High-titer Production of Olivetolic Acid and Analogs in Engineered Fungal Host using a non-Plant Biosynthetic Pathway

Abstract

Microbial synthesis of cannabinoids and related molecules requires access to the intermediate olivetolic acid (OA). While plant enzymes have been explored for E. coli and yeast biosynthesis, moderate yields and shunt product formation are major hurdles. Here, based on the chemical logic to form 2,4-dihydroxybenzoate containing natural products, we discovered a set of fungal tandem polyketide synthases (PKSs) that can produce OA and the related octanoyl-primed derivative sphaerophorolcarboxylic acid (SA) in high titers using the model organism Aspergillus nidulans. This new set of enzymes will enable new synthetic biology strategies to access microbial cannabinoids.

Graphical Abstract

INTRODUCTION

Cannabinoids are a large class of bioactive natural products originally derived from the Cannabis sativa plant that regulate the cannabinoid receptors CB1 and CB2 of the human endocannabinoid system.¹ While the most well-known cannabinoids are Δ⁹ -tetrahydrocannabinol (Δ⁹ -THC) and cannabidiol (CBD), intermediates of the plant cannabinoid biosynthetic pathway such as tetrahydrocannabinolic acid (THCA), cannabidiolic acid (CBDA) and cannabigerolic acid (CBGA), are also bioactive and useful as ingredients in cannabinoid based medicines (CBMs) (Figure 1).^2,3,4 In recent years, there has been significant interest from the synthetic biology community to produce these cannabinoids using microbial and cell-free strategies, because of i) challenges associated with chemical synthesis; ii) inconsistent and relatively low production of cannabinoids from plants; and iii) flexibilities in engineering the pathway to access rare or unnatural cannabinoids.⁵

Figure 1.

Resorcylic acid natural products produced from Cannabis sativa and from fungi. A) C. sativa cannabinoid biosynthetic pathway. The β-resorcylic acid, olivetolic acid (OA), is produced from the olivetolic acid synthase (OAS) and olivetolic acid cyclase (OAC). OA is processed to the final products THCA and analogues. In the absence of OAC, side products (pyrone and olivetol) emerge. B) Tandem fungal iterative polyketide synthases produce resorcylic acid lactones.

To date, these approaches all rely on the plant pathway as shown in Figure 1A. The pathway starts with biosynthesis of the first intermediate olivetolic acid (OA), using two plant enzymes olivetolic acid synthase (OAS) and olivetolic acid cyclase (OAC). OAS elongates the hexanoyl starter unit from hexanoyl-CoA with malonyl-CoA to yield a tetraketide intermediate. In the absence of OAC, shunt products such as the pyrone and decarboxylated olivetol can form. OAC performs a regioselective aldol cyclization and product release to form OA.⁶ OA is then geranylated to form CBGA and is further cyclized by dedicated enzymes to the more advanced cannabinoids. Using the OAS and OAC pair, both E. coli and S. cerevisiae have been engineered to produce OA and advanced cannabinoids.^7,8 Gonzalez and coworkers demonstrated ~80 mg/L of OA can be produced from an E. coli strain that accumulates high levels of hexanoyl-CoA, while Keasling and coworkers demonstrated complete reconstitution of THCA and CBDA in yeast starting with OAS and OAC.^7,8 Despite these advances, using the OAS and OAC in the microbial system suffer from relatively lower yield of OA (in yeast) and shunt product (pyrone and olivetol) accumulation. Furthermore, new entries into the field are limited by the intellectual property restrictions associated with OAS and OAC of plant origin. Therefore, alternative pathways that can afford OA and analogs in microbial host are highly desired.

OA is a 6-alkyl substituted 2,4-dihydroxybenzoic acid, also known as a β-resorcylic acid.^9,10 The β-resorcylate moiety appears frequently in fungal polyketides, including the well-known zearalenone, hypothemycin, and radicicol (Figure 1B).^11.12 In these molecules, the carboxylate group is esterified to form a fused macrolactone, and the resulting molecules are known as resorcylic acid lactones (RALs). The biosynthesis of RALs from fungi has been extensively studied, including contributions from our group.^11,13–17 Two iterative fungal polyketide synthases (PKSs) are required to forge the RAL scaffold: a highly reducing PKS (HRPKS) generates the reduced polyketide unit containing the terminal hydroxyl group that becomes the macrocyclizing nucleophile. The chain is transferred to a nonreducing PKS (NRPKS) and further elongated and undergoes aldol cyclization to form the enzyme bound resorcylic thioester. A fused thioesterase (TE) domain in the NRPKS then performs macrocyclization to release the RAL (e.g. monocillin II in Figure 1B).^16,17 Considerable structural diversity at the C6 position in the RAL can be generated by using different HRPKSs that can synthesize various reduced products.¹¹

Based on the shared β-resorcylate moiety found in OA and in fungal RALs, we hypothesize that fungal biosynthetic pathways that encode a pair of HRPKS and NRPKS may be able to produce OA or related molecules that vary in C6 substitutions. Given the potential to mix-and-match different HRPKS and NRPKS enzymes, such dual-PKS pathways may be engineered to produce different OA derivatives, including those that can lead to rare or unnatural cannabinoids. In this work, we performed genome mining of fungal PKS biosynthetic clusters and identified a set of HRPKS/NRPKS containing pathways that also encodes an unprecedented ψACP (pseudo-acyl carrier protein)-TE enzyme that can release β-resorcylic acids such as OA and sphaerophorolcarboxylic acid (SA). Heterologous expression of the pathway in A. nidulans led to a high-titer production of OA and SA.

RESULTS AND DISCUSSION

Identification of potential OA producing pathways in fungi.

The terminal TE domains in the NRPKSs that produce RALs are responsible for the macrocyclization reaction. In order to produce resorcylic acid instead of RALs, the releasing enzyme must catalyze a hydrolysis reaction instead of esterification. In fungal PKSs, TEs that catalyze hydrolytic release have been characterized and are typically free-standing enzymes.¹⁸ With this in mind, we performed genome mining of sequenced fungal genomes for biosynthetic gene clusters that encode a HRPKS, a NRPKS, and a standalone TE. Among the clusters identified by antiSMASH,¹⁹ one set of homologous clusters satisfied this particular criterion (Figure 2A).

Figure 2.

Genome mining of microbial clusters that can synthesize olivetolic acid and related compounds. A) Homologous clusters were elucidated through genome mining of the ψACP-TE gene. All the homologous clusters found contained a NRPKS, a HRPKS, and a ψACP-TE. Functional ACPs contain the hallmark DSL sequence where the serine is post-translationally modified with phosphopantetheine (pPant) and is therefore activated to shuttle the acyl chain. Because the ψACP sequences do not have that serine residue, they cannot be post-translationally modified with pPant and are therefore proposed to be inactive. B) LC-MS traces of Aspergillus nidulans expressing different biosynthetic clusters and combinations of genes. i) Heterologous expression of Ma_OvaABC from M. anisopliae produced compounds 1–4. ii) Combinatorial biosynthesis of OvaBC with StcJ and StcK fatty acid synthase from A. nidulans selectively produced OA. iii) Heterologous expression of To_OvaABC from T. inflatum produces compounds 1–3; iv) Heterologous expression of Mr_OvaABC from M. rileyi produced primarily compounds 1; v) Heterologous expression of Ti_OvaABC from T. islandicus selectively produces compound 3. C) structures of compounds 1-4.

The ova cluster from Metarhizium anisopliae encodes a typical HRPKS (Ma_OvaA) and a NRPKS (Ma_OvaB) that is not fused to a terminal TE domain. Instead, a didomain enzyme Ma_OvaC containing an N-terminal ACP and a C-terminal TE is present in the cluster. Further sequence analysis of the ACP domain showed the well-conserved DSL triad in all functional ACPs, in which the serine is post-translationally phosphopantetheinylated, is mutated to NQI.^20–22 This suggests the ACP domain is unlikely to carry out the canonical function of acyl chain shuttling, thus the enzyme is designated as a ψACP-TE. Previously, a ψACP-methyltransferase (MT) fusion enzyme was found in a fungal PKS pathway, in which the ψACP facilitates protein-protein interactions between the NRPKS and the ψACP- MT to enable methylation of the growing polyketide intermediate.²³ Hence, we hypothesize the ψACP domain in Ma_OvaC may have a similar role in facilitating the catalytic function of the TE domain on a PKS-bound intermediate. The M. anisopliae cluster contains additional genes encoding a transcriptional factor and a flavin-dependent monooxygenase. Alignment of homologous clusters from various fungal species (Metarhizium rileyi, Talaromyces islandicus, and Tolypocladium inflatum) showed that HRPKS, NRPKS, and ψACP-TE are conserved (Figure S1), including the inactivated ACP triad (Figure 2A). None of these clusters have been characterized and no product has been reported in the literature. Based on these analyses, we predict that the trio of HRPKS, NRPKS, and ψACP-TE will make resorcylic acids that are structurally related to OA.

Heterologous expression of candidate pathways led to OA and SA production.

To examine the product profile of the ψACP-TE containing pathways, we heterologously expressed Ma_OvaA, B, and C in the model fungus Aspergillus nidulans A1145 ΔSTΔEM strain.²⁴ This strain has been used in reconstitution of fungal biosynthetic pathways, and contains genetic deletions that inactivated biosynthesis of endogenous metabolites sterigmatocystin and emericellamide B.²⁴ Each of the three genes was cloned into separate episomal vectors, transformed into A. nidulans, and the resulting transformants were grown on CD-ST agar plates (see Methods). Following 5 days of growth in CD-ST, the sample media were extracted and analyzed by liquid chromatography-mass spectrometry (LC-MS).

Coexpression of Ma_OvaA-C in A. nidulans produced four metabolites 1-4 (Figure 2B). The molecular weights (MWTs) as indicated by LC-MS for these compounds are 1: 252; 2: 250; 3: 224; and 4: 222. The MWT, retention time and UV absorption of 3 all matched to those of OA. To obtain compounds for structural determination, we first optimized the culturing conditions to get high titers from shake flask culture. By examining the organic extracts from cells and media separately, we determined that most of the compounds were secreted into the media. We also observed that a spore inoculum size of 10⁴ spores/mL led to the highest titers of the four compounds whereas inoculum sizes of 10⁸ spores/mL and higher gave low production of target compounds (Figure S4). Notably, when the molecules were produced at high titers, A. nidulans adopted a morphology of globular pellets. As the titer drops upon increased inoculum size, A. nidulans grew as dispersed filaments (Figure S4). This is in agreement with previous reports that Aspergillus niger accumulated a high titer of citric acid when the fungus grew as globular pellets.²⁵

From this simple optimization, these metabolites were purified from a large-scale culture and characterized by NMR (Supporting Figures S11–S25 and Tables S3–S5). Compound 3 (80 mg/L) was confirmed to be OA, while 1 (~ 1400 mg/L) was determined to be the C6-heptyl substituted 2,4-dihydroxybenzoic acid, i.e. sphaerophorolcarboxylic acid (SA) (Figure 2, Table 1). Compounds 2 (140 mg/L) showed a slightly red-shifted λ_max, together with a decrease in MWT of 2 mu compared to 1, indicating there is an extra degree of unsaturation that is conjugated with the aromatic ring. NMR analysis confirmed the presence of an olefin in the C6 alkyl substitution (Figure 2C). While 4 was not isolated due to its lower titer (estimated to be 300 μg/L), based on the UV absorption and −2 mu decrease in MWT compared to 3, we propose the structure to be olefin-containing version of 3 (Figure 2C). Significantly, formation of 3 unveils a new microbial pathway to the cannabinoid precursor. The most abundantly produced SA (1) is a precursor to the octanoyl-primed, rare analog of cannabinoids, such as tetrahydrocannabiphorol (THCP). THCP represents the most potent natural CB1 and CB2 modulators isolated to date, with K_i of 1.2 nM and 6.2 nM, respectively.²⁶ Hence the M. anisopliae pathway provides a facile route to access SA that can be further elaborated to the rare THCP and related molecules.

Table 1.

A summary of titers of olivetolic acid and analogues from heterologous expression in A. nidulans.

	1 (mg/L)*	2 (mg/L)	3 (mg/L)	4 (mg/L)
Ma_OvaA-C	1400 (+/− 80)	140 (+/− 20)	80 (+/− 10)	0.3
StcJ/K + Ma_OvaB-C	0	0	5 (+/− 1)	0
To_OvaA-C	750 (+/− 20)	75 (+/− 10)	40 (+/− 10)	2 (+/− 1)
Mr_OvaA-C	600 (+/− 10)	60 (+/− 10)	30 (+/− 5)	0.5 (+/− 1)
Ti_OvaA-C	0	0	60 (+/− 10)	0
Ma_OvaA + Ti_OvaB-C	75	0	4	0
Ti_OvaA + Ma_OvaB-C	0	0	0	0

^*Measurement of titers was performed after 5 days of culturing in 25 mL of CD-ST media in a 125 mL flask at 28°C and 250 rpm.

The biosynthesis of 1-4 confirms the hypothesis that a freestanding TE enzyme in a HRPKS/NRPKS-containing biosynthetic gene cluster is indicative of a resorcylic acid pathway. The biosynthesis of 1-4 requires all three enzymes, as we observed that omitting any of the three in A. nidulans completely abolished the biosynthesis of 1-4 (Supporting Figure S5). Based on the mechanism of dual PKS pathways, we assigned the functions of the enzymes as shown in Figure 3A. HRPKS Ma_OvaA can produce a mixture of four different starter units, octanoyl-, 2-octenoyl-, hexanoyl-, and 2-hexenoyl-thioester, with octanoyl-thioester being the most abundant. The α,β-unsaturated starter units result from the enoylreductase (ER) domain of Ma_OvaA not functioning in the last iteration prior to the chain transfer to Ma_OvaB. Transfer of any of these four starter units to Ma_OvaB, facilitated by the SAT domain, is followed by three additional rounds of chain elongation and aldol cyclization by the product template (PT) domain to yield the resorcylyl-thioester attached to the ACP domain of Ma_OvaB. ψACP-TE Ma_OvaC then performs thioester hydrolysis to give 1-4. The interaction between Ma_OvaB and Ma_OvaC may be enhanced by the ψACP domain as previously described for the ψACP-MT system,²³ although this requires more detailed biochemical characterization.

Figure 3.

Proposed biosynthetic pathway of olivetolic acid and its analogues from M. anisopliae ARSEF23 and proposed olivetolic acid biosynthesis pathway of the homologous pair from T. islandicus. A) HRPKS, Ma_OvaA, produces an ACP bound starter unit that is promiscuous in alkyl chain length and saturation degree. NRPKS, Ma_OvaB, accepts any of the four starter units and further processes each into a resorcylyl-thioester. This is followed by hydrolysis of the thioester by Ma_OvaC to generate olivetolic acid 3 and its analogues (1, 2, and 4). B) T. islandicus Ti_OvaA homolog only produces an ACP bound hexanoyl that is accepted by the SAT domain of the Ti_OvaB to form exclusively olivetolic acid 3.

Strategies towards finding an exclusive OA-producing pathway.

While the M. anisopliae cluster is able to make an abundant amount of SA, OA, the precursor to common cannabinoids, is produced as a minor product. It is desirable to obtain a pathway that can produce OA exclusively for further optimization. Since removing HRPKS Ma_OvaA abolished the production of all related compounds, we reasoned that supplying a strain expressing Ma_OvaB and OvaC with a hexanoyl starter unit could lead to exclusive OA production. First, 1 mM N-acetylcysteamine thioester (SNAC) of hexanoate was supplied to a 25-mL A. nidulans culture on day 2 and cultured for 5 days to test if the KS-domain of OvaB can directly capture it as the starter unit. However, when analyzing the sample on days 3, 4, and 5, no product can be detected from the culture, indicating hexanoyl-SNAC was either not taken up by the cell or was not accepted (Figure S6). Next, we attempted to directly generate hexanoyl-CoA as a primer for Ma_OvaB by feeding 1 mM hexanoic acid and coexpressing the Cannabis sativa acyl activating enzyme (CsAAE1) in combination with Ma_OvaB and Ma_OvaC. CsAAE1 was previously expressed in yeast to generate the hexanoyl-CoA for OAS incorporation.⁸ We cultured the transformed strain in 25 mL of CD-ST media in a 125-mL flask for 5 days following feeding with 1 mM of hexanoic acid on day 2. However, no OA or related product was observed (Figure S6), suggesting Ma_OvaB requires a hexanoyl starter unit attached to an upstream ACP, and small molecule mimics are not compatible.

To pair Ma_OvaB with a HRPKS-like enzyme that can supply an ACP-bound hexanoyl starter unit, we turned to the related biosynthetic pathways of aflatoxin B1 and sterigmatocystin. Both utilize a pair of fatty acid synthase-like enzymes to synthesize a hexanoyl starter unit, which is then transferred by the SAT domain of the NRPKS for elongation and cyclization to yield the intermediate norsolorinic acid.^27,28 In the A. nidulans sterigmatocystin pathway, the StcK/StcJ enzymes have been paired with an NRPKS from asperfuranone pathway to afford hexanoyl-primed hybrid products.²⁹ To test if crosstalk between StcJ/StcK and Ma_OvaB can take place, we coexpressed these enzymes with Ma_OvaC in the engineered A. nidulans ΔSTΔEM strain. Metabolic analysis showed that only OA is produced by this host at a titer of 5 mg/L. The lowered titer of OA using this combinatorial biosynthetic approach is likely due to nonnative and suboptimal protein-protein interactions between the ACP domain of StcJ/StcK and the SAT domain of Ma_OvaB. Nevertheless, this simple mix-and-match attempt showed that the product profile can indeed be manipulated by using different acyl donors.

Homologous clusters to the M. anisopliae cluster were discovered in T. inflatum, M. rileyi, and T. islandicus through bioinformatic analysis (Figure 2A). Although the clusters all share OvaA, B and C, their sequence identities are different (Supplementary Figure S1), indicating the potential to generate resorcylic acids with different C6 substituents. While OvaA, B, and C from T. inflatum (To_OvaABC) and M. rileyi share high homology (~84%−89%) with the M. anisopliae cluster, those from T. islandicus (Ti_OvaABC) share low homology (~46–52%). Heterologous expression of To_OvaABC in A. nidulans cultured in CD-ST revealed a similar product profile to Ma_OvaABC, albeit with lower titer (750 mg/L for 1 and 40 mg/L for 3) (Figure 2B, Table 1). Similarly, heterologous expression of Mr_OvaABC showed that 1-4 can be detected from A. nidulans extract at lower titer (600 mg/L for 1 and 30 mg/L for 3) (Table 1). Interestingly, when we heterologously expressed Ti_OvaABC genes from T. islandicus, 3 was exclusively produced at ~ 60 mg/L, which is ~12 fold higher than that from the StcJ/StcK and Ma_OvaBC combination (Figure 2B, trace ii; Table 1). Therefore, we propose that Ti_OvaA, differentiated from Ma_OvaA, selectively produces an ACP-bound hexanoyl starter unit, which leads to the exclusive formation of 3 (Figure 3B). Ti_OvaABC therefore represents a new pathway to produce olivetolic acid in microbial hosts.

Next, we explored combinatorial mix-and-match of OvaA-C from M. anisopliae and T. islandicus to determine if an exclusive 3-producing strain can be attained. We generated strains that coexpressed Ma_OvaA with Ti_OvaBC, as well as Ti_OvaA with Ma_OvaBC. Heterologous expression of Ma_OvaA with Ti_OvaBC produced 1 and 3 at 75 mg/L and 4 mg/L, respectively (Table 1). This is almost 19-fold less than heterologously expressing Ma_OvaABC. This result indicates that NRPKS does not have selectivity towards different C8 or C6 starter units. On the other hand, pairing of Ti_OvaA with Ma_OvaB-C led to abolishment of 1-4 production, suggesting the unnatural pair is not compatible.

There might be two factors contributing to the low titer (Ma_OvaA + Ti_OvaBC) and abolishment of production (Ti_OvaA + Ma_OvaBC): i) The most plausible cause is the disruption of specific protein-protein interactions between the domains participating in acyl chain transfer; and ii) rates of intermediates and final product formation by different clusters are different, in that each has been evolutionarily optimized to serve their respective biological hosts. The unnatural pairing can compromise the performance of overall biosynthesis by a “rate-limiting” component such as Ti_OvaBC. Aside from the molecular recognition of the acyl chain by the SAT domain active site, a successful acyl chain transfer also requires complementary protein-protein interactions between HRPKS ACP domain and NRPKS SAT domain.^30,31 Previous studies have shown that although for some non-cognate HRPKS-NRPKS pairs, acyl chain could occur and new products emerged without any protein engineering efforts,³² for others replacement of SAT domain is necessary to compensate for the otherwise undermined inter-domain communication.^29,33 While the sequence identity between ACPs of Ti_OvaA and Ma_OvaA is 62%, the identities between NRPKS SAT of Ti_OvaB and Ma_OvaB is lowered to 48%. Such moderate sequence identity between the SAT domains implies that the recognition sites between ACP from noncognate HRPKS and SAT can be weakened or even abolished. Therefore, to generate a combination that exclusively produces 3 at a higher titer, protein engineering endeavors that improve the compatibility between unnatural HRPKS and NRPKS enzymes are necessary. Alternatively, further genome mining of related clusters may lead to one that can produce olivetolic acid robustly in a heterologous host.

CONCLUSIONS

We have discovered a novel platform to produce olivetolic acid and its analogues from filamentous fungi. The platform consists of a HRPKS and a NRPKS, known to produce resorcylic acid moieties in tandem, and a separate thioesterase enzyme. This platform represents a new strategy to produce these cannabinoid precursors in microbes without relying on OAS and OAC found in Cannabis sativa.

METHODS

Strain and culture conditions

Metarhizium anisopliae ARSEF23, Tolypocladium inflatum, Metarhizium rileyi, and Talaromyces islandicus were all separately grown on PDA (potato dextrose agar) for 3 days and then transferred to liquid PDB (PDA medium without agar) for isolation of genomic DNA. Aspergillus nidulans 1145 was used as the model host for heterologous expression. Aspergillus nidulans 1145 was first grown on CD plates (10 g/L glucose, 20 g/L agar, 50 mL/L 20X nitrate salts, 1 mL/L trace elements) at 28°C and then cultured in 25 mL of CD-ST medium (20 g/L starch, 20 g/L casamino acids, 50 mL/L 20x nitrate salts, and 1 mL/L trace elements) in a 125 mL flask at 28°C and 250 rpm. 120 g of NaNO₃, 10.4 g of KCl, 10.4 g of MgSO₄•7H₂O, 30.4 g of KH₂PO₄ were dissolved in 1 L of distilled water in order to make 20x nitrate salts. 2.20 g of ZnSO₄•7H2O, 1.10 g of H₃BO₃, 0.50 g of MnCl₂•4H2O, 0.16 g of FeSO₄•7H2O, 0.16 g of CoCl₂•5H₂O, 0.16 g of CuSO₄•5H2O, and 0.11 g of (NH₄)₆Mo₇O₂₄•4H₂O were dissolved in 100 mL of distilled water with the pH being adjusted to 6.5 in order to make trace element solution.

Plasmid construction and expression

Plasmids pYTU, pYTP, and pYTR containing auxotrophic markers for uracil (pyrG), pyridoxine (pyroA), and riboflavin (riboB), respectively, were digested and used as backbones to insert genes. Genes expressed (OvaA, OvaB, and OvaC) were amplified through PCR using genomic DNA of Metarhizium anisopliae ARSEF23, Tolypocladium inflatum, Metarhizium rileyi, and Talaromyces islandicus as templates. StcJ/K genes were amplified through PCR using genomic DNA of Aspergillus nidulans. A glaA promoter and trpC terminator were amplified through PCR using pYTR as a template. The PCR fragments were transformed in yeast, and through homologous recombination, the plasmids pYTU-glaA- OvaB-trpC, pYTP-glaA-OvaC, and pYTR-glaA-OvaA-trpC were generated. Yeast transformation was performed using Frozen-EZ Yeast Transformation II Kit^™ (Zymo Research). The plasmids were extracted from yeast and transformed into E. coli TOP10 by electroporation to isolate single plasmids. After extraction from E. coli, plasmid sequences were confirmed by sequencing. All three plasmids (pYTU-glaA-OvaB-trpC, pYTP-glaA-OvaC-, pYTR-glaA-OvaA-trpC) were transformed into A. nidulans using standard protocols to form the olivetolic acid producing strain.²⁴ The strain was then cultured in 10 mL of CD-ST medium in a 50 mL falcon tube and kept in a shaker at 28 °C and 250 rpm overnight. The next day 25 μL of the culture was inoculated into 25 mL of CD-ST medium in a 125 mL flask and kept in a shaker at 28 °C and 250 rpm.

C. Sativa acyl activating enzyme (CsAAE1) was ordered as a gene block from Integrated DNA Technologies (IDT). CsAAE1 was amplified through PCR and cloned onto the pYTR backbone containing the gpdA promoter. After following the previously outlined protocol to construct and confirm the sequence of plasmids, we transformed the pYTR-gpda-CsAAE1, pYTU-glaA-OvaB-trpC, and pYTP-glaA-OvaC plasmids into A. nidulans. After 5 days of growth at 37°C on CD sorbitol plates, the strain was cultured in 25 mL of CD-ST medium in 125 mL flasks at 28°C and 250 rpm. On day 2, 1 mM hexanoic acid was added into the culture and analysis of the sample was done on days 3, 4, and 5.

Aspergillus nidulans heterologous expression

To produce protoplasts, Aspergillus nidulans 1145 was grown on CD agar plates with 10 mM of uridine, 5 mM of uracil, 0.5 μg/mL of pyridoxine HCl and 2.5 μg/mL of riboflavin at 37 °C for 4 days. Aspergillus nidulans 1145 spores were then inoculated in 25 mL of CD liquid media containing 10 mM of uridine, 5 mM of uracil, 0.5 μg/mL of pyridoxine HCl and 2.5 μg/mL of riboflavin in a 250 mL flask and grown at 28 °C and 250 rpm for 16 hours. The mycelia were harvested by centrifugation at 4300 × g for 20 min and washed with osmotic buffer (1.2 M of MgSO₄, 10 mM of sodium phosphate, pH 5.8). After harvesting by centrifugation once more, mycelia were then transferred to a 250-mL flask containing 10 mL of osmotic buffer with 30 mg lysing enzymes from Trichoderma and 20 mg yatalase. The mycelia were digested for 5 hours at 28 °C and 80 rpm. The cells were then transferred to a 50-mL centrifuge tube, were overlaid gently by 10 mL of trapping buffer (0.6 M of sorbitol, 0.1 M of Tris-HCI, pH 7.0), and centrifuged at 5300 rpm for 20 min at 4°C. Two layers appeared with the protoplasts at the interface of the two layers. The protoplasts were collected and placed in a sterile 15 mL falcon tube and washed with 10 mL of STC buffer (1.2 M of sorbitol, 10 mM of CaCl₂, 10 mM of Tris-HCI, pH 7.5) and centrifuged for 10 min at 4300 × g and 4°C. The protoplasts were then suspended in 1 mL of STC buffer, aliquoted in 60 μL increments in 1.5 mL microcentrifuge tubes and stored at −80 °C.

For transformation, 2 μL of each plasmid needed for heterologous expression was added to the 60 μL aliquots of Aspergillus nidulans 1145 protoplasts and then kept on ice for 60 min. The aliquots were then mixed with 600 μL of PEG solution (60% PEG, 50 mM of calcium chloride, and 50 mM of Tris-HCl, pH 7.5), incubated for 20 min, and plated on CD sorbitol agar plates (CD solid medium with 1.2 M of sorbitol and the appropriate supplements, according to the markers on the plasmids). Plates were then incubated at 37 °C for 3–5 days.

For hexanoyl-SNAC feeding, the A. nidulans strain containing pYTU-glaA-OvaB-trpC and pYTP-glaA-OvaC plasmids was cultured in 25 mL of CD-ST in a 125-mL flask at 28°C and 250 rpm for 5 days. On day 2, 1 mM hexanoyl-SNAC was fed into the cultures and analysis of the sample was done on days 3, 4, and 5.

Sample Preparation, Detection, Isolation, and Quantification

Individual colonies from transformation plates were cultured in 25 mL of liquid CD-ST media in 125-mL flasks and grown for six days. On the sixth day, 500 μL of cells and media were extracted with 800 μL of an ethyl acetate acid mix (1: 9 methanol to ethyl acetate with .1% formic acid). The extracted sample was then dried and placed in 50 μL of methanol and then loaded onto the LC-MS.

LC-MS analyses were performed using a Shimadzu 2020 LC-MS (Phenomenex® Kinetex, 1.7 μm, 2.0 × 100 mm, C-18 column) using positive and negative mode electrospray ionization. The elution method was a linear gradient of 5–95% (v/v) acetonitrile/water in 13.25 min, followed by 95% (v/v) acetonitrile/water for 4.75 min with a flow rate of 0.3 mL/min. The LC mobile phases were supplemented with 0.1% formic acid (v/v).

Large-scale production of compounds for the purpose of isolation and structural determination was carried out by cultivating transformants in 1 L CD-ST. After 5 days of growth at 28 °C, the media was extensively extracted with acidified ethyl acetate. The extract was concentrated under reduced pressure. Purification was carried out as previously reported with slight modifications.²³ Briefly, the residue was loaded to a Redisep Rf Gold Reversed-phase C18 column on a Teledyne Combi-Flash system. Subsequently, HPLC purifications were performed with a Phenomenex® Kinetex column (5μ, 10.0 × 250 mm, C18) using a Shimadzu UFLC system. For HPLC purification, a flow-rate of 4 ml/min with solvent A (0.1% formic acid in water) and B (0.1% formic acid in acetonitrile) was used. NMR spectra of 1-3 were acquired on a Bruker AV500 spectrometer with a 5 mm dual cryoprobe (¹H 500 MHz, ¹³C 125 MHz).

Quantification of the compounds was done by first making a standard curve on the HPLC. Known concentrations of isolated compound were analyzed on the HPLC and a standard curve was constructed correlating the area under the UV peak corresponding to the compound. Cultured samples were then extracted and analyzed on the HPLC where the area under the UV peak was used to calculate the concentration of the sample.