MODIFIED TERPENE SYNTHASES AND THEIR USE FOR PRODUCTION OF PSEUDOPTEROSIN INTERMEDIATES AND/OR PSEUDOPTEROSINS

Abstract

The present invention pertains to novel modified terpene synthases and their use for a preparation method for pseudopterosin intermediates and/or pseudopterosins. The method is based on the use of a modified terpene synthase comprising at least one modified amino acid residue, which enables a terpene synthase-catalyzed increased production of pseudopterosin intermediates and/or pseudopterosins from Geranylgeranyl pyrophosphate as starting material. The new modified terpene synthase of this invention allow the production of pseudopterosin intermediates, such as Isoelisabethatriene A, Isoelisabethatriene B, Erogorgiaene, or Seco-Pseudopterosin and/or the production of pseudopterosins, such as Pseudopterosin A, in a cost-efficient, economical, and sustainable manner. Also provided are nucleic acids, encoding for the modified terpene synthases of this invention, as well as expression vectors capable of expressing said nucleic acids and host cells comprising the same.

Claims

1. A modified terpene synthase comprising at least one modified amino acid residue as compared to an amino acid sequence corresponding to an unmodified wild type terpene synthase according to any one of SEQ ID NOs: 1 to 5, wherein said at least one modified amino acid residue is located in an α-helix structure being part of, or close to, an active site pocket of the terpene synthase, and wherein said at least one modified amino acid residue is an amino acid with a hydrophobic side chain and/or an amino acid with a polar uncharged side chain.

2. The modified terpene synthase according to claim 1, wherein the modified terpene synthase has at least 75% sequence identity to the unmodified wild type terpene synthase according to any one of SEQ ID NOs: 1 to 5.

3. The modified terpene synthase according to claim 1, wherein the modified terpene synthase catalyzes the production of at least one pseudopterosin intermediate and/or the production of at least one pseudopterosin; from Geranylgeranyl pyrophosphate (GGPP) in a host cell in an amount that is greater than the amount of said pseudopterosin intermediate and/or said pseudopterosin produced from GGPP by the unmodified wild type terpene synthase having the amino acid sequence according to any one of SEQ ID NOs: 1 to 5 in the same host cell and under the same conditions, and/or wherein the modified terpene synthase catalyzes the production of at least one side product from GGPP in a host cell in an amount that is smaller than the amount of said side product produced from GGPP by the unmodified wild type terpene synthase having the amino acid sequence according to any one of SEQ ID NOs: 1 to 5 in the same host cell and under the same conditions.

4. The modified terpene synthase according to claim 1, wherein the terpene synthase is a Hydropyrene synthase (HpS) comprising the amino acid sequence according to SEQ ID NO: 1, a class I terpene synthase from Streptomyces melanosporofaciens comprising the amino acid sequence according to SEQ ID NO: 2, a Diterpene synthase comprising the amino acid sequence according to SEQ ID NO: 3, a Trichodiene synthase comprising the amino acid sequence according to SEQ ID NO: 4, or a Clavulatriene synthase comprising the amino acid sequence according to SEQ ID NO: 5.

5. The modified terpene synthase according to claim 1, wherein said at least one modified amino acid residue is a substitution of a wild type amino acid residue selected from: (i) methionine at position 71, (ii) methionine at position 75, (iii) glycine at position 182, (iv) histidine at position 184, (v) methionine at position 300, and (vi) methionine at position 304, in the amino acid sequence of the unmodified wild type HpS according to SEQ ID NO: 1, or wherein said at least one modified amino acid residue is a substitution of a wild type amino acid residue located at an equivalent position of any of (i) to (vi) in the amino acid sequence of an unmodified wild type terpene synthase according to any one of SEQ ID NOs: 2 to 5.

6. The modified terpene synthase according to claim 1, wherein the modified terpene synthase comprises at least one substitution selected from the group consisting of: (i) a substitution of methionine for tyrosine at position 71, (ii) a substitution of methionine for phenylalanine at position 75, (iii) a substitution of methionine for leucine at position 75, (iv) a substitution of glycine for alanine at position 182, (v) a substitution of glycine for phenylalanine at position 182, (vi) a substitution of histidine for alanine at position 184, (vii) a substitution of histidine for phenylalanine at position 184, (viii) a substitution of methionine for isoleucine at position 300, (ix) a substitution of methionine for isoleucine at position 304, (x) a substitution of methionine for threonine at position 304, and (xi) a substitution of methionine for cysteine at position 304, in the amino acid sequence of the unmodified wild type HpS according to SEQ ID NO: 1, or wherein said modified terpene synthase comprises at least one substitution at an amino acid residue located at an equivalent position of any of (i) to (xi) in the amino acid sequence of an unmodified wild type terpene synthase according to any one of SEQ ID NOs: 2 to 5.

7. The modified terpene synthase according to claim 1, wherein the amino acid sequence of said modified terpene synthase further comprises one or more amino acid deletions, substitutions, and/or additions at positions other than at position 71, 75, 182, 184, 300, and/or 304 according to the amino acid sequence of the unmodified wild type terpene synthase according to SEQ ID NO: 1, or other than the at least one substitution at said equivalent position of an unmodified wild type terpene synthase according to any one of SEQ ID NOs: 2 to 5.

8. The modified terpene synthase according to claim 1, wherein the modified terpene synthase comprises an amino acid sequence according to any one of SEQ ID Nos. 8 to 18, or wherein the modified terpene synthase has at least 75% sequence identity to an amino acid sequence according to any one of SEQ ID Nos. 8 to 18 or wherein the modified terpene synthase consists of an amino acid sequence according to any one of SEQ ID Nos. 8 to 18.

9. A nucleic acid, encoding a modified terpene synthase according to claim 1, or an expression vector capable of expressing said nucleic acid.

10. A recombinant host cell comprising the modified terpene synthase according to claim 1, a nucleic acid encoding a modified terpene synthase according to claim 1, or an expression vector capable of expressing said nucleic acid.

11. A method for producing a modified terpene synthase according to claim 1, the method comprising culturing a host cell that comprises the modified terpene synthase according to claim 1, or expresses a nucleic acid encoding a modified terpene synthase according to claim 1, or an expression vector capable of expressing said nucleic acid and isolating the modified terpene synthase from the host cell or its culture medium.

12. (canceled)

13. A method for producing at least one pseudopterosin intermediate, and/or for producing at least one pseudopterosin, the method comprising the steps of: a) Providing an intermediate generated from a Geranylgeranyl pyrophosphate (GGPP); b) Providing a modified terpene synthase comprising at least one modified amino acid residue as compared to the amino acid sequence corresponding to an unmodified wild type terpene synthase according to any one of SEQ ID NOs: 1 to 5, wherein said at least one modified amino acid residue is located in an α-helix structure being part of or close to an active site pocket of the terpene synthase, and wherein said at least one modified amino acid residue is an amino acid with a hydrophobic side chain and/or an amino acid with a polar uncharged side chain, and c) Destabilizing the intermediate of step a) by said at least one modified amino acid residue of the modified terpene synthase, thereby producing at least one pseudopterosin intermediate and/or at least one pseudopterosin.

14. The method according to claim 13, wherein said method further comprises the step of modifying said at least one pseudopterosin intermediate and/or said at least one pseudopterosin, wherein said modifying comprises a modification selected from a functionalization, oxidation, hydroxylation, methylation, glycosylation, lipid-conjugation, and combinations thereof.

15. A pseudopterosin intermediate and/or a pseudopterosin produced by a method according to claim 13.

16. A modified terpene synthase according to claim 1, wherein said amino acid with a hydrophobic side chain is alanine, valine, isoleucine, leucine, methionine, phenylalanine, tyrosine, or tryptophan, and/or wherein said amino acid with a polar uncharged side chain is threonine, cysteine, asparagine, glutamine, or serine.

17. The modified terpene synthase according to claim 3, wherein said modified terpene synthase catalyzes the production of Elisabethatriene, Isoelisabethatriene A, Isoelisabethatriene B, Erogorgiaene, Seco-Pseudopterosin, and/or Pseudopterosin A.

18. The method according to claim 3, wherein said side product is Hydropyrene (HP) or Hydropyrenol (HP-ol).

19. The method according to claim 13, wherein said amino acid with a hydrophobic side chain is alanine, valine, isoleucine, leucine, methionine, phenylalanine, tyrosine, or tryptophan, and/or wherein said amino acid with a polar uncharged side chain is threonine, cysteine, asparagine, glutamine, or serine.

20. The method according to claim 13, wherein the intermediate that is produced is Elisabethatriene, Isoelisabethatriene A, Isoelisabethatriene B, Erogorgiaene, or Seco-Pseudopterosin and/or the pseudopterosin is Pseudopterosin A.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0113] The figures show:

[0114] FIG. 1 shows structures of elisabethatriene (A), IE A (B), IE B (C), HP (D) and HPol (E); B (12%), C (9%), D (52%) and E (26%) are products of the hydropyrene synthase (HpS) (percentage of product spectrum of wild-type HpS in brackets; carbon numbering as previously described by Kohl and co-workers 4).

[0115] FIG. 2 shows early biosynthetic intermediates leading to pseudopterosins; (A) endogenous coral pathway of pseudopterosin production starting with elisabethatriene via erogorgiaene and seco-pseudopterosins; (B) Proposed pathway using HpS from S. clavuligerus encompassing isoelisabethatriene A; R1,2=sugar moiety.

[0116] FIG. 3 shows conserved motifs of class I terpene synthases with their respective catalytic function (Mg.sup.2+-coordinating residues in bold) and the corresponding amino acid sequences for CotB2 and HpS.

[0117] FIG. 4 shows (A) Variants linked to respective mutation sites (white: variants with wild type-like product spectrum; dark grey: inactive variants; light grey: variants with altered product spectrum); (B) Secondary structure of HpS with highlighted mutation sites. (C) Secondary structure of HpS with highlighted mutation sites, tilted 90° compared to the structure in B. (D) Primary structure and secondary structure elements of HpS (cylinders and lines indicate alpha helices and beta sheets, respectively; dots indicate mutation sites, wherein the code corresponds to the amino acid code of (C) and (D); highlighted parts of the primary structure show conserved motifs (DDXXD, NSE; WXXXXXRY).

[0118] FIG. 5 shows the relative proportion of the main products of the reaction catalysed by HpS and its variants (displayed as percental ratio of the areas of the respective GC-FID product peaks); HpS variants displayed in order of increasing IE A content.

[0119] FIG. 6 shows a structural model of catalytically active residues in HpS. Distance of 75M to catalytically important carbocations C1 (5.0 Å), C11 (4.7 Å) and C7 (7.6 Å), as well as to 71M (6.5 Å; putatively active as dative bond). The numbering of carbon atoms in the intermediate is based on the numbering of GGPP.

[0120] FIG. 7 shows important cyclisation intermediates; (A) GGPP; (B) C10 carbocation intermediate; (C) C1 carbocation intermediate (D): C7 carbocation intermediate; numbering of carbon atoms based on numbers of GGPP.

[0121] FIG. 8 shows (A) Lipase-mediated epoxidation of IE B to the new natural product 1R-epoxy-5,14-elisabethadiene. (B) Lipase-mediated IE A specific conversion of IE A to (−)-erogorgiaene.

[0122] FIG. 9 shows the postulated biosynthetic pathway of pseudopterosin biosynthesis, leading from the geranylgeranyl diphosphate to pseudopterosin via elisabethatrienol and erogorgiaene (R=—H or -sugar).

[0123] FIG. 10: Sequence alignment CotB2 and HpS. The boxes are used as a code for identical and/or similar amino acids, as well as amino acids with no corresponding match. The following code is used: Identical amino acids are highlighted by the following . Highly similar amino acids are highlighted by the following . Similar amino acids are highlighted by the following . Amino acids with no corresponding match are highlighted by the following .

[0124] FIG. 11: Sequence alignment of HpS and ABS. Conserved residues are highlighted by a squared box on top of the aligned sequences.

EXAMPLES

[0125] Certain aspects and embodiments of the invention will now be illustrated by way of example and with reference to the description, figures and tables set out herein. Such examples of the methods, uses and other aspects of the present invention are representative only, and should not be taken to limit the scope of the present invention to only such representative examples.

The examples show:

Example 1: HpS Model-Based Mutagenesis Strategy

[0126] Previous studies reported the formation of early pseudopterosin precursors IE A and B by wild-type (wt) HpS, but with low yields. Initial in vitro studies with HpS revealed a plausible cyclisation mechanism for GGPP conversion towards the products IE A, IE B, HP and HP-ol (FIG. 1). None of these previous studies specifically aimed for selective production of pseudopterosins' precursors IE A and IE B.

[0127] Knowledge-based HpS structure-function studies require a model to delineate a consolidated mutagenesis strategy. Thus, a homology model of the closed complex of HpS synthase was generated by applying the Web tool I-Tasser (https://zhanglab.ccmb.med.umich.edu/I-TASSER/). The predicted structure was further analysed and modified within the environment of UCSF Chimera software package including Modeller software package for comparative modelling (http://www.cgl.ucsf.edu/chimera/). As previously described by Hirte et al., all substrate docking studies were predicted by AutoDock Vina..sup.5,6 For comparative alignment of secondary structure of terpene synthases HHPred applying HMM/HMM comparisons and Ali2D including PSIPRED and MEMSAT software package was used..sup.7

[0128] While class I terpene synthases, such as HpS, share low primary sequence similarity, these enzymes display a significant homology in secondary and tertiary structural features, forming a common α-barrel protein scaffold. Class I terpene synthase catalysis is primed by initial binding and orientation of GGPP via its diphosphate (PP) moiety to a conserved Mg.sup.2+ triade in the active site, characterised by the canonical (DDXX(X)D) motif, and which is located in the centre of the α-barrel. Substrate binding initiates active site closure by an induced fit mechanism and subsequent Mg.sup.2+-mediated PP hydrolysis, generating a highly reactive, priming carbocation. Solvent water is expelled during active site closure creating a hydrophobic microenvironment that prevents an uncontrolled nucleophilic attack on the carbocation. Moreover, specific amino acid residues lining the active site also pre-shape the priming carbocation, thereby significantly influencing the terminal terpene product profile. The inventors reasoned that for subsequent site directed mutagenesis of residues in proximity (3-8 Å) to the docked substrate replacement by more polar or more spacious non-polar residues should allow for quenching of the carbocation intermediate and restrict free folding of the HP skeleton.

[0129] The subsequent intramolecular carbocation rearrangement cascade and terminal cyclization can then commence through C1-C6-, C1-C7-, C1-C10-, C1-C11-, C1-C14- or C1-C15-bond forming reactions, which are modulated by the relative double bond reactivity of the priming carbocation. In addition to the inherent carbocation reactivity, the local electrostatic environment created by the substrate-derived PP moiety, as well as transient electronic and ionic interactions with amino acids of the active site, drive and control successive carbocation rearrangements along the reaction trajectory towards an enzyme specific terminal product profile. Specifically, terminal cyclisation is induced by amino acid-mediated deprotonation or addition of a water molecule to the final carbocation. These concerted enzyme-substrate interactions facilitate an intense diversity of stereochemically complex diterpene macrocycles, all being derived from the universal precursor GGPP.

[0130] As no HpS crystal structure is available, a homology model was constructed, employing the high resolution crystal structure of the taxonomically and secondary structure-related.sup.7 (FIG. 10) class I terpene synthase CotB2 (PDB-ID 6GGI) from Streptomyces melanosporofaciens as a template. The Streptomyces-derived diterpene synthase CotB2, which converts GGPP via cyclooctat-9-en-7-ol to the anti-inflammatory agent cyclooctatin, belongs to the best characterized diterpene synthases to date. CotB2, catalysing the cyclization of GGPP to tricyclic cyclooctat-9-en-7-ol, has been subject to extensive mutagenesis studies. To date, CotB2 is the only class I (di)terpene synthase, for which a closed, catalytically relevant structure containing a trapped diterpene reaction intermediate, is available. Computational interrogation of this unique structure in synergy with extensive QM/MD simulations provided detailed insights into the dynamic CotB2 reaction mechanism, highlighting a concerted network of catalytically essential amino acid lining its active site. Therefore, CotB2 represents an ideal template for a comprehensive HpS structure-function analysis.

[0131] An initial CotB2/HpS structural comparison indicated that all catalytically relevant class I structural motifs are conserved (FIG. 3). However, the canonical class I DDXXD motif, responsible for initial binding and orientation of the substrate's (GGPP) diphosphate (PP) moiety in the active site, is altered in both CotB2 and HpS to DDXD (.sup.110DDMD) and DDXXXD (.sup.82DDRAID), respectively. Interestingly, such extensive modifications of the highly conserved DDXXD motif are rare in class I terpene synthases (TPSs). However, the addition of a single amino acid (X) has also been reported in other TPSs, such as selina-3,7(11)-diene synthase (.sup.82DDGYCE) and (+)-T-muurolol synthase (.sup.83DDEYCD). Other active site motifs are also conserved in HpS and CotB2 (FIG. 3 and FIG. 10), including the NSE triad and the class I TPS specific WXXXXXRY motif. The CotB2 and HpS-specific amino acid sequences for each catalytically relevant motif are listed in FIG. 3.

[0132] Interestingly, a more extensive HpS structural interrogation revealed the distinctive presence of five unique methionine residues (.sup.71M, .sup.75M, .sup.188M, .sup.300M and .sup.304M) inside or in the immediate vicinity of the putative HpS active site. A feature that has not been reported or experimentally evaluated for any TPS. The catalytic relevance of these residues is largely unknown, although a computational (QM) study of Fusarium sporotrichioides trichodiene synthase (TdS) implicates a methionine residue in interactions with TdS-specific carbocation reaction intermediates. Thus, these methionine residues were included in the mutational strategy to elucidate HpS structure-function relationships to selectively establish the biosynthetic pseudopterosin precursors IE A and B as the main GGPP cyclisation products.

[0133] Relevant active site residues selected for mutagenesis are listed in Table 3.

TABLE-US-00003 TABLE 3 Comparison of HpS and CotB2 active site residues used to delineate the HpS mutagenesis strategy. Amino acid residues were chosen due to their potential to alter the product range or stabilize the carbocation intermediate. Residues in bold show identical amino acids in HpS and CotB2. HpS Cot B2 HpS Cot B2 L 54 V 80 G 182 D 180 Y 58 S 84 H 184 G 182 M 71 V 99 M 188 W 186 M 75 N 103 M 300 L 281 Y 78 T 106 M 304 N 292 A 79 F 107 W 307 W 288 Y 153 F 156 R 313 R 294 R 179 R 177

[0134] The relevant active site residues selected for mutagenesis listed in table 3 include .sup.307W and .sup.313R residues of the conserved .sup.307WXXXXXRY motif. Conservative substitutions of these residues in CotB2 have previously been shown to modulate the product spectrum.

Example 2: Tailoring E. coli for HpS-Derived Diterpene Production

[0135] An engineered E. coli host harbouring a metabolically balanced two-plasmid terpene production system was employed for HpS expression, which allows for rapid mutagenesis of wt class I HpS and subsequent screening for altered product profiles..sup.8 For terpene extraction, technical grade ethanol, ethyl acetate and hexane were purchased from Westfalen AG (Minster, Germany). For all other purposes, highest purity grade chemicals were used. Acetonitrile, ethyl acetate, hexane, methanol, propionic acid, and media components were obtained from Roth chemicals (Karlsruhe, Germany). Immobilized Lipase B from C. antarctica (CalB), CDCl.sub.3, Benzene-d.sub.6 and urea hydrogen peroxide were purchased from Sigma-Aldrich (St. Louis, USA).

[0136] E. coli strain DH5a was used for plasmid generation and cloning. It was cultivated at 37° C. in Luria-Bertani medium. Terpenes were produced with E. coli strain ER2566. During shaking flask experiments E. coli ER2566 was grown at 23° C. in either Luria-Bertani or R-Media supplemented with 30 g L.sup.−1 glucose and 5 g L.sup.−1 yeast extract. In case of fermentation experiments, E. coli ER2566 was cultivated in R-Media supplemented with 30 g L.sup.−1 glycerol and 5 g L.sup.−1 yeast extract. Chloramphenicol (30 μg mL.sup.−1) and Kanamycin (50 μg mL.sup.−1) were added as required.

[0137] All genes encoding diterpene synthase (Uniprot: SCLAV_p0765) from S. clavuligerus (ATCC 27064) were cloned into pACYC-based expression vector system. All genes and primers were synthesized by Eurofins Genomics GmbH (Ebersberg, Germany). Genes were codon-optimised for E. coli by use of the GeneOptimizer™ software.

[0138] Overnight pre-culture was used to inoculate the fermenters of a DASGIP® 1.3 L parallel reactor system (Eppendorf AG, Germany) (OD.sub.600=0.1). Cultivation temperature was kept constant at 23° C. Stirring velocity, airflow, oxygen content and feeding protocol were set as previously described..sup.8 Feed solution consisted of 600 g L.sup.−1 glycerol, 5 g L.sup.−1 yeast extract, 35 g L.sup.−1 collagen, 20 g L.sup.−1 MgSO.sub.4, 0.3 g L.sup.−1 Thiamine-HCl, 5 ml L.sup.−1 1M Ammonium iron(III) citrate, 20 ml L.sup.−1 100× trace elements solution (pH=7.0) as described previously..sup.8 To monitor terpene production, samples were taken at different time points.

[0139] Co-transformation of the plasmid carrying the codon-optimised HpS gene, together with a separate plasmid harbouring bottleneck enzymes of E. coli terpene biosynthesis, led to efficient production of functional HpS. The balanced carbon flux and terpene precursor supply in the tailored E. coli host allowed native HpS to efficiently convert GGPP to HP, HP-ol, IE A and IE B (total terpene yield 55.56 f 2.01 mg/l, FIG. 2). The HpS-harbouring E. coli production system provided rapid and simplified cultivation with high product yields.

Example 3: Diterpene-Directed Product Screening of HpS Variants

[0140] The inventors reasoned that for subsequent site directed mutagenesis of residues in proximity (3-8 Å) to the docked substrate replacement by more polar or more spacious non-polar residues should allow for quenching of the carbocation intermediate and restrict free folding of the HP skeleton. A tailored design of the active site of an enzyme allows the generation of a hydrophilic environment, thereby enabling water molecules access to the active site. As a result, the water molecules in the active site can quench a carbocation, whereby a hydroxyl group at the active site is generated. A library of HpS mutants (FIG. 4) was expressed in E. coli, and diterpene products were extracted from the cultivation broth..sup.8 To extract products and other compounds from shaking flask experiments and samples taken from fermentation units an equivalent volume of solvent (ethanol:ethyl acetate:hexane=1:1:1) was added to the culture broth and mixed for 2 h at room temperature. The solution was centrifuged for 5 min at 8000 rpm to separate the upper organic phase to be analysed by GC-FID and GC-MS.

[0141] The whole fermentation broth was extracted by addition of the same volume of ethanol. The first process step was carried out on a rotary shaker (80 rpm) at 20° C. for 12 h. Subsequently, the extraction mixture was centrifuged for 20 min at 7000 rpm to separate the supernatant from the cell debris. Via addition of ethyl acetate (50% of supernatant volume) a second extraction step on the rotary shaker (80 rpm) was started (20° C. for 5 h). After 5 hours the same amount of hexane was added, and the extraction process continued for further 2 h. Finally, the phases were separated by a separation funnel and the organic phase was evaporated.

[0142] The flash chromatography system PLC 2250 (Gilson, USA) allowed for a separation between the fatty acid residues and the terpene fraction. To this end, the solvents hexane (A) and ethyl acetate (B) were pumped with a flowrate of 10 mL min.sup.−1 at room temperature over a Luna 10 μm Silica (2) 100A column. The following gradient was applied: 100% A for 15 min, increasing B in one step to 100%, holding 100% B for 15 min and then applying 100% A for 30 min. Eluted compounds were analysed by a diode array and an ELSD detector which was flushed with nitrogen gas at 40° C. Fractions of interest were reduced by nitrogen flow to approximately 2 ml. Terpene concentration was measured using GC-FID. Fractions containing IEs were mixed with acetonitrile (ACN). Subsequently hexane and ethyl acetate were carefully evaporated until only acetonitrile remained.

[0143] To further purify the IEs dissolved in ACN, the samples were injected into an Ultimate 3000 UHPLC system (Thermo Scientific, USA) containing a binary pump, a diode array detector, an automated fraction collector, and a Jetstream b1.18 column oven. Separation of isoelisabethatrienes from hydropyrenol, hydropyrene and other terpene derivatives (maximum terpene content of 25 mg) was carried out at 30° C. oven temperature using H.sub.2O (A) and ACN (B) as solvents with a flowrate of 2.2 mL min.sup.−1 on a NUCLEODUR® C18 HTec 250/10 mm 5 μm column with a guard column NUCLEODUR® C18 HTec 10/8 mm and guard column holder 8 mm (Macherey-Nagel GmbH & Co. KG, Germany). The separation gradient started with 30% B for 5 min, then it increased within 55 min to 1000% B. 100% B was maintained for further 60 min.

[0144] To separate IE A from B, the same HPLC system was equipped with a NUCLEODUR® C18 Isis 250/10 mm 5 μm column with guard column NUCLEODUR® C18 Isis 10/8 mm and guard column holder 8 mm (Macherey-Nagel GmbH & Co. KG, Germany). The mobile phase consisted of H.sub.2O (A) and MeOH (B). The following program was applied: 30% B for 5 min, then increase to 100% B within 55 min to remain for another 35 min. The oven temperature was set to 30° C. After liquid-liquid extraction with hexane purified compounds were stored at −20° C.

[0145] Analysis and quantification of terpenes was performed using a Trace GC-MS Ultra system with DSQII (Thermo Scientific, USA). One microliter (1/10 split) of sample was injected by a TriPlus auto sampler onto a SGE BPX5 column (30 m, I.D 0.25 mm, film 0.25 μm) with an injector temperature of 280° C. Helium was used as carrier gas with a flow rate of 0.8 ml/min. Initial oven temperature was set to 50° C. for 2 min. The temperature was increased to 320° C. at a rate of 10° C./min and then held for 3 min. MS data were recorded at 70 eV (EI). Masses were recorded in positive mode in a range between 50 and 650. GC-FID analysis was performed in the same way.

[0146] Compounds for NMR studies were dissolved either in CDCl.sub.3 or benzene-d.sub.6. .sup.13C NMR spectra were measured with a Bruker Avance-III 500 MHz spectrometer equipped with a cryo probe head (5 mm CPQNP, 1H/13C/31P/19F/29Si; Z-gradient). .sup.1H NMR spectra as well as the 2D experiments (HSQC, HMBC, COSY, NOESY) were obtained on an Avance-I 500 MHz system with an inverse probehead (5 mm SEI, .sup.1H/.sup.13C; Z-gradient). The temperature was set to 300 K. Resulting data was processed and analysed by TOPSPIN 3.0 or MestreNova. Chemical shifts were given in ppm relative to CDCl.sub.3 (S=7.26 ppm for .sup.1H and 6=77.16 ppm for .sup.13C spectra) or benzene-d-.sub.6 (S=7.16 ppm for .sup.1H and 6=128.06 ppm for .sup.13C spectra). The total terpene yields of catalytically viable HpS mutants was comparable to that of the wild type enzyme. Subsequently, all enzyme mutants were evaluated for variations in their product spectrum with respect to wt HpS. A specific focus was given to enhanced IE A and/or B generation.

[0147] Mutations Y153A, Y153F, G182K, and W307F did not affect the product spectrum. In contrast, mutations L54A, Y58A, M71R, M71P, M71G, Y78A, A79F, Y153R, R179A, M188G, M188A, M188K, M188Y, M300G, M304D, W307A, W307G and R313A inactivated HpS, indicating that the mutated amino acids are essential for catalysis. Most notably, variants M71Y, M75F, M75L, G182A, G182F, H184A, H184F, M300I, M304I and M304C displayed an altered product spectrum with respect to wt HpS (FIG. 4). Mutation .sup.179R, .sup.307W and .sup.313R either resulted in a non-active variant or displayed the native HpS product spectrum. Therefore, .sup.179R, .sup.307W and .sup.313R are likely to be essential for catalysis, which is consistent with previous reports for CotB2.

[0148] Interestingly, mutants M71Y, M75F, M75L, M300I, and M304C targeting HpS-specific methionine residues displayed the most pronounced shifts in the diterpene product profile when compared to wt HpS (FIG. 5). Mutations M300I and M304C lead to a decrease in IE A production and a concomitant increase in IE B. A more significant effect is observed for mutations M71Y, M75F and M75L. Each variant displays a significantly enhanced yield of IE A and B, with a concomitant reduction in HP and HPol production. The most prominent effect is observed for mutations of M75. Notably, M75F showed the highest IE B yield, whereas M75L displayed the highest IE A yield.

[0149] Since IE A is a biosynthetic pseudopterosin intermediate.sup.24, its increased yield in the M71Y, M75F and M75L variants is highly encouraging for the ongoing effort to generate a sustainable pseudopterosin production platform. Mutants M75F and M75L shifted the product spectrum towards an IE isomer as their major product. M75L is the most promising mutation for pseudopterosin production due to its particularly high IE A yield. Therefore, this HpS variant was termed isoelisabethatriene synthase (IES) and used for all downstream efforts to generate advanced pseudopterosin intermediates.

Example 4: In Silico Driven Mechanistic Considerations for IE Generating Mutants

[0150] As mutations of .sup.71M and .sup.75M significantly modulated the HpS product spectrum towards IE production it was essential to evaluate the chemical mechanisms that induce these effects. Interestingly, no methionine-carbocation interactions within a distance of ˜8 Å have been suggested to be important in the catalytic mechanism of CotB2. Notably, the residue equivalent to .sup.75M in HpS is .sup.103N in CotB2. The latter was proposed to coordinate a water molecule that terminates the CotB2 cyclization cascade or form a dipole-charge interaction during the cyclization reaction. Interestingly, the N103A variant CotB2 featured a 3,7,12-dolabellatriene as the major cyclisation product. CotB2 has one methionine (.sup.189M) that lines its active site but whose replacement by cysteine does not directly interfere with catalysis. The only report that describes the effect of a metionine on terpene synthase catalysis is a computational study of trichodiene synthase, in which 73M is stabilizing selected carbocation conformations via a dative sulphur-carbocation bond. Hence, substrate tumbling, and premature deprotonation is prevented. It is thus plausible that methionine residues (especially .sup.75M) in HpS also aid in the stabilization of carbocation intermediates, providing a route which is crucial for opening a distinct reaction pathway (e.g. en-route to HP derivatives; FIG. 6). Therefore, mutagenesis of .sup.71M and .sup.75M in HpS can reroute GGPP cyclisation to either HP or IE derivatives.

[0151] The initial cyclisation step of the HpS-specific mechanism for GGPP cyclisation to HP, HP-ol, IE A and IE B comprises a 1,10-ring closure, which generates a carbocation at C11. Subsequently, the carbocation (FIG. 7) can be channeled in two distinct reaction pathways. The primary route proceeds via the stable Markovnikov C11 carbocation followed by a 1,3-hydride shift to form the less stable anti-Markovnikov C1 carbocation, which ultimately provides HP or HPol, respectively. In contrast, the formation of IE derivatives requires a 1,3-hydride migration, forming a carbocation at position C7, which leads either to IE A via a 1,2-hydrid shift and deprotonation or to IE B via simple deprotonation. For effective biotechnological pseudopterosin precursor supply, it is crucial to evolve the product spectrum of HpS towards a specific production of IE isomers. As the C11 carbocation is the essential intermediate for changing the preferred pathway to the desired IE A, it is crucial to prevent the 1,3-hydrid shift towards the less stable anti-Markovnikov C1 carbocation.

[0152] The route towards HP derivative formation requires that the anti-Markovnikov C1 carbocation is stabilized within the active site. In wild-type HpS .sup.75M is in close proximity (˜5.0 Å) to both C1 and C11 carbons, and it is plausible that in the C1:C11 carbocation transition the proximal .sup.71M plays a stabilizing role via a dative Met-Met interaction (FIG. 6). Hence, the performed mutagenesis of .sup.71Met and .sup.75Met are likely destabilizing the key C1 intermediate and therefore showing a significant product shift towards Isoelisabethatriene A and B.

Example 5: Identification of Hydroxylated IE Derivatives

[0153] While IE A and B are primary biosynthetic pseudopterosin precursors, especially the oxidised IE A forms represents advanced pseudopterosin precusors. Culture broth extracts of E. coli expressing IES were evaluated for the presence of oxidised IE derivatives using a GC-MS based screening method. Inspection of GC-MS spectra identified a compound with MS spectral similarity to IE but with extended retention time (retention time (Rt) (IE A): 20.46 min; Rt (IE B): 20.87 min; Rt (unknown compound): 22.28 min) and an parent ion mass (m/z) of 290, indicating the presence of a hydroxyl-moiety (data not shown). The putative hydroxylated IE derivative can potentially arise by controlled water capture within the HpS active site, which facilitates carbocation quenching along the reaction trajectory. Analogous data have been reported for the class I germacradien-4-ol sesquiterpene synthase.

[0154] Moreover, the presence of the aromatized IE derivative, erogorgiaene, a key intermediate in coral-based pseudopterosin biosynthesis, was confirmed by comparison with an authentic GC-MS standard isolated from A. elisabethae coral tissue. However, as erogorgiaene could not be detected, when the E. coli extract was analysed directly after the extraction process, it is plausible that oxygen exposure of the analysed extract initiated an oxidative transformation of IE A or B to erogorgiaene. As erogorgiaene is an advanced intermediate in pseudopterosin biosynthesis, the current data are consistent with previous reports indicating that hydroxylated elisabethatriene derivatives are direct erogorgiaene precursors in the pseudopterosin biosynthetic pathway.

Example 6: Chemo-Enzymatic IE A and B Oxidation—a Route to Advanced Pseudopterosin Precursors

[0155] As erogorgiaene formation is a crucial step in pseudopterosin biosynthesis, its definitive biosynthetic origin was probed by development of a selective in-vitro chemo-enzymatic oxidation approach with IE A and B as substrates. Recently, selective functionalization of the macrocyclic diterpene hydrocarbons dolabellatriene and taxadiene via lipase-mediated oxidation reactions has been reported..sup.8 Consequently, in a lipase-mediated and chemo-enzymatic assay IE A and B were oxidized to establish whether oxyfunctionalization, and therefore activation of the IE hydrocarbon skeleton, is part of the pseudopterosin biosynthetic pathway.

[0156] 250 μg mL.sup.−1 IE A or B was mixed in 5 mL ethyl acetate with 1 μl concentrated propionic acid, 2 mg mL.sup.−1 immobilised CalB and 2 mg mL.sup.−1 urea-hydrogen peroxide. Reaction was performed at 22° C. and 1000 rpm in a thermo shaker (Eppendorf AG; Germany). At different time points, samples were taken to monitor the reaction progress by GC-MS analysis.

[0157] CalB reaction was stopped at appropriate time points by separation of immobilised CalB from reaction mixture by filtration. The remaining solution was diluted with hexane (1:4) and filtrated through filter paper. Final product purification occurred in two steps:

[0158] In case of IE A, the reaction mixture was first purified by flash chromatography. Hence, the solvents hexane (A) and ethyl acetate (B) were applied at 10 mL min.sup.−1 to a Luna 10 μm Silica (2) 100A column. After 10 min 100% A, solvent B was increased within 5 min to 100%. Finally, another 30 min the system was operated with 100% A. Subsequently, the fractions were further purified by a preparative HPLC system equipped with a NUCLEODUR® C18 HTec 250/10 mm 5 μm column with Guard column NUCLEODUR® C18 HTec 10/8 mm and guard column holder 8 mm (Macherey-Nagel GmbH & Co. KG, Germany). The method used an oven temperature of 30° C. and the solvents H.sub.2O (A) and ACN (B) at a flowrate of 2.2 mL min.sup.−1. The gradient started with 30% B for 5 min to increase afterwards to 100% B within 55 min. This solvent level was maintained for 60 min.

[0159] When purifying products originating from the reaction using IE B, the process also starts with a flash chromatography. This time the gradient was altered to: 1% B for 10 min, increase of B within 41 min to 40%, stay at 40% B for 1 min, further increase to 100% B within the next 3 min and final remain at this level for 10 min. Afterwards the column was cleaned with 100% A for 30 min. Again, the second step consists of a preparative HPLC purification. The solvents remain H.sub.2O (A) and ACN (B), but the following gradient was used: 40% B for 5 min, increase of B to 100% in 30 min and a stay at 100% B for 60 min.

[0160] To ensure future process scalability under economic boundary conditions, the inventors employed the industrially well-established lipase Cal B. The mild lipase-mediated IE oxidation was carried out in ethyl acetate in the presence of urea-hydrogen peroxide with propionic acid, which generates the reactive oxidant. The reaction was initiated by in situ generation of per-oxo carboxylic acid as the reactive oxidant, which targets olefinic IE bonds either in re- or si-face conformations to afford a racemic mixture of oxidised products. Reaction progress was monitored by GC-FID analysis, while GC-MS was applied to identify IE A and B specific oxidation products (FIG. 8).

Example 7: Identification of the IE B-Specific Oxidation Products and IE A-Specific Conversion to Erogorgiaene

[0161] While GC-FID allowed kinetic reaction profiling, parallel GC-MS analysis indicated that the lipase-mediated IE B oxidation resulted in a time dependent formation of IE B mono- (m/z 288) and IE B diepoxides (m/z 304), respectively. To enhance product selectivity towards formation of the IE B mono-epoxide, the reaction was terminated after 120 min (yield of 41%). Subsequently, a 2D-HPLC protocol allowed for 1D and 2D NMR spectroscopy-based structure elucidation of the putative IE B-derived mono-epoxide. .sup.13C NMR analysis provided characteristic epoxide-type chemical shifts for C1 and C9 at 62.66 and 64.21 ppm, respectively. Comprehensive NMR signal assignment confirmed the IE B monoepoxide as the new natural product 1R-epoxy-5,14-elisabethadiene (EED, FIG. 8).

[0162] The epoxidation of the IE B diterpene carbon skeleton enables various downstream biotechnological and chemical functionalization strategies to access a diversified chemical space. As most bioactive terpenoids contain at least one functional group, subsequent modification of EED and other IEs is a fundamental step towards the sustainable generation of new pharmaceutical agents. Various approaches for hydroxyl group functionalization at the bicyclic pseudopterosin carbon skeleton have been applied to generate pseudopterosin derivatives and pseudopteroxazoles, which both were active against M. tuberculosis and other pathogens.

[0163] Lipase-mediated oxidation rapidly (90 min) transformed IE A into a single new compound (yield: 69%). Synchronous GC-MS analysis indicated that this compound was the aromatic pseudopterosin precursor erogorgiaene (data not shown). For structural confirmation, the putative erogorgiaene was purified via an optimised 2D-HPLC method and subsequently subjected to 1D and 2D NMR spectroscopy. The resulting NMR signals of the purified compound were in agreement with reported data for (+)-erogorgiaene. While NOESY experiments resolved the relative erogorgiaene stereochemistry, the absolute configuration remained elusive. However, the absolute stereochemistry of the primary HpS cyclisation products was previously resolved using isotopically labelled substrates and CD-spectrophotometric cyclisation product detection. The analysis indicated that HpS converts GGPP to the ((−)-IE A enantiomer, while the A. elisabethae coral-derived counterpart constitutes (+)-IE A. Similarly, it was deduced that the lipase-based oxidation of HpS derived (−)-IE A leads to the formation of (−)-erogorgiaene, while the coral-derived compound constitutes the (+)-erogorgiaene enantiomer.

[0164] The rapid (−)-erogorgiaene formation, precluded observation of any epoxidised IE A intermediates via GC-MS. However, mechanistic considerations imply that (−)-IE A oxidation proceeds via initial epoxidation of the C9-C10 double bond, followed by protonation of the resulting epoxide and a subsequent dehydration, which induces a spontaneous ring system aromatization to afford (−)-erogorgiaene.

[0165] This mechanistic sequence is supported by detection of elisabethatriene as well as a transient hydroxylated elisabethatriene derivative in crude A. elisabethae coral extracts. The spontaneous dehydration of the hydroxylated intermediate to erogorgiaene has been proposed as an essential step in the pseudopterosin biosynthesis (FIG. 9). In analogy, the observed chemo-enzymatic conversion of IE A to (−)-erogorgiaene employs the same mechanism. Since erogorgiaene has potent activity against M. tuberculosis (with reported MICs as low as 32.25 μg/ml), the current technology platform can provide a scalable and sustainable access to this interesting natural product. In light of the accelerated evolution of infectious diseases and the lack of new molecular leads for advanced antibiotic therapy, this platform addresses the essential need for preparedness to fight infection epidemics.

REFERENCES

[0166] 1. Newton, C. G. et al. Pseudopterosin synthesis from a chiral cross-conjugated hydrocarbon through a series of cycloadditions. Nature chemistry 7, 82-86; 10.1038/NCHEM.2112 (2015). [0167] 2. Davies, H. M. L. & Walji, A. M. Direct synthesis of (+)-erogorgiaene through a kinetic enantiodifferentiating step. Angewandte Chemie (International ed. in English) 44, 1733-1735; 10.1002/anie.200462227 (2005). [0168] 3. Dixit, M., Weitman M., Gao, J., Major, D. T. Chemical Control in the Battle against Fidelity in Promiscuous Natural Product Biosynthesis: The Case of Trichodiene Synthase. ACS Catal. 2017 Jan. 6; 7(1):812-818. doi: 10.1021/acscatal.6b02584. [0169] 4. Kohl, A. C., Ata, A. & Kerr, R. G. Pseudopterosin biosynthesis-pathway elucidation, enzymology, and a proposed production method for anti-inflammatory metabolites from Pseudopterogorgia elisabethae. Journal of industrial microbiology & biotechnology 30, 495-499; 10.1007/s10295-003-0076-7 (2003). [0170] 5. Trott, O. & Olson, A. J. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. Journal of computational chemistry 31, 455-461; 10.1002/jcc.21334 (2010). [0171] 6. Hirte, M., Meese, N., Mertz, M., Fuchs, M. & Brück, T. B. Insights Into the Bifunctional Aphidicolan-16-ß-ol Synthase Through Rapid Biomolecular Modeling Approaches. Frontiers in chemistry 6, 101; 10.3389/fchem.2018.0010 (2018). [0172] 7. Zimmermann, L. et al. A Completely Reimplemented MPI Bioinformatics Toolkit with a New HHpred Server at its Core. Journal of molecular biology 430, 2237-2243; 10.1016/j.jmb.2017.12.007 (2018). [0173] 8. Hirte, M. et al. From microbial upcycling to biology-oriented synthesis. Combining whole-cell production and chemo-enzymatic functionalization for sustainable taxanoid delivery. Green Chem 20, 5374-5384; 10.1039/c8gc03126f (2018).