General experimental procedures
Semipreparative HPLC was performed on an Agilent 1260 instrument, equipped with a diode-array detector, UHPLC-MS runs were done on an Agilent 1290 Infinity II chromatograph, interfaced to an Agilent 6130 single quadrupole mass detector. HR-ESI-MS spectra were recorded in positive mode on a Thermo Scientific Exactive Orbitrap instrument. UV/Vis spectra were recorded from λ = 200–700 nm with diode array detectors connected with the respective chromatographs, chromatograms were extracted at λ = 450 nm.
Microbial strains and growth conditions
Escherichia coli XL1-blue was used for routine cloning and was cultivated in LB supplemented with 50 µg ml−1 carbenicillin, if required. Laetiporus sulphureus (s.l.) JMRC SF012599 was provided by the Jena Microbial Resource Collection (JMRC) and was routinely maintained on MEP medium (per liter: malt extract 30 g, soytone peptone 3 g, agar 18 g) for 14 days at 25 °C. To produce laetiporic acids, L. sulphureus was cultivated on 150 YPD agar plates (per liter: yeast extract 10 g, soytone peptone 20 g, d-glucose 20 g, agar 18 g) at 20 °C for 14–21 days. Fruiting bodies of L. sulphureus were collected in Jena, Germany, on willow trees along the Saale river, in September 2019. Aspergillus strains used for transformation were A. niger ATNT16ΔpyrGx24  and A. nidulans FGSC A4.
A. niger transformants tPS01 and tPS02 were cultivated on Aspergillus minimal medium agar plates (AMM)  supplemented with 5 mM l-glutamine at 30 °C for 5–7 days. Media for ATNT16ΔpyrGx24 were supplemented with 10 mM uridine. To produce laetiporic acids in recombinant A. niger, transformants tPS01 (vector control) and tPS02 (polyene producer) were pre-cultivated overnight in 30 Erlenmeyer flasks, each filled with 50 ml YPD medium, at 30 °C and 140 rpm. The main culture was a 30 × 1 l fermentation (AMM containing 200 mM d-glucose and 50 mM l-glutamine), inoculated with 50 ml pre-culture each. To induce lpaA expression, 30 mg l−1 doxycycline was added after 18 h, and cultivation was continued for additional 48 h. A. nidulans FGSC A4 and mutant tMG01 were maintained on AMM plates supplemented with 5 mM l-glutamine at 37 °C for 3 days. Plates for tMG01 were supplemented with 0.1 µg ml−1 pyrithiamine hydrobromide. To produce laetiporic acids in A. nidulans, the strains were cultivated in 100 l auto-inducing AMM, prepared with 200 mM ethanol and 10 mM d-glucose as carbon sources, at 30 °C and 140 rpm for 72 h. Details on fungal strains are given in Supplementary Table S1. Aspergillus conidia were harvested with 10 ml sterile water and the suspension was filtered by a cell strainer (40 µm, EASYstrainer). Media were inoculated at a titer of 1 × 106 conidia per milliliter.
cDNA cloning and construction of lpaA expression plasmids
L. sulphureus mycelium was grown in liquid YPD medium at 20 °C and 140 rpm for 7 days, harvested, and ground under liquid nitrogen. RNA was isolated using the SV Total RNA Isolation Kit (Promega). Residual genomic DNA was digested by Baseline-ZERO DNase (Biozym). Reverse transcription was carried out with anchored oligo-dT18 primers and RevertAid Reverse Transcriptase (ThermoFisher). The lpaA coding sequence was PCR-amplified from the first strand reaction, using the oligonucleotides oCL46 and oCL47 (Supplementary Table S2), using method A (Supplementary Table S3). The gel-purified fragment was ligated into pJET1.2 (Thermo) to yield plasmid pCL10 (Supplementary Table S4) which was sequenced (GenBank accession number MT304701) to verify accurate amplification and then served as template for subsequent PCRs. A tag-free version (8190 bp) was expressed in A. nidulans, while in A. niger, a gene for a hexahistidine fusion protein was used (8229 bp).
The lpaA coding sequence was PCR-amplified (method B, Supplementary Table S3) from pCL10 using oMG459 and oMG460 (Supplementary Table S2) in order to introduce PacI sites at either end of the fragment. The A. niger expression vector pSMX2-URA , allowing for doxycycline-inducible gene expression, was modified by PCR-mediated ligation (oligonucleotides oMG457/oMG458) to incorporate a PacI restriction site in the multiple cloning site to create vector pPS01 (Supplementary Tables S2, S4). Both the insert and pPS01 were restricted with PacI and ligated to create the lpaA expression vector pPS03. Plasmids pPS01 (vector) and pPS03 (lpaA expression plasmid) were used to transform A. niger.
To construct an alcohol-inducible lpaA expression vector, the vector backbone of plasmid pMD03  as well as the lpaA coding sequence inserted in plasmid pCL10 were amplified (method B, Supplementary Table S3) using oligonucleotides oMG468/oMG469 and oMG471/oMG472, respectively. Both fragments were ligated using the NEBuilder HiFi DNA Assembly Cloning Kit (NEB) to yield lpaA expression plasmid pMG49, which was used to transform A. nidulans. Details of plasmids are described in Supplementary Table S4.
Transformation of Aspergillus species
Protoplast transformation of A. niger and A. nidulans was carried out as previously described [7, 9]. In brief, protoplasts were obtained by incubation of mycelium with VinoflowPro (1.1 g per 20 ml volume) for 4 h in YAT buffer (0.6 M KCl, 50 mM maleic acid, pH 5.5) and 10 µg of plasmid DNA (pPS01 and pPS03 for A. niger, pMG49 for A. nidulans) were used for polyethylene glycol-mediated transformation. A. niger transformants (tPS01 and tPS02) were selected by uracil prototrophy, while A. nidulans transformants (tMG01) were selected by pyrithiamine resistance in presence of 0.1 µg ml−1 pyrithiamine. Integration of the lpaA gene was confirmed by PCR (Supplementary Table S3, methods C and D).
UHPLC analysis of laetiporic acids
Methanolic crude extracts of mycelia from Aspergilli and Laetiporus mycelia and carpophores were centrifuged, filtered, and were subjected to UHPLC measurements. Method B (Supplementary Table S5) was used for initial screening of extracts of positive transformants or selection of fractions containing laetiporic acids during the purification procedure (see below). Method B was also applied for polyene quantification in growth inhibition assays.
Purification of laetiporic acids
One-hundred and fifty Laetiporus agar plates were diced, lyophilized, and extracted with acetone (3 × 5 l). Aspergillus mycelia were collected, washed with water, and lyophilized. Mycelia were ground to a fine powder and extracted six times with methanol and subsequently twice with acetone (50 ml per 1 g dry biomass and extraction).
Aspergillus and Laetiporus extracts were filtered through cellulose round filters and evaporated to dryness. The dry residue was dissolved in 2 l water and repeatedly extracted with a total of 12 l ethyl acetate. The organic phase was evaporated. The residue was dissolved in 400 ml methanol, and 40 ml aliquots were subjected to size exclusion chromatography on Sephadex LH-20 (60 × 4 cm) with methanol as eluent. Three fractions (FI-III) were obtained, containing laetiporic acid (LA)-A, LA-B in F-I, LA-C in F-II, and LA-D and traces of other derivatives in F-III. All fractions were subjected to reversed phase semi-preparative HPLC, using methods C (F-I and F-II) and D (F-III) (Supplementary Table S5). Isolated compounds were lyophilized and dissolved in methanol. Final work up was accomplished under slightly basic conditions (method E for LA-A and LA-B, method F for LA-D, Supplementary Table S5).
To test for photoisomerization, 100 µg ml−1 of Aspergillus-produced laetiporic acids A1, B1, or B2, respectively, or laetiporic acid A1 from L. sulphureus were continuously exposed to light for 24 h (or in the dark for control). After exposure, the solutions were chromatographically analyzed by UHPLC-MS (method A, Supplementary Table S5).
Growth inhibition assays
Cultivations were carried out in triplicate at 30 °C and 140 rpm with a conidial titer of 1 × 106 per ml in 30 ml medium. A. niger tPS01 and tPS02 were cultivated in AMM (+200 mM d-glucose and 50 mM l-glutamine) for 42 h. Doxycycline hydrochloride (0, 7.5, 15, 30, 60, or 120 µg ml−1) was added immediately after inoculation. A. nidulans strains were cultivated in AMM without glucose, but with 200 mM ethanol as sole carbon source (inducing condition) for 72 h. For delayed gene expression, d-glucose (0, 2.5, 5, 10, or 50 mM) was added prior to inoculation. AMM with 200 mM d-glucose (repressing condition) served as negative control. After cultivation, the mycelium was lyophilized, weighted, ground to a fine powder, and extracted with 1 ml methanol for 5 min in an ultrasonic bath. After centrifugation (10 min, 13,000 g), an aliquot of 5 µl was subjected to HPLC analysis (method A, Supplementary Table S5) and polyene signals were manually integrated (λ = 450 nm, tR = 4–8 min). Polyenes were quantified against a calibration curve with respective SEC-purified polyenes as authentic reference standards.
Protoplast toxicity assays
To obtain polyene-containing and control extracts, A. nidulans was cultivated in AMM (+10 mM d-glucose and 200 mM ethanol) at 30 °C and 140 rpm for 72 h. Lyophilized mycelium (5 g) was ground to a powder and extracted five times with 200 ml methanol. Extracts were purified via size exclusion chromatography with Sephadex LH-20 as described above. Fractions were analyzed with UHPLC-MS, and fractions containing laetiporic acids were pooled for tMG01 and added in concentrations from 31 µg ml−1 to 4 mg ml−1. As negative control, appropriate SEC fractions of A. nidulans wild type were pooled accordingly. The eluates were dried under reduced pressure and residues were suspended in sterile YAT buffer.
To produce protoplasts, A. nidulans mycelium was filtered, washed with sterile YAT buffer, and incubated in lysis solution (1.3 g VinoTaste Pro (Novozymes), 0.1 g lysing enzymes from Trichoderma harzianum (Sigma), 0.1 g Yatalase (Takara), in 20 ml YAT buffer) at 30 °C and 70 rpm for 3 h. Protoplasts were filtered through sterile Miracloth, washed with YAT buffer, counted, and diluted to a final titer of 2 × 104 protoplasts per ml. 100 µl (2 × 103 cells) of the cell suspension were gently mixed with the same volume of extracts (A. nidulans FGSCA4 or tMG01), and protoplasts were incubated on ice for 3 h. Suspensions were carefully plated on osmotic AMM plates (with 1.2 M sorbitol, 100 mM d-glucose, 20 mM l-glutamine, pH 6.5) and incubated at room temperature for 96 h. Colony forming units from four independent experiments were counted.