCFA) are an appealing biorefinery-driven target of fatty acid (FA) production. BCFAs typically have lower melting points compared to straight chain FAs, making them useful in lubricants and biofuels. Actinobacteria, especially Streptomyces species, have unique secondary metabolism that are capable of producing not only antibiotics, but also high percentage of BCFAs in their membrane lipids. Since biosynthesis of polyketide (PK) and FA partially share common pathways to generate acyl-CoA precursors, in theory, Streptomyces sp. with high levels of PK antibiotics production can be easily manipulated into strains producing high levels of BCFAs. To increase the percentage of the BCFA moieties in lipids, we redirected acyl-CoA precursor fluxes from PK into BCFAs using S. coelicolor M1146 (M1146) as a host strain. In addition, 3-ketoacyl acyl carrier protein synthase III and branched chain α-keto acid dehydrogenase were overexpressed to push fluxes of branched chain acyl-CoA precursors towards FA synthesis. The maximum titer of 354.1 mg/L BCFAs, 90.3% of the total FA moieties, was achieved using M1146_dD-B, fadD deletion and bkdABC overexpression mutant of M1146 strain. Cell specific yield of 64.4 mg/L/g_cell was also achieved. The production titer and specific yield are the highest ever reported in bacterial cells, which provides useful insights to develop an efficient host strain for BCFAs.

Xanthocidin Derivatives from the Endophytic Streptomyces sp. AcE210 Provide Insight into Xanthocidin Biosynthesis

Xanthocidin and six new derivatives were isolated from the endophytic Streptomyces sp. AcE210. Their planar structures were elucidated by 1D and 2D NMR spectroscopy as well as by HRMS. The absolute configuration of one compound was determined by using vibrational circular dichroism spectroscopy (VCD). The structural similarities of xanthocidin and some of the isolated xanthocidin congeners to the methylenomycins A, B, and C suggested that the biosynthesis of these compounds might follow a similar route. Feeding studies with isotopically labelled [¹³ C₅ ]-l-valine showed that instead of utilizing acetyl-CoA as starter unit, which has been proposed for the methylenomycin biosynthesis, Streptomyces sp. AcE210 employs an isobutyryl-CoA starter unit, resulting in a branched side chain in xanthocidin. Further evidence for a comparable biosynthesis was given by the analysis of the genome sequence of Streptomyces sp. AcE210 that revealed a cluster of homologues to the mmy genes involved in methylenomycin biosynthesis.

Increased valinomycin production in mutants of Streptomyces sp. M10 defective in bafilomycin biosynthesis and branched-chain α-keto acid dehydrogenase complex expression

Streptomyces sp. M10 is a valinomycin-producing bacterial strain that shows potent bioactivity against Botrytis blight of cucumber plants. During studies to increase the yield of valinomycin (a cyclododecadepsipeptide) in strain M10, additional antifungal metabolites, including bafilomycin derivatives (macrolide antibiotics), were identified. To examine the effect of bafilomycin biosynthesis on valinomycin production, the bafilomycin biosynthetic gene cluster was cloned from the genome of strain M10, as were two branched-chain α-keto acid dehydrogenase (BCDH) gene clusters related to precursor supply for bafilomycin biosynthesis. A null mutant (M10bafm) of one bafilomycin biosynthetic gene (bafV) failed to produce bafilomycin, but resulted in a 1.2- to 1.5-fold increase in the amount of valinomycin produced. In another null mutant (M10bkdFm) of a gene encoding a subunit of the BCDH complex (bkdF), bafilomycin production was completely abolished and valinomycin production increased fourfold relative to that in the wild-type M10 strain. The higher valinomycin yield was likely the result of redistribution of the metabolic flux from bafilomycin to valinomycin biosynthesis, because the two antibiotics share a common precursor, 2-ketoisovaleric acid, a deamination product of valine. The results show that directing precursor flux toward active ingredient biosynthesis could be used as a prospective tool to increase the competence of biofungicides.

Bacterial fatty acid synthesis and its relationships with polyketide synthetic pathways

This review presents the most thoroughly studied bacterial fatty acid synthetic pathway, that of Escherichia coli and then discusses the exceptions to the E. coli pathway present in other bacteria. The known interrelationships between the fatty acid and polyketide synthetic pathways are also assessed, mainly in the Streptomyces group of bacteria. Finally, we present a compendium of methods for analysis of bacterial fatty acid synthetic pathways.

Alteration of the fatty acid profile of Streptomyces coelicolor by replacement of the initiation enzyme 3-ketoacyl acyl carrier protein synthase III (FabH)

The first elongation step of fatty acid biosynthesis by a type II dissociated fatty acid synthases is catalyzed by 3-ketoacyl-acyl carrier protein (ACP) synthase III (KASIII, FabH). This enzyme, encoded by the fabH gene, catalyzes a decarboxylative condensation between an acyl coenzyme A (CoA) primer and malonyl-ACP. In organisms such as Escherichia coli, which generate only straight-chain fatty acids (SCFAs), FabH has a substrate preference for acetyl-CoA. In streptomycetes and other organisms which produce a mixture of both SCFAs and branched-chain fatty acids (BCFAs), FabH has been shown to utilize straight- and branched-chain acyl-CoA substrates. We report herein the generation of a Streptomyces coelicolor mutant (YL/ecFabH) in which the chromosomal copy of the fabH gene has been replaced and the essential process of fatty acid biosynthesis is initiated by plasmid-based expression of the E. coli FabH (bearing only 35% amino acid identity to the Streptomyces enzyme). The YL/ecFabH mutant produces predominantly SCFAs (86%). In contrast, BCFAs predominate (approximately 70%) in both the S. coelicolor parental strain and S. coelicolor YL/sgFabH (a deltafabH mutant carrying a plasmid expressing the Streptomyces glaucescens FabH). These results provide the first unequivocal evidence that the substrate specificity of FabH observed in vitro is a determinant of the fatty acid made in an organism. The YL/ecFabH strain grows significantly slower on both solid and liquid media. The levels of FabH activity in cell extracts of YL/ecFabH were also significantly lower than those in cell extracts of YL/sgFabH, suggesting that a decreased rate of fatty acid synthesis may account for the observed decreased growth rate. The production of low levels of BCFAs in YL/ecFabH suggests either that the E. coli FabH is more tolerant of different acyl-CoAs substrates than previously thought or that there is an additional pathway for initiation of BCFA biosynthesis in Streptomyces coelicolor.

An acyl-CoA dehydrogenase is involved in the formation of the Δcis3 double bond in the acyl residue of the lipopeptide antibiotic friulimicin in Actinoplanes friuliensis

The lipopeptide antibiotic friulimicin, produced by Actinoplanes friuliensis, is an effective drug against Gram-positive bacteria, such as methicillin-resistant Staphylococcus epidermidis and Staphylococcus aureus strains. Friulimicin consists of a cyclic peptide core of ten amino acids and an acyl residue linked to an exocyclic amino acid. The acyl residue is essential for antibiotic activity, varies in length from C13 to C15, and carries a characteristic double bond at position Δcis3. Sequencing of a DNA fragment adjacent to a previously described fragment encoding some of the friulimicin biosynthetic genes revealed several genes whose gene products resemble enzymes of lipid metabolism. One of these genes, lipB, encodes an acyl-CoA dehydrogenase homologue. To elucidate the function of the LipB protein, a lipB insertion mutant was generated and the friulimicin derivative (FR242) produced by the mutant was purified. FR242 had antibiotic activity lower than friulimicin in a bioassay. Gas chromatography showed that the acyl residue of wild-type friulimicin contains a double bond, whereas a saturated bond was present in FR242. These results were confirmed by the heterologous expression of lipB in Streptomyces lividans T7, which led to the production of unsaturated fatty acids not found in the S. lividans T7 parent strain. These results indicate that the acyl-CoA dehydrogenase LipB is involved in the introduction of the unusual Δcis3 double bond into the acyl residue of friulimicin.

Production of Branched-Chain Alkylprodiginines in S. coelicolor by Replacement of the 3-Ketoacyl ACP Synthase III Initiation Enzyme, RedP

The enzyme RedP is thought to initiate the biosynthesis of the undecylpyrolle component of the antibiotic undecylprodiginine produced by Streptomyces coelicolor. RedP has homology to FabH, which initiates fatty acid biosynthesis by condensing the appropriate acyl-CoA starter unit with malonyl ACP. We have generated a redP-deletion mutant of S. coelicolor M511 (SJM1) and shown that it produces reduced levels of prodiginines and two new analogs, methylundecylprodiginine and methyldodecylprodiginine. Incorporation studies with perdeuterated valine were consistent with these being generated using methylbutyryl-CoA and isobutyryl-CoA as starter units, respectively. Plasmid-based expression of a streptomycete fabH in the SJM1 mutant led to restoration of overall prodiginine titers but the same overall ratio of undecylprodiginines and novel prodiginines. Thus, the redP FabH can be replaced by FabH enzymes with different substrate specificities and provides a method for generating novel prodiginines.

Monomethyl branched-chain fatty acids play an essential role in Caenorhabditis elegans development

Monomethyl branched-chain fatty acids (mmBCFAs) are commonly found in many organisms from bacteria to mammals. In humans, they have been detected in skin, brain, blood, and cancer cells. Despite a broad distribution, mmBCFAs remain exotic in eukaryotes, where their origin and physiological roles are not understood. Here we report our study of the function and regulation of mmBCFAs in Caenorhabditis elegans, combining genetics, gas chromatography, and DNA microarray analysis. We show that C. elegans synthesizes mmBCFAs de novo and utilizes the long-chain fatty acid elongation enzymes ELO-5 and ELO-6 to produce two mmBCFAs, C15ISO and C17ISO. These mmBCFAs are essential for C. elegans growth and development, as suppression of their biosynthesis results in a growth arrest at the first larval stage. The arrest is reversible and can be overcome by feeding the arrested animals with mmBCFA supplements. We show not only that the levels of C15ISO and C17ISO affect the expression of several genes, but also that the activities of some of these genes affect biosynthesis of mmBCFAs, suggesting a potential feedback regulation. One of the genes, lpd-1, encodes a homolog of a mammalian sterol regulatory element-binding protein (SREBP 1c). We present results suggesting that elo-5 and elo-6 may be transcriptional targets of LPD-1. This study exposes unexpected and crucial physiological functions of C15ISO and C17ISO in C. elegans and suggests a potentially important role for mmBCFAs in other eukaryotes.

Product diversity and regulation of type II fatty acid synthases

Fatty acid biosynthesis is catalyzed in most bacteria by a group of highly conserved proteins known as the type II fatty acid synthase (FAS II) system. FAS II has been extensively studied in the Escherichia coli model system, and the recent explosion of bioinformatic information has accelerated the investigation of the pathway in other organisms, mostly important human pathogens. All FAS II systems possess a basic set of enzymes for the initiation and elongation of acyl chains. This review focuses on the variations on this basic theme that give rise to the diversity of products produced by the pathway. These include multiple mechanisms to generate unsaturated fatty acids and the accessory components required for branched-chain fatty acid synthesis in Gram-positive bacteria. Most of the known mechanisms that regulate product distribution of the pathway arise from the fundamental biochemical properties of the expressed enzymes. However, newly identified transcriptional factors in bacterial fatty acid biosynthetic pathways are a fertile field for new investigation into the genetic control of the FAS II system. Much more work is needed to define the role of these factors and the mechanisms that regulate their DNA binding capability, but there appear to be fundamental differences in how the expression of the pathway genes is controlled in Gram-negative and in Gram-positive bacteria.Key words: fatty acid synthase, bacteria.

MeaA, a putative coenzyme B12-dependent mutase, provides methylmalonyl coenzyme A for monensin biosynthesis in Streptomyces cinnamonensis

The ratio of the major monensin analogs produced by Streptomyces cinnamonensis is dependent upon the relative levels of the biosynthetic precursors methylmalonyl-coenzyme A (CoA) (monensin A and monensin B) and ethylmalonyl-CoA (monensin A). The meaA gene of this organism was cloned and sequenced and was shown to encode a putative 74-kDa protein with significant amino acid sequence identity to methylmalonyl-CoA mutase (MCM) (40%) and isobutyryl-CoA mutase (ICM) large subunit (36%) and small subunit (52%) from the same organism. The predicted C terminus of MeaA contains structural features highly conserved in all coenzyme B12-dependent mutases. Plasmid-based expression of meaA from the ermE* promoter in the S. cinnamonensis C730.1 strain resulted in a decreased ratio of monensin A to monensin B, from 1:1 to 1:3. Conversely, this ratio increased to 4:1 in a meaA mutant, S. cinnamonensis WM2 (generated from the C730.1 strain by insertional inactivation of meaA by using the erythromycin resistance gene). In both of these experiments, the overall monensin titers were not significantly affected. Monensin titers, however, did decrease over 90% in an S. cinnamonensis WD2 strain (an icm meaA mutant). Monensin titers in the WD2 strain were restored to at least wild-type levels by plasmid-based expression of the meaA gene or the Amycolatopsis mediterranei mutAB genes (encoding MCM). In contrast, growth of the WD2 strain in the presence of 0.8 M valine led only to a partial restoration (<25%) of monensin titers. These results demonstrate that the meaA gene product is significantly involved in methylmalonyl-CoA production in S. cinnamonensis and that under the tested conditions the presence of both MeaA and ICM is crucial for monensin production in the WD2 strain. These results also indicate that valine degradation, implicated in providing methylmalonyl-CoA precursors for many polyketide biosynthetic processes, does not do so to a significant degree for monensin biosynthesis in the WD2 mutant.

Identification of a cyclohexylcarbonyl CoA biosynthetic gene cluster and application in the production of doramectin

The side chain of the antifungal antibiotic ansatrienin A from Streptomyces collinus contains a cyclohexanecarboxylic acid (CHC)-derived moiety. This moiety is also observed in trace amounts of omega-cyclohexyl fatty acids (typically less than 1% of total fatty acids) produced by S. collinus. Coenzyme A-activated CHC (CHC-CoA) is derived from shikimic acid through a reductive pathway involving a minimum of nine catalytic steps. Five putative CHC-CoA biosynthetic genes in the ansatrienin biosynthetic gene cluster of S. collinus have been identified. Plasmid-based heterologous expression of these five genes in Streptomyces avermitilis or Streptomyces lividans allows for production of significant amounts of omega-cyclohexyl fatty acids (as high as 49% of total fatty acids). In the absence of the plasmid these organisms are dependent on exogenously supplied CHC for omega-cyclohexyl fatty acid production. Doramectin is a commercial antiparasitic avermectin analog produced by fermenting a bkd mutant of S. avermitilis in the presence of CHC. Introduction of the S. collinus CHC-CoA biosynthetic gene cassette into this organism resulted in an engineered strain able to produce doramectin without CHC supplementation. The CHC-CoA biosynthetic gene cluster represents an important genetic tool for precursor-directed biosynthesis of doramectin and has potential for directed biosynthesis in other important polyketide-producing organisms.