Fig 1.
Biosynthesis of archetypal aromatic polyketides.
The central dark grey pane (extension units) illustrates biosynthesis of the nascent poly-β-ketide chain by the KS/CLF heterodimer that remains tethered to the ACP. The red moiety represents the starter unit, which can be altered for increased product diversity. The left blue pane shows the enzymatic constituents necessary to prime type II polyketide biosynthesis (priming units), e.g., phosphopantetheinylation of ACP from apo-ACP to holo-ACP catalyzed by a 4′ PPTase, and acylation of the holo-ACP with the starter unit, e.g., via an MCAT, acyl-ACP synthetase, or via self-malonylation. The top right light grey pane (tailoring units) illustrates the typical primary tailoring enzymes. The bottom right yellow pane indicates the point at which the modified polyketide chain is released from the ACP. The free metabolite can be further modified by secondary tailoring enzymes that do not interact with the polyketide-ACP intermediate, e.g., glycosyltransferases, prenyltransferases, or halogenases, to introduce chemical diversity. *Only primary tailoring enzymes are shown schematically. ACP, acyl carrier protein; ARO, aromatase; CLF, chain length factor; CYC, cyclases that are responsible for biosynthesis of the aromatic carbon core scaffold; KR, ketoreductase; KS, ketosynthase; MCAT, malonyl-CoA:ACP transacylase; PPTase, phosphopantetheinyl transferase.
Fig 2.
Schematic representation of aromatic polyketide biosynthesis.
(a) Expected octaketide shunt metabolites after each biosynthetic step are designated by grey dotted line. Biosynthetic steps are confined to individual grey boxes that proceed in the order of biosynthesis. Circular arrows within each box represent the ability to functionally substitute biosynthetic enzymes for homologues. (b) Examples of plausible biosynthetic pathway perturbations: b1, the exchange of an octaketide producing PKS heterodimer with a decaketide producer (XU exchange); b2, compound maturation despite loss of KR (TU modification); and b3, alteration of polyketide starter unit by exchanging PU enzyme constituents as well as functional exchange of an aromatase/cyclase and loss of supplementary enzymes (PU and TU exchange). Supplementary enzymes can be variable in function. (c) Structures of AQ256 (1) and its dianthrone (13). ACP, acyl carrier protein; Aro/Cyc, aromatase/cyclase; CLF, chain length factor; DMAC, 3,8-dihydroxy-methylanthraquinone carboxylic acid; KS, ketosynthase; PKS, polyketide synthase; PU, priming unit; TU, tailoring unit; XU, extension unit.
Fig 3.
Identification and expression of soluble KS/CLF heterodimers in E. coli.
(a and b) Phylogeny of KS and CLF sequences from a dataset of 58 characterised type II PKSs derived from the MiBIG repository, respectively. The clades representing canonical type II KS and CLF are denoted by red and yellow wedges, respectively. Both alignments include FabF sequences from S. avermitilis, E. coli, and B. subtilis. Red and yellow dots denote His6-AntE and AntD, respectively. This colouring is conserved throughout the figure. (c) Denaturing PAGE showing soluble protein extracted from E. coli BL21(DE3) (lane 1) and E. coli BL21(DE3) pBbA2k-plumPKS harbouring the mPKS from P. luminescens (lane 2) induced at 30°C. Lanes 3 and 4 mirror those of 1 and 2; however, they show soluble protein expressed when incubated at 20°C. (d) Denaturing PAGE gel of AntD and E purified by IMAC. Lane 1: protein flow through, lane 2: protein eluted at 50 mM imidazole, lane 3: 500 mM imidazole column wash. (e) Western blot of purified AntDE protein showing signal corresponding to a single AntE band. Lane 1: PAGE ladder, lane 2: purified AntDE protein, and lane 3: His-tagged mCherry (approximately 29 kDa) fusion protein as positive control. All numbers correspond to standard protein ladders and are defined in kDa. Theoretical size of His6-AntE and AntD is 42.43 kDa and 46.16 kDa, respectively. CLF, chain length factor; His, Histidine tag; IMAC, immobilised metal ion affinity chromatography; KS, ketosynthase; MiBIG, Minimum Information about a Biosynthetic Gene cluster; mPKS, minimal PKS; PKS, polyketide synthase.
Fig 4.
Expression of AntDEFBG in E. coli.
(a and b) Three-dimensional and two-dimensional EICs of all observable masses between 301.0661 and 301.0714 m/z. The theoretical mass of expected shunt metabolites AUR367 (5) and SEK4 (2) is [M-H2O+H]+ 301.07067. (c and d) Three-dimensional and two-dimensional EICs of all observable masses between 319.0764 and 319.086 m/z. The theoretical mass of expected shunt metabolites SEK4 (2) and SEK4b (3) is [M+H]+ 319.08099, 15 ppm tolerance. Samples I through V are filtered supernatant from culture of (I) E. coli BL21, (II) E. coli BL21 pBbB1a-GFP pACYCDuet-1 (empty vector control), (III) E. coli BL21 pBbB1a-plumPKS pACYCDuet-1 (AntDEF), (IV) E. coli BL21 pBbB1a-plumPKS pACYC93 (AntDEFB), and (V) E. coli BL21 pBbB1a-plumPKS pACYC8893 (AntDEFBG). Extracted ion count was normalised to final cell density (OD600). All bolded numbers correspond to Fig 2. EIC, extracted ion chromatogram; OD600, optical density measured at 600 nm.
Fig 5.
Anthraquinone identification and characterisation.
(a) EIC in negative ionisation mode for shunt metabolites described in Fig 2A, in addition to anthraquinones produced by P. luminescens TT01 and predicted pathway end compounds. All EICs represent theoretical mass for [M-H]− ± 5 ppm. Red, blue, and black lines represent EICs of E. coli BL21 (host control), E. coli BL21 pACYCDuet-1 (plasmid control), and E. coli BL21 pACYCAnthraquinone normalized to final cell density. EICs displaying masses from positive ionisation mode are detailed in S17 Fig. (b) Schematic diagram showing intramolecular couplings between nuclei of E. coli–produced AQ256, which were determined by COSY (green), HSQC (grey), and HMBC (blue scale) two-dimensional NMR spectroscopies. Coupling constants are detailed in Materials and methods. COSY, correlation spectroscopy; DMAC, 3,8-dihydroxy-methylanthraquinone carboxylic acid; EIC, extracted ion chromatogram; HMBC, heteronuclear multiple bond correlation; HSQC, heteronuclear single quantum correlation; NMR, Nuclear magnetic resonance.
Fig 6.
Complementation of the anthraquinone BGC with actinorhodin components.
(a) Typical exometabolome HPLC profiles of E. coli BL21, E. coli BL21 expressing empty vector pACYCDuet-1, KR antA knockout mutant pACYCAntΔAntA, wild-type Sco5086 KR complemented plasmid pACYCAntwtKR, refactored Sco5086 KR complemented plasmid pACYCAntrefKR, and pACYCAnthraquinone compared with fully characterised AQ256 standards and M9 growth media at 434 nm. The UV-Vis spectrum for peaks designated 1 are as follows: AQ256 standard λmax at 216, 264, 282, 434, and 583 nm; E. coli BL21 pACYCAnthraquinone λmax at 215, 263, 283, and 434 nm; E. coli pACYCAntrefKR λmax at 215, 283, and 435; and E. coli BL21 pACYCAntwtKR λmax at 216, 282, 435, and 585 in agreement with AQ256 (S15 Fig). Spectra were limited to 215–600 nm. (b) Typical exometabolome HPLC profiles of E. coli BL21, E. coli BL21 expressing empty vector pACYCDuet-1, ARO/CYC knockout mutant pACYCAntΔAntH, wild-type Sco5090 ARO/CYC complemented plasmid pACYCAntwtCYC, refactored Sco5090 ARO/CYC complemented plasmid pACYCAntrefCYC, and pACYCAnthraquinone compared with standards as above. Compound reference numbers are as described in Fig 2: 1: AQ256; 11: Aloesaponarin II; and 10: DMAC. The UV-Vis spectrum for the 3 most abundant peaks are as follows: 1: λmax at 216, 264, 282, 434, and 583 nm in agreement with AQ-256 (S15 Fig); 10: λmax at 217, 408, and 650 nm; 11: λmax at 214, 277, 409, and 582 nm, with tR of 16.004 min (960 s), 14.62 min (877.2 s), and 16.742 min (1004.5 s), respectively. (c) Typical exometabolite HPLC profile of E. coli BL21, E. coli BL21 expressing empty vector pACYCDuet-1, an α/β hydratase knockout mutant pACYCAntΔAntI, E. coli BL21 harbouring pACYCAnthraquinone, and anthraquinone standards. All numbering and λmax Figs are as in panel b. ARO, aromatase; BGC, biosynthesis gene cluster; CYC, cyclase; DMAC, 3,8-dihydroxy-methylanthraquinone carboxylic acid; HPLC, high resolution liquid chromatography; KR, ketoreductase; tR, retention time; UV-Vis, UV-visible.
Fig 7.
Introducing chemical diversity through supplementation and substitution of secondary tailoring enzymes.
A schematic representation of an AntA-I plug-and-play biosynthetic scaffold extended with enzymes sourced from a range of organisms. Metabolites produced through addition of O-methyltransferase (from M. truncatula) and RadH (from Chaetomium chiversii) are shown in red and blue boxes correspondingly. Enzymes within AntA-I plug-and-play core biosynthetic pathway are represented in light grey boxes. Additional non-cognate secondary tailoring enzymes are shown in darker grey boxes. Dotted lines represent deviation away from the natural AQ256 biosynthetic pathway. ACP, acyl carrier protein; CLF, chain length factor; KS, ketosynthase.