Anthracaurisin ADiscovery of a Cytotoxic Nor-Sesquiterpene Dimer from Patagonian Anthracophyllum discolor Authors: Bastien Petit, Ines Michrafy, Jordan Nafie, Grayce E. Dyer, Erin M. Marshall, Cristian Riquelme, Jaime R. Cabrera-Pardo, James A. Strother, Sandra Loesgen Published: 2026 ABSTRACT Fungal natural products, also called secondary metabolites (SMs), continue to inspire chemists and drug development. Around 30,000 fungal SMs are known, mainly from filamentous fungi of the phylum Ascomycota. Only 5% of all fungal species are described today, and even fewer have been cultured or chemically screened, which makes fungi an untapped reservoir for chemical discovery. The phylum Basidiomycota harbors chemically talented fungi such as the psilocybin producer Psilocybe cubensis (Agaricales, Hymenogastraceae), but other members like the genus Anthracophyllum (Agaricales, Omphalotaceae) remain largely genetically and chemically underexplored although new species are being discovered frequently. Here, we present the chemical and bioactivity exploration of a Chilean Anthracophyllum discolor fungus. We discovered anthracaurisin A, the first fungal nor-sesquiterpene homodimer with potent cytotoxic activity in vitro and in vivo, isolated one new and several known illudins, and established their absolute configurations. Fungi are prolific producers of natural products, also called secondary metabolites (SMs). An estimated 30,000 fungal SMs are known, mainly from filamentous fungi of the phylum Ascomycota. Estimates indicate that only 5% of all extant fungal species have been described, and even fewer have been cultured or chemically screened, suggesting that fungi remain an untapped reservoir for SMs discovery. From Ascomycota alone, which includes well-studied genera such as Aspergillus, Fusarium, and Penicillium, a conservatively estimated 1.4 million compounds remain to be discovered. The chemical diversity of fungal metabolites and their potential for human health applications can be seen in clinically used drugs from Ascomycota such as the immunosuppressing agent cyclosporin from Tolypocladium inflatum, the cholesterol-lowering lovastatin from Aspergillus terreus, and the antibiotic penicillin from Penicillium rubens. In recent years, more SMs from the phylum Basidiomycota have been discovered, and there has been reinvigorated interest in psilocybin from Psilocybe cubensis for its biosynthesis and to develop and treat neurological disorders. The genus Anthracophyllum (Agaricales, Omphalotaceae) includes about 12 species of pleurotoid/marasmioid Basidiomycetes fungi growing in tropical regions of the world. Our literature review revealed that new Anthracophyllum species are being continuously discovered, but no genome sequences are available for the family of Omphalotaceae, and only a few partial gene loci amplifications can be found for Anthracophyllum lateritium, A. archeri, A. sinensis, and A. sontraense. , In the past, morphological criteria placed this genus in the Marasmiaceae family, but modern taxonomic criteria place Anthracophyllum within the Omphalotaceae family. , Macroscopically, Anthracophyllum species are well-known for the pigmentation of their basidiomes, with colors from orange to reddish-brown but also black or dark blue depending on the species. A blue-green coloration is typically observed when treating the basidiomes with aqueous alkali due to the presence of various terphenyl derivatives including cycloleucomelon. , Another natural product study has reported the occurrence of sesquiterpenoids including the class of aristolane sesquiterpene dimers called aurisins. Aside from this, very little is known regarding the SMs synthesized by Anthracophyllum, which has been studied for its ability to degrade organic matter and its bioremediation potential. − In an ongoing effort to discover new and bioactive natural products, we screened a set of Chilean–Patagonian basidiomycetes for antimicrobial, cytotoxic, and pain-relieving activities. , The screening identified an Anthracophyllum discolor extract (internal strain name JCP1718), grown on rice-based media, with potent antibacterial and cytotoxic activities. Mass spectrometry (MS)-based dereplication efforts supported the presence of potentially unique chemical features. Here, we report the isolation and bioactivity of six terpenoids related to the illudane class of natural products. The known illudanes (1, 2, 3, and 5) were subjected to stereochemical analysis using ECD and VCD spectroscopy supported by DFT-based computations to unequivocably assign their absolute configuration for the first time. We also found one new illudane (4) as well as a unique nor-sesquiterpene dimer scaffold with potent cytotoxicity, which we named anthracaurisin A (6). Results and Discussion A set of Patagonian Basidiomycetes was collected in Chile and identified based on macroscopic and microscopic observations, which was supported by ITS sequencing. These fungi were grown in two media conditions (solid rice and liquid malt), each extracted using two different solvents (ethyl acetate and methanol). The resulting extracts were screened in multiple bioassays: (i) for antimicrobial activity against human pathogens (including Staphylococcus aureus, multidrug-resistant S. aureus, Escherichia coli, and Candida albicans), (ii) for inhibition of cell viability using a human colon cancer cell line (HCT-116), and (iii) for in vivo effects in a zebrafish acute toxicity assay. One extract, A. discolor grown on rice and extracted with ethyl acetate, exhibited potent antistaphylococcal activity and toxic activity in vitro and in vivo. In parallel, all extracts were subjected to a chemical dereplication approach using high-resolution LC-MS and LC-MS/MS with the aim of avoiding the isolation of known fungal metabolites. The bioactive A. discolor extract exhibited unique MS features with m/z values of 435.2559 (ESI +) and 553.2831 (ESI −), which could not be matched with known metabolites from our in-house UV database and online/commercial databases. , Moreover, the mass spectrometric feature was considerably more abundant in the solid rice extracts than the liquid malt extracts, suggesting its occurrence to be highly dependent on the growth medium conditions. Structure Elucidation of Compounds 1 to 6 Six compounds were isolated from the ethyl acetate rice extract of A. discolor, including the known illudins M (1), B (2), S (3), and dichomitin A (5), together with the new 15-O-acetylilludin S (4) and anthracaurisin A (6). Their structures (Figure ) were elucidated via extensive NMR and HRESIMS experiments. Stereochemical assignments were achieved using NOE NMR and ECD/VCD spectroscopic experiments, supported by DFT-based computations. Multiple stereoisomers of illudins 1–3 were previously isolated, however, with limited spectral data, and the absolute configuration of dichomitin A (5) was subjected to revision in the past. To unequivocally assign the absolute configuration of 1, 2, 3, and 5, we carried out a series of VCD and ECD experiments in conjunction with measuring their optical rotation. Structures of compounds 1 to 6. Compounds 4 and 6 are previously undescribed. Compound 1 was identified as illudin M based on mass spectrometry and NMR data (details in SI S49). However, its specific rotation (−44°) was significantly different from the published specific rotation of illudin M ((2R,6S)-illudin M, −284°). , Multiple stereoisomers of illudin M have been described recently, including (2S,6R)-illudin M with a reported specific rotation of +96° (named (3S,7R)-illudin M in the cited literature) and (2S,6S)-illudin M with a specific rotation of +59° (named (3S,7S)-illudin M in the cited literature). For further details on carbon numbering systems for illudins, please see Figure S66 in the SI. Despite similar experimental conditions, none of these specific rotation values matched ours (Table S52). We decided to use experimental and computational VCD and ECD approaches, which revealed compound 1 to be the original published illudin M ((2R,6S)-illudin M) by comparison to its DFT-calculated and experimental ECD/VCD data (Figure , details in SI S53–S55). Vibrational circular dichroism (VCD) is a powerful technique for the study of stereochemistry in the solution phase without derivatization. Although ECD has been previously used for the stereochemical assignments of illudane sesquiterpenoids, to our knowledge, VCD is used here for the first time. − Multiple stereoisomers coexist in illudane literature, and the specific rotations alone are not sufficient to determine their stereochemistry. We believe our ECD/VCD-based approach toward the absolute configuration of illudins will be helpful to the community. Experimental (red) and calculated (blue) VCD and ECD spectra of known compounds 1, 2, 3, and 5. Compound 2 shared a number of similar NMR signals with compound 1 (details in SI S49). Its mass spectrometry data revealed its nominal mass to be 34 Da greater than that of compound 1, which could suggest two additional hydroxy groups and a double bond reduction. Extensive 2D NMR analysis revealed the two supernumerary hydroxys to be located at C-4/C-8 and the reduction to involve the cross-conjugated dienone system spanning C-8, C-9, C-4, C-5, and C-1 of the illudin backbone. Two stereoisomers of this structure were previously described, known as illudin B and H. , ROESY NMR and specific rotation of compound 2 (−18°) shared many similarities with illudin B (−9°), but these resemblances were not considered sufficient for a stereochemical assignment given the 16 possible stereoisomers for this structure. Similarly to compound 1, we performed VCD and ECD spectral analyses of compound 2 supported by DFT-based computations, which confirmed 2 to be illudin B ((2R,4S,6S,8R)-illudin B) (Figure , details in SI S56–S58). Compound 3 was identified as illudin S based on mass spectrometry and NMR data (details in SI S49). Reminiscent of compound 1, the specific rotation of compound 3 (−104°) was significantly different from the reported specific rotation of illudin S ((2R,6R,7S)-illudin S, – 358°) , and another stereoisomer of illudin S reported recently from the Basidiomycete Granulobasidium vellereum. Again, experimental and computational VCD and ECD approaches were used to unambiguously assign compound 3 to have the 2R,6R,7S configuration of illudin S (Figure , details in SI S59–S61). Compound 4’s NMR data were almost superimposable to that of compound 3, except for a key chemical shift difference at C-15 with δC 69.3/δH 3.39, 3.33 ppm for compound 3 and δC 70.6/δH 3.97, 3.93 ppm for compound 4 (Tables and S49). 2D NMR experiments allowed the identification of an extra O-acetyl moiety located at C-15 in compound 4, as demonstrated by the HMBC cross-peak of H2-15 δH 3.97, 3.93 to δC 172.6 (15-O-Ac, CO) (Figure ). This was further supported by mass spectrometry evidence, in which compound 4 showed 42 additional mass units compared to compound 3. The relative and absolute configuration of compound 4 seemed to be identical to compound 3 based on NMR chemical shifts, NOESY correlations, and specific rotations of similar sign and magnitude (compare Tables with S49 for NMR, compare Figures with S51 for NOESY correlations, and compare the new compounds description with Table S52 for specific rotations). VCD and ECD experiments confirm this statement and assign the three stereocenters of 4 to be in the 2R,6R,7S configuration (Figure ). Based on the above data, compound 4 was named 15-O-acetylilludin S. Although this compound has been previously synthesized as part of structure–activity studies on illudins, here is the first report of 4 as a natural product. ,, Key HMBC, COSY, and NOESY correlations of previously undescribed compounds 4 and 6. Experimental (red) and calculated (blue) VCD and ECD spectra of 15-O-acetylilludin S (4) and anthracaurisin A (6). 1 1H and 13C NMR Spectroscopic Data for Compound 4, 600 MHz, Methanol-d 4, Cryoprobe, 1.7 mm Tubes 4positionδC, typeδH 1201.4, C 277.6, C 332.8, C 4135.6, C 5139.4, C 675.6, CH4.59, s754.1, C 8141.5, CH6.39, s9137.1, C 1024.7, CH3 1.36, s118.9, CH2 1.09, m; 0.44, m126.2, CH2 0.97, m; 0.79, m1314.3, CH3 1.68, s1416.6, CH3 1.18, s1570.6, CH2 3.97, d (11.0); 3.93, d (11.0)15-O-Ac (CO)172.6, C 15-O-Ac (Me)20.5, CH3 1.99, s 1H and 13C NMR Spectroscopic Data for Compound 4, 600 MHz, Methanol-d 4, Cryoprobe, 1.7 mm Tubes Compound 5 NMR and mass spectrometry data were similar to the previously reported dichomitin A, also called dichomitol (Table S50). , Its chemical shifts and NOESY correlations were identical to dichomitin A, suggesting that both compounds shared the same relative and absolute configuration (Figure S51). However, the specific rotation of compound 5 (−5°) was of opposite sign of dichomitin A (+6.3°) (Table S52). VCD and ECD experiments assigned the absolute configuration of compound 5 to be 2S,3R,8S,9R,11R, identical to the dichomitin A structure published in 2004, however here with a different specific rotation value (Figure ). This case illustrates nicely how stereochemical assignments based on NMR and specific rotation data can benefit from expanded spectral analyses and computations. Compound 6 exhibited a molecular formula of C32H42O8, deduced from its HRESIMS data showing a [M–H]− ion peak of m/z 553.2831 (calcd for C32H41O8 –, 553.2807; mass error – 4.34 ppm). An extensive analysis of its NMR data suggested 6 to be a dimer formed by the fusion of two illudane sesquiterpenoid monomers, which were not isolated nor detected from the fungus (Figure ). Each monomer exhibited four methyls CH3-11 (δC/δH 31.1/1.06), CH3-12 (δC/δH 15.7/1.25), CH3-13 (δC/δH 24.1/1.04), and CH3-14 (δC/δH 12.3/1.06), one hydroxy-methine CH-7 (δC/δH 66.9/5.92), one hydroxy-olefinic carbon C-8 (δC 164.4), one ketone C-1 (δC 204.4), one ester moiety (δC 171.1), and a gem-cyclopropyl unit at CH-5 (δC/δH 31.6/0.97) and CH-6 (δC/δH 26.5/1.32) (Table ). The dimer is connected at C-2 based on the HMBC cross-peak between CH-2 and its dimeric equivalent CH-2′, at δC/δH 49.0/3.03 (Figure ). 2 1H and 13C NMR Spectroscopic Data for (P) and (M) Atropisomers of 6, 500 MHz, Chloroform-d, J in Hz, ov., Overlapped P- 6 M- 6 positionδC, typeδH δC, typeδH 1/1′204.4, C 205.5, C 2/2′50.0, CH2.18, ov.49.0, CH3.03, br d (12.0)3/3′42.8, CH2.65, m40.6, CH1.97, m4/4′41.0, C 41.2, C 5/5′31.6, CH0.97, d (8.6)31.4, CH1.01, ov.6/6′26.5, CH1.32, t (7.5)26.7, CH1.39, t (7.5)7/7′66.9, CH5.92, br d (5.8)66.5, CH5.98, br d (5.8)8/8′164.4, C 163.5, C 9/9′117.2, C 116.6, C 10/10′21.4, C 21.4, C 11/11′31.1, CH3 1.06, br s30.9, CH3 1.06, br s12/12′15.7, CH3 1.25, br s16.5, CH3 1.14, br s13/13′24.1, CH3 1.04, br s24.3, CH3 1.11, br s14/14′12.3, CH3 1.06, ov.13.3, CH3 0.85, br d (6.1)7/7′-O-Ac (CO)171.1, C 170.9, C 7/7′-O-Ac (Me)21.0, CH3 2.15, br s21.0, CH3 2.15, br s8/8′-OH 6.80–7.60, bs s 6.80–7.60, br s 1H and 13C NMR Spectroscopic Data for (P) and (M) Atropisomers of 6, 500 MHz, Chloroform-d, J in Hz, ov., Overlapped Compound 6’s NMR data showed a phenomenon of duplication of some of the proton signals and most of the carbon signals with a consistent ratio of 1:0.6 between each signal and its duplicate equivalent, particularly apparent on the 13C spectrum of 6 in CDCl3 (Figure ). 10 to 35 ppm region of 13C NMR spectrum of anthracaurisin A (6) exhibits doubled signals for C-5/5′, C-6/6′, C-11/11′, C-12/12′, C-13/13′, and C-14/14′ of the P and M atropisomers. (A) 13C NMR spectrum of 6 in CDCl3 at 298 K. (B) 13C NMR spectrum of 6 in DMSO-d 6 at 298 K, (C) 323 K, (D) 348 K, and (E) 373 K. Variable temperature (VT) NMR experiments in DMSO-d 6 at 323, 348, and 373 K allowed coalescence of the previously duplicated 1H and 13C NMR signals at higher temperature, confirming that the two halves are identical, but that the compound exists as a pair of distinct M or P atropisomers (rotamers along the C-2/C-2′ axis) in solution, on the NMR time scale, at room temperature (Figure , Table S39). We named the (P) and (M) atropisomers accordingly, following the axial chirality nomenclature, , and assigned different chemical shifts to the atropisomers based on hydrogen bonding (Figure ). Structures of the P and M atropisomers of anthracaurisin A (6). (A) Definition of the A, B, A′, B′ axes of 6 and Newman projections of the two atropisomers, P-6 (left) and M-6 (right). (B) Perspective representation of P-6 (left) and M-6 (right) structures. 13C and 1H chemical shift differences for P and M conformers at C-2/C-2′ and C-3/C-3′ are emphasized in blue color and may be attributed to different, possible hydrogen bonding patterns (red dashed lines). (C) Wireframe representation of DFT-calculated lower-energy conformers of 6 generated in Avogadro software with hydrogen bonding in red dashed lines. (D) Flat representation of P-6 (left) and M-6 (right). The relative and absolute configuration of 6 was established by VCD and ECD approaches supported by DFT calculations to be (2R,2′R,3S,3′S,4S,4′S,5R,5′R,6S,6′S,7R,7′R) (details S46–48). Briefly, computed IR and VCD spectra at the B3PW91/cc-pVTZ DFT level were compared to their experimental counterparts both visually and via algorithmic (CompareVOA, BioTools, Jupiter, FL) analysis (Figure , details Figures S46–S48). Quantitative comparisons of the experimental and DFT-derived IR and VCD spectra resulted in high similarity (Sfg) and enantiomeric similarity index values for the (2R,2′R,3S,3′S,4S,4′S,5R,5′R,6S,6′S,7R,7′R) configuration and a high confidence level (99%) assignment (Figure S48). To support the relative configuration, we calculated six alternate structures, varying the dimer connection stereochemistry, as well as inverting each stereocenter or pair of stereocenters in the cyclopropyl case, on both halves of the dimer, to see if any improvements to the spectral agreement were possible. In each case, the spectra were visibly a worse match and yielded lower similarity values. In addition, we obtained UV–vis and ECD data for this compound, and the DFT-calculated spectra for the (2R,2′R,3S,3′S,4S,4′S,5R,5′R,6S,6′S,7R,7′R) configuration produced an excellent match to the experiment. Our stereochemical assignment is additionally supported by literature data: the aurisins are structurally related sesquiterpene dimers, produced by Basidiomycota genera Panus, Neonothopanus, and Anthracophyllum, and the aurisins share several stereocenters in the same configurations. ,, Compound 6 was thus named anthracaurisin A after the producing strain and the related class of aurisin terpenoids. We were able to detect anthracaurisin A in rice cultures extracted with dichloromethane and acetonitrile, supporting the notion that this is a natural product and not an artifact of the ethyl acetate extraction. Biosynthetic Hypothesis Despite being related to the aurisins, 6 lacks a carbon per monomeric subunit (C14 monomer for anthracaurisin A instead of C15 monomer for the aurisins); therefore, it falls within the rare class of nor-sesquiterpene dimers. We propose that the biosynthesis starts with a sesquiterpene maaliane cation, a precursor for all aristolene monomers with the characteristic gem-dimethylpropane moiety also found in aurisins. Starting from the aristolene scaffold, we propose an early installment of the ketone and hydroxy groups, as seen in other members of this class of compounds (Figure ). Ring contraction from a six-membered cyclohexanone to a five-membered ring has been extensively studied in the gibberellin biosynthesis, which is catalyzed by a cytochrome P450 enzyme (CYP114) via a semipinacol rearrangement. We propose a similar ring contraction, followed by oxidation of the excised aldehyde carbon and decarboxylation of the acid. The resulting radical species could undergo radical dimerization and give rise to 6. , Putative biosynthetic pathway for anthracaurisin A (6). Sesquiterpene dimers are rare in fungi, and the first report appears to be trichoderonic acid B, a dimer of the heptelidic acid sesquiterpene isolated from a Japanese Trichoderma virens. More heptelidic acid dimer analogues were later characterized from the pleurotoid mushroom Lentinellus ursinus and a termite-associated T. virens. , Other examples are the disydonols A-C, phenolic bisabolane sesquiterpenoid dimers isolated from a marine Aspergillus species. We next focused on nor-sesquiterpene homodimers from fungi (i.e., sesquiterpene dimers composed of two nor-sesquiterpene subunits), and to the best of our knowledge, anthracaurisin A (6) is the first and only fungal example. Plants, however, are able to produce nor-sesquiterpene homodimers like mexicanin F isolated from Helenium mexicanum and valeriadimer A from Valeriana officinalis. , Plants are also known to synthesize nor-sesquiterpene heterodimers (i.e., sesquiterpene dimers composed of one nor-sesquiterpene subunit and one sesquiterpene subunit) such as dinardokanshone A and valeriadimer B. , We believe anthracaurisin A is the first nor-sesquiterpene homodimer in fungi and one of the very few sesquiterpene dimer natural products. Biological Activity We tested anthracaurisin A’s antimicrobial activity and cytotoxic potential in vitro, as well as its toxicity in zebrafish in vivo. For the antimicrobial assays, we tested a panel of human-pathogenic Gram-positive bacteria (Staphylococcus and Enterococcus spp.), Gram-negative bacteria (Escherichia and Pseudomonas spp.), and fungi (Candida and Cryptococcus spp.). Anthracaurisin A displayed significant antimicrobial activity against the Gram-positive bacteria and the fungi, with MIC values comprised between 4 μg/mL (7.2 μM) and 8 μg/mL (14.4 μM). No activity against the selected Gram-negative bacteria was observed (Table ). 3Antimicrobial Activity of Compound 6 against Various Human Pathogens (MIC90, μg/mL),, pathogen6kanamycin vancomycin ampicillin amphotericin B capsofungin S. aureus 4 2 S. aureus (methicillin-res.)4 1 S. aureus (multidrug-res.)4 1 En. faecium 8 0.5 E. coli >250 4 P. aeruginosa >25032 C. albicans 4 1 C. albicans (fluconazole-res.)4 1 C. auris 8 0.5 Cr. neoformans 16 2 aValues are MIC in μg/mL, obtained from three independent experiments performed in duplicates. S., Staphylococcus; En., Enterococcus; E., Escherichia; P., Pseudomonas; C., Candida; Cr., Cryptococcus. bAverage value over three independent trials with MIC ranging from 2 to 8 μg/mL.cPositive controls. Antimicrobial Activity of Compound 6 against Various Human Pathogens (MIC90, μg/mL),, Values are MIC in μg/mL, obtained from three independent experiments performed in duplicates. S., Staphylococcus; En., Enterococcus; E., Escherichia; P., Pseudomonas; C., Candida; Cr., Cryptococcus. Average value over three independent trials with MIC ranging from 2 to 8 μg/mL. Positive controls. Cytotoxic activity on cancer cells was evaluated via the NCI-60 cell line panel screening at a one dose concentration of 10 μM. In the presence of anthracaurisin A, all cell lines exhibited considerable growth reduction, with percentage growth ranging from 0.25% (cell line HL-60­(TB)) to 57.19% (cell line NCI-H322M) compared to the negative control (no drug, 100% growth). Average growth of the cell line panel was 10.3%, demonstrating the broad-spectrum cytotoxic activity of anthracaurisin A against various cancer cell lines (Figure ). This was further corroborated by our in-house MTT-based cell viability assay using HCT-116 (colon carcinoma), MCF-7 (breast adenocarcinoma), and PC3 (prostate adenocarcinoma) cancer cell lines. Against these cells, IC50 values of 0.3, 2.3, and 0.5 μM of 6 were established (Table ). Lastly, the in vivo effects of anthracaurisin A were examined using a zebrafish embryo acute toxicity, yielding an LC50 value of 8.6 μM, similar to other mycotoxins discovered by our team (Figure ). Overall, anthracaurisin A has broad cytotoxic activity in the lower micromolar range and exhibits more potency than its aurisin congeners. , Future studies on its cellular mode of action are underway. Noteworthy, the monomer illudin S and its derivatives are potent DNA alkylating agents and completed a phase II clinical trial in late 2022 for oncology application (NCT03643107). , NCI one-dose 60-cell line panel screening of anthracaurisin A (6). Data are reported as a mean graph of the percent growth of treated cells with 10 μM anthracaurisin A (6). Growth is relative to the no-drug control and relative to the time zero number of cells. NCI identification number: NSC # D-856337/1; Experimental ID # 2507HT86. In vivo toxicity of anthracaurisin A (6) in zebrafish assay. Mortality dose–response curve for anthracaurisin A (6) in zebrafish embryo acute toxicity assay (LC50= 8.6 ± 3 μM, n = 30). 4Cytotoxic Activity of Compound 6 against Various Cancer Cell Lines (IC50, μM), mammalian cell line 6 mensacarcin HCT-116 (human colon carcinoma)0.3 ± 0.12.7 ± 1.5MCF-7 (human breast cancer)2.3 ± 0.71.9 ± 1.1PC3 (human prostate cancer)0.5 ± 0.21.4 ± 0.5aValues are mean IC50 ± standard deviations in μM, obtained from three independent experiments performed in triplicate.bPositive control. Cytotoxic Activity of Compound 6 against Various Cancer Cell Lines (IC50, μM), Values are mean IC50 ± standard deviations in μM, obtained from three independent experiments performed in triplicate. Positive control. Experimental Section General Experimental Procedures Illudin M (1), illudin S (3), and anthracaurisin A (6) 1D and 2D NMR spectra were recorded at 500 MHz for 1H NMR and 125 MHz for 13C NMR at 298 K on a Bruker Ascend 500 instrument equipped with a 5 mm probe (Bruker Corporation, Billerica, MA, USA). Chloroform-d was used for illudin M and anthracaurisin A and methanol-d 4 for illudin S. Using the same instrument, 1D and 2D 1H and 13C NMR spectra under increasing temperature of 323, 348, and 373 K were recorded for anthracaurisin A in dimethyl sulfoxide-d 6. Illudin B (2), 15-O-acetylilludin S (4), and dichomitin A (5) 1D and 2D NMR spectra were recorded at 600 MHz for 1H NMR and 150 MHz for 13C NMR at 298 K on a Bruker Avance Neo instrument equipped with a 1.7 mm TCI Micro CryoProbe in methanol-d 4. Solvents were used as internal standards and referenced as follows: δH 7.26 ppm, δC 77.16 ppm for chloroform-d; δH 3.31 ppm, δC 49.00 for methanol-d 4 and δH 2.50 ppm, δC 39.52 for dimethyl sulfoxide-d 6 . Optical rotations were measured on a JASCO P-2000 polarimeter at the sodium D line (589 nm), and they were averaged over ten integration periods (JASCO Applied Sciences, Halifax, Canada). We did not establish the enantiomeric excess or purity of 1–6 via chiral chromatography, as most natural products are made entiopure. Therefore, we cannot exclude that these compounds exist as scalemic mixtures. IR spectra were acquired on a Bruker α II FTIR spectrophotometer, and UV spectra were acquired using a Thermo Scientific Evolution 220 UV–visible spectrophotometer. VCD spectra were acquired in chloroform-d with a ChiralIR 2X Dual PEM FT-VCD spectrometer, set to 4 cm–1 resolution, with PEM (both 1 and 2) maximum frequency set to 1400 cm–1 (BioTools Inc., Jupiter, FL, USA). ECD spectra were acquired in acetonitrile using a Chirascan V100 instrument (Applied Photophysics, Leatherhead, England) in three trials at 1 s per step, and the results were averaged. Low-resolution LC-MS experiments were conducted on an Agilent 1100 series HPLC coupled to an Agilent 6130 Quadrupole LC/MS mass spectrometer with an electrospray ionization (ESI) source (Agilent Technologies, Santa Clara, CA, USA). Chromatography was performed using a Kinetex C18 column (150 mm × 4.6 mm, 5 μm, 100 Å, Phenomenex, Torrance, CA, USA), the oven was turned off, and the sample injection volume was 10–20 μL. A binary gradient consisting of water (eluent A) and acetonitrile (eluent B) (both +0.05% formic acid (FA)) at a constant flow rate of 800 μL/min was used. The gradient was applied as follows: 0–2 min, 10% B; 35 min, 100% B; 38 min, 100% B; 40 min, 10% B; 42 min, 10% B. For data acquisition and subsequent qualitative analysis, Agilent ChemStation software was used. High-resolution LC-MS and LC-MS/MS experiments were conducted on an Agilent 1290 Infinity II series UPLC coupled to an Agilent 6546 Q-TOF mass spectrometer with an electrospray ionization (Jet Stream ESI) source. Liquid chromatography was performed using a Kinetex C18 column (50 mm × 2.1 mm, 2.6 μm, 100 Å, Phenomenex, Torrance, CA, USA), the oven temperature was set to 40 °C, and the sample injection volume was 3–5 μL. A binary gradient consisting of water (eluent A) and acetonitrile (eluent B) (both +0.05% FA) at a constant flow rate of 400 μL/min was used. The gradient was applied as follows: 0–0.25 min, 10% B; 10.00 min, 95% B; 13.00 min, 95% B; 13.1 min, 10% B; 15.00 min, 10% B. For data acquisition and subsequent qualitative analysis, Agilent MassHunter software was used. Preparative HPLC experiments were carried out on an Agilent 1260 Infinity II series apparatus exhibiting a Luna C18 column (AXIA, New Column 250 mm × 21.1 mm, 5 μm, 100 Å, Phenomenex, Torrance, CA, USA). Semipreparative HPLC experiments were performed on an Agilent 1100 series apparatus equipped with a Luna C18 column (AXIA, New Column 250 mm × 10 mm, 5 μm, 100 Å, Phenomenex, Torrance, CA, USA). Fungal genomic DNA for barcoding experiments was extracted using the DNeasy Blood and Tissue Kit (Qiagen, Germantown, MD, USA). PCR was performed in a MyCycler thermal cycler (Bio-Rad Laboratories Inc., Hercules, CA, USA). PCR products were submitted to DNA gel electrophoresis in 1% agarose gel in the presence of GelRed nucleic acid stain (Biotium Inc., Fremont, CA, USA) and visualized using a Bio-Rad UV Transilluminator 2000 or an Azure Gel Imager 400 (Azure Biosystems, Dublin, CA, USA). All solvents were purchased from Thermo Fisher (Thermo Fisher Scientific Inc., Waltham, MA, USA) or VWR (VWR International, Radnor, PA, USA). Fungal Strain Collection and Identification The fungus was found on dead indeterminate branches at Fundo La Cantera y El Guindo, Concepción, Biobío Region, Chile (−36.837173° S, −73.025636° W) and collected in August 2019. It was later identified as A. discolor based on morphological features. A voucher specimen (VALD-F 0003) has been deposited in the VALDF herbarium of the Universidad Austral de Chile in Valdivia, Chile. To further identify the species by genomic tools, its ITS was sequenced. Briefly, a small culture of the fungus was grown for 3 weeks in 125 mL liquid malt-based medium. The mycelium was collected, and genomic DNA was extracted using the DNeasy Blood and Tissue Kit from Qiagen following the manufacturer’s guidelines. The ITS region was amplified by PCR using primers ITS1 and ITS4 in association with the commercial master mix MangoMix (Meridian Bioscience, Cincinnati, OH, USA). Resulting amplicons were then sequenced by Sanger sequencing (Azenta Life Science, South Plainfield, NJ, USA). In total, eight ITS sequences were retrieved, each originating from an independent replicate of DNA extraction. All sequences were pairwise aligned with the reference databases FunCBS, EF1, ITS, Unite, Mycokey, fusarioidID, Morchella, PasteurYeasts, PasteurMOLDS, and MIRRI in MycoBank. All sequences aligned with Anthracophyllum spp. sequences, sometimes specified to A. archeri or A. lateritium. Out of the eight sequences aligned with the databases, four showed a similarity score of 94% or above with at least one Anthracophyllum sp. (Figure S65). Fermentation, Extraction, and Isolation The fungus was grown in two 500 mL glass mason jars filled with autoclaved rice medium (50% v/v rice to 50% v/v DI water) for 4 weeks without shaking. After this period of fermentation, the jars were extracted with ethyl acetate by maceration and manual stirring followed by filtration through a paper filter. The resulting liquid was concentrated in vacuo to obtain the organic extract (974 mg). A portion of this extract (200 mg) was subjected to preparative HPLC applying an acetonitrile/water gradient from 10 to 50% acetonitrile in 35 min, followed by a 98% acetonitrile wash for 10 min, on a Phenomenex Luna column (C18(2), 5 μm, 100 Å, AXIA, 250 mm × 21.1 mm) at a flow rate of 28.32 mL/min. Compound 1 (3.6 mg (yield 18.0 mg/L), R t 23.6 min), 2 (0.5 mg (yield 2.5 mg/L), R t 15.9 min), 3 (6.0 mg (yield 30.0 mg/L), R t 11.1 min), 6 (30.0 mg (yield 150.0 mg/L), R t 39.2 min), and a mixture of 4 and 5 (3.0 mg, R t 20.7 min) were isolated. The later was further purified by semipreparative HPLC employing an acetonitrile/water gradient composed of 20% isocratic acetonitrile for 35 min, followed by a 98% acetonitrile wash for 7 min, on a Phenomenex Luna column (C18(2), 5 μm, 100 Å, AXIA, 250 mm × 10 mm) at a flow rate of 5.0 mL/min. This experiment allowed the isolation of compound 4 (1.3 mg (yield 6.5 mg/L), R t 26.2 min) and 5 (1.8 mg (yield 9.0 mg/L), R t 33.4 min). 15-O-acetylilludin S (4): yellowish oil; [α]D 22 −80° (c 0.1, MeOH); UV/vis (MeOH) λmax (log ε) 203.4 (3.87), 222.4 (3.95), 316.0 (3.31) nm (Figure S10); ECD (acetonitrile) λmax (Δε) 216 (+64.6), 249 (−36.6), 271 (−2.2), 340 (−7.5) nm; IR νmax 3457, 2939, 1700, 1609, 1375, 1237, 1107, 1032, 945, 825, 764 cm–1 (Figure S9); 1H and 13C NMR (Table ); ESIMS m/z 351.1 [M + FA–H]−, m/z 341.1 [M + Cl]−, m/z 289.1 [M–H2O + H]+, m/z 307.1 [M + H]+, m/z 324.1 [M + NH4]+, m/z 635.3 [2M + Na]+ (Figure S7); HRESIMS m/z 245.1192 [M–acetyl–H2O–H]− (calcd for C15H17O3 –, 245.1183), m/z 289.1435 [M–H2O + H]+ (calcd for C17H21O4 +, 289.1434), m/z 329.1361 [M + Na]+ (calcd for C17H22O5Na+, 329.1359) (Figure S8). Anthracaurisin A (6): yellowish oil; [α]D 22 +64° (c 0.1, MeOH), + 76° (c 0.1, CHCl3); UV/vis (MeOH) λmax (log ε) 201.3 (3.77), 286.0 (4.15) nm (Figure S45); ECD (acetonitrile) λmax (Δε) 185 (+11.4), 217 (+0.5), 279 (+15.0), 318 (−6.1) nm; IR νmax 2967, 1751, 1683, 1619, 1462, 1373, 1226, 1179, 1133, 1027, 935, 861, 795 cm–1 (Figure S44); 1H and 13C NMR (Tables , S37, and S47); ESIMS m/z 553.2 [M–H]−, m/z 495.2 [M–acetyl–H2O + H]+, m/z 435.2 [M–2acetyl–2H2O + H]+ (Figure S40); HRESIMS m/z 553.2831 [M–H]− (calcd for C32H41O8 –, 553.2807), m/z 575.2643 [M + Na–2H]− (calcd for C32H40O8Na–, 575.2626), m/z 511.2714 [M–acetyl–H]− (calcd for C30H39O7 –, 511.2701), m/z 493.2601 [M–acetyl–H2O–H]− (calcd for C30H37O6 –, 493.2596) (Figure S41). VCD and ECD Measurements and Computations Compounds subjected to VCD analyses were solubilized in chloroform-d, which was run through a small plug of activated basic alumina immediately before use. The resulting solution was transferred to a liquid IR cell (BaF2, 100 μm cell path) and placed in the measurement chamber. The VCD was then acquired (BioTools Inc. ChiralIR 2X Dual PEM FT-VCD spectrometer) at 4 cm–1 resolution, with PEM (both 1 and 2) maximum frequency set to 1400 cm–1. The sample was measured for several hours in 1 h blocks. The IR data from the first block were solvent and water vapor subtracted and then offset to zero at 2000 cm–1. The VCD data blocks were averaged, and from this was subtracted a previously measured solvent baseline. Finally, the VCD spectrum was offset to zero at 2000 cm–1. The VCD noise data was block averaged and used without further processing. Compounds subjected to ECD measurements were solubilized in acetonitrile and transferred to a 500 μm UV–vis cell. The absorbance was checked, and the solution was diluted until an absorbance of 0.8 was achieved for the UV–vis max absorbance (285 nm). The UV–vis and ECD spectra were then acquired (Applied Photophysics Chirascan V100 spectrometer) in three trials at 1 s per step, and the results were averaged. The solvent was measured under the same conditions and was used to correct both the UV–vis and ECD spectra. For each compound, the different candidate stereoisomers were constructed using ComputeVOA (BioTools, Jupiter, FL, USA). A thorough conformational search was performed at the MM level using the MMF94 force field in a 7 kcal/mol energy window. The conformers found were subjected to DFT level optimization and frequency calculation with Gaussian ′09 (Gaussian Inc., Wallingford, CT, USA) using the B3LYP/6–31G­(d) and B3PW91/6–31G­(d) methods. The resulting lowest energy unique conformations were reoptimized using the B3LYP/cc-pVTZ and B3PW91/cc-pVTZ methods, and the IR and VCD frequencies recalculated at these levels. The resulting spectra from all methods were Boltzmann averaged (using both free energy and electronic energy), plotted at 5 cm–1 resolution, and then uniformly x-axis scaled (values obtained using CompareVOA and varied with basis set and functional from 0.968 to 0.984) for comparison to the experimental IR and VCD spectra. The two different weighing methods (free energy and electronic energy) were compared to the experimental data. The same approach was applied to each of the alternate structures, and the same method was used to produce spectra for these stereoisomers. The B3PW91/cc-pVTZ calculations gave the best match to the experiment and are used in all plots and comparisons. The optimized structures of conformers obtained from the VCD calculation were subjected to UV–vis/ECD calculation using the TD-DFT method in Gaussian ′09 at the CAM-B3LYP/TZVP level for 200 excited states. The results were Boltzmann averaged using free energy and plotted at 0.333 eV HWHH and then uniformly scaled for comparison to the experiment. Cell Culture Assays Colon cancer (HCT-116), breast adenocarcinoma (MCF-7), and prostate adenocarcinoma (PC-3) cells were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). Dulbecco’s modified Eagle’s medium (DMEM), RPMI 1640, phosphate-buffered saline (PBS), trypsin/EDTA (0.25%/2.21 mM), and penicillin/streptomycin solution were obtained from Thermo Fisher Scientific. Fetal bovine serum (FBS) was obtained from R&D systems. HCT-116 and MCF-7 cells were cultivated in DMEM and PC-3 cells in RPMI 1640, each supplemented with 10% (v/v) fetal bovine serum, penicillin (100 U/mL), and streptomycin (100 μg/mL). The cell lines were incubated at 37 °C, 5% CO2. Cells were plated into 96-well plates at different densities (HCT-116:7000 cells/well; PC3:7000 cells/well, MF7:9000 cells/well). The passage number for cells used in the experiments never exceeded 15. All cell lines were tested mycoplasma-negative by real-time PCR (Myco Solutions mycoplasma detection kit, Akron Biotech, Boca Raton, FL, USA). Growth inhibition and cytotoxicity in the different cell lines were measured by the reduction of the tetrazolium salt MTT by metabolically active cells. Cells were plated into 96-well plates and maintained overnight before treatment was started with the addition of mensacarcin to each well. After the designated time, MTT (5 mg/mL in PBS) was added to each well at a final concentration of 0.5 mg/mL. The plates were incubated for 2 h at 37 °C. The medium was removed, cells were lysed, and the purple formazan product was solubilized by the addition of 50 μL of DMSO. Absorbance was measured at 550 nm with a microplate reader (Synergy HTX, Biotek, Winooski, VT, USA). Metabolic activity of vehicle-treated cells (0.5% DMSO unless otherwise stated) was defined as 100% cell growth. Antimicrobial Assays For antibacterial activity in cell-based assays, established protocols were followed. , Gram-positive bacteria, including Enterococcus faecium (ATCC 49032), Staphylococcus aureus (ATCC 25923), methicillin-resistant S. aureus (ATCC BAA-41), and multidrug-resistant S. aureus (ATCC BAA44), two Gram-negative pathogens, Pseudomonas aeruginosa (ATCC 15442) and Escherichia coli (ATCC 8739), as well as several fungi such as Candida albicans (ATCC 90027), fluconazole-resistant C. albicans (ATCC 10231), Cryptococcus neoformans (ATCC 14116), and C. auris (CDC B11903) were tested. Vancomycin, kanamycin, ampicillin, amphotericin B, fluconazole, and caspofungin were used as positive controls. DMSO served as the negative control. MICs were determined based on two independent experiments performed in duplicates. For the single-dose assay, compounds, mixtures, and antibiotic controls were tested at 125 μg/mL. Microbial growth rates were measured after 16 h by absorbance at 620 nm using a Biotek Synergy 96-well plate reader (BioTek, Winooski, VT, USA). All human pathogens used in the study were acquired from the ATCC. Zebrafish Assay In Vivo Toxicity Zebrafish adults (Danio rerio, AB strain) were reared under standard conditions (28 °C, 14:10 light:dark) and crossed to produce embryos. Embryos with intact chorions were transferred into 96-well plates (1 embryo/well) at 24 h postfertilization (hpf) using 50 μL of medium (HEPES-buffered E3:5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2, 0.33 mM MgSO4, 1 mM HEPES, titrated to 7.2 pH). Compound 6 was dissolved in DMSO, serially diluted using HEPES-buffered E3, and then added to well plates (final conc.: 0, 2.0, 5.0, 15, 25, and 50 μM in 1% DMSO). Plates were incubated for 48 h (72 hpf; 28 °C, 14:10 light:dark), after which the number of surviving embryos was counted. Each assay examined eight embryos per concentration, and the assay was replicated three times using animals from three clutches. The results were similar across replicates, and so survivorship is plotted from pooled data, with error bars representing 68% confidence intervals (like SEM) found by fitting results for each concentration to a binomial distribution (MATLAB R2020b). EC50 values were calculated by fitting unpooled survivorship values to the Hill equation, and 95% confidence intervals were calculated using a bootstrap analysis. Protocols were approved by the University of Florida’s Institutional Animal Care and Use Committee. Supplementary Material