Molecular Networking and Cultivation Profiling Reveals Diverse Natural Product Classes from an Australian Soil-Derived Fungus Aspergillus sp. CMB-MRF324

This study showcases the application of an integrated workflow of molecular networking chemical profiling (GNPS), together with miniaturized microbioreactor cultivation profiling (MATRIX) to successfully detect, dereplicate, prioritize, optimize the production, isolate, characterize, and identify a diverse selection of new chemically labile natural products from the Queensland sheep pasture soil-derived fungus Aspergillus sp. CMB-MRF324. More specifically, we report the new tryptamine enamino tripeptide aspergillamides E–F (7–8), dihydroquinoline-2-one aflaquinolones H–I (11–12), and prenylated phenylbutyrolactone aspulvinone Y (14), along with an array of known co-metabolites, including asterriquinones SU5228 (9) and CT5 (10), terrecyclic acid A (13), and aspulvinones N-CR (15), B (16), D (17), and H (18). Structure elucidation was achieved by a combination of detailed spectroscopic and chemical analysis, biosynthetic considerations, and in the case of 11, an X-ray crystallographic analysis.


Introduction
During our ongoing investigations into the chemistry of Australian microbes, we have reported numerous new structurally diverse natural products from many genera of fungi obtained from geographically dispersed and varying substrates. For example, our discoveries from the genus Aspergillus include a 2003 account of the depsi-cyclohexapeptide aspergillicins A-E from a Tasmanian estuarine sediment-derived Aspergillus carneus MST-MF156 [1]; a 2005 account of the heterocyclic dipeptide aspergillazines A-E from a remote outback Queensland soil-derived A. unilateralis MST-F8675 [2]; a 2009 account of the nor-methyl fumiquinazoline cottoquinazoline A and cyclopentapeptide cotteslosins A-B from a Western Australian beach sand-derived A. versicolor MST-MF495 [3]; a 2014 account of the P-glycoprotein inhibitory diketomorpholine shornephine A from a Queensland intertidal marine sediment-derived Aspergillus sp. CMB-M081F [4]; and a 2017 account of the nitro depsi-tetrapeptide diketopiperazine waspergillamide A from a Queensland mud dauber wasp-derived Aspergillus sp. CMB-W031 [5].
In recent years, these microbial biodiscovery efforts have been enhanced by the use of an integrated workflow consisting of global natural products social (GNPS) molecular networking [6] to detect, dereplicate, and prioritize, and a miniaturized microbioreactor methodology (MATRIX) [7] employing multiple different media under solid agar as well as static and shaken broth conditions to optimize the production of novel natural products. For example, in 2021 we reported a brown rice cultivation of a Queensland sheep pasture soil-derived Aspergillus sp. CMB-MRF324 yielding the acetylcholinesterase inhibitory meroterpene millmerranones A-F (1-6) ( Figure 1), with millmerranone A being noteworthy as a rare example of a natural cyclic carbonate [8]. At that time, the GNPS analysis of a This current report describes just such an investigation, where fractionation of the CMB-MRF324 SDA extract yielded new linear tripeptide aspergillamides E-F (7)(8) as equilibrating E/Z amide rotamers incorporating a rare tryptamine enamine moiety, along with the known asterriquinones SU5228 (9) and CT5 (10) (Figure 1). In addition, we provide an account of a GNPS-MATRIX analysis of CMB-MRF324 that prompted discovery This current report describes just such an investigation, where fractionation of the CMB-MRF324 SDA extract yielded new linear tripeptide aspergillamides E-F (7)(8) as equilibrating E/Z amide rotamers incorporating a rare tryptamine enamine moiety, along with the known asterriquinones SU5228 (9) and CT5 (10) (Figure 1). In addition, we provide an account of a GNPS-MATRIX analysis of CMB-MRF324 that prompted discovery of the new dihydroquinoline-2-one aflaquinolones H (11) and I (12), the known terrecyclic acid A (13), and the new prenylated phenylbutyrolactone aspulvinone Y (14), and known aspulvinones N-CR (15), B (16), D (17) and H (18) (Figure 1). The detection, optimized production, isolation, and structure elucidation of 7-18 are outlined below.

Results and Discussion
A GNPS molecular network analysis of a selection of EtOAc extracts prepared from ×27 solid phase plate cultivations of pasture-derived fungi (SDA) and bacteria (ISP-2 agar) revealed two molecular families (clusters) unique to the fungus CMB-MRF324, both correlating with prominent peaks in the extract UPLC-DAD chromatogram (Figure 2A-C). Subsequent scaled-up SDA cultivation of CMB-MRF324 followed by extraction with EtOAc, solvent trituration and reversed phase chromatography ( Figure S1) yielded the new tryptamine enamino tripeptide aspergillamides E-F (7-8) along with the known asterriquinones SU5228 (9) and CT5 (10) (Figure 1). Similarly, a GNPS based re-analysis of a brown rice cultivation of this same CMB-MRF324 fungus revealed a molecular family featuring potentially new chemistry, also evident in the UPLC-DAD chromatogram along with two other classes of metabolite ( Figure 3A,B). Subsequent reversed phase and gel (LH-20) chromatography fractionation of a scaled up brown rice cultivation ( Figure S2) yielded the new dihydroquinoline-2-one aflaquinolones H (11) and I (12) and prenylated phenylbutyrolactone aspulvinone Y (14), along with the known terrecyclic acid A (13) and aspulvinones N-CR (15), B (16), D (17), and H (18) (Figure 1). Structures for the known natural products 9 (Table S5 and Figures S19-S20) [9], 10 (  [17,18], and 18 ( Figures S56-S57) [19] were established by spectroscopic analysis and comparison to data in the literature. Of interest, although 10 possesses a plane-of-symmetry, this was not manifested in the 1D NMR data acquired in DMSO-d6 ( Figure S21) or methanol-d4 ( Figure S22), but was evident in the data acquired in CDCl3 ( Figures S23 and S24). A possible explanation for Of interest, although 10 possesses a plane-of-symmetry, this was not manifested in the 1D NMR data acquired in DMSO-d 6 ( Figure S21) or methanol-d 4 ( Figure S22), but was evident in the data acquired in CDCl 3 ( Figures S23 and S24). A possible explanation for this solvent effect (which has not been previously described) is that H-bonding in DMSO-d 6 and methanol-d 4 restricts/slows intramolecular rotations sufficiently to generate distinct and NMR detectable populations of atropisomeric stereoisomers, differentiating between a racemic mixture of P,P and M,M enantiomers on the one hand, and the meso P,M diastereomer on the other. As such, the apparent loss of symmetry in the NMR data is misleading and is perhaps better characterized as the detection of two separate isomer populations, each retaining a plane-of-symmetry. These phenomena suggest a need for caution when assessing the biological activity (and attempting any in silico modelling) of otherwise pure achiral (symmetric or asymmetric) compounds where H-bond mediated solvent effects (i.e., induced by aqueous assay media) can present as mixed populations of atropisomeric stereoisomers. An account of the structure elucidation of the new natural products 7-8, 11-12, and 14 is outlined below.
Molecules 2022, 27, x FOR PEER REVIEW 4 of 18 this solvent effect (which has not been previously described) is that H-bonding in DMSO-d6 and methanol-d4 restricts/slows intramolecular rotations sufficiently to generate distinct and NMR detectable populations of atropisomeric stereoisomers, differentiating between a racemic mixture of P,P and M,M enantiomers on the one hand, and the meso P,M diastereomer on the other. As such, the apparent loss of symmetry in the NMR data is misleading and is perhaps better characterized as the detection of two separate isomer populations, each retaining a plane-of-symmetry. These phenomena suggest a need for caution when assessing the biological activity (and attempting any in silico modelling) of otherwise pure achiral (symmetric or asymmetric) compounds where H-bond mediated solvent effects (i.e., induced by aqueous assay media) can present as mixed populations of atropisomeric stereoisomers. An account of the structure elucidation of the new natural products 7-8, 11-12, and 14 is outlined below. HRESI(+)MS measurement on 7 returned a molecular formula (C27H32N4O3, Δmmu-1.9) requiring 14 double bond equivalents (DBE). The 1D NMR (DMSO-d6) data for 7 (Tables 1, 2, S1, and S2 and Figures 4 and S5-S10) revealed an isomeric mixture (ratio 1:0.6). Analysis of NMR resonances attributed to the major isomer 7a revealed a mono substituted benzene, a C-2 mono substituted indole, an E 1,2-disubstituted double bond (J8, 9 15 Hz), and three carbonyl carbons accounting for all DBE. Diagnostic 2D NMR correlations allowed assembly of the tripeptide scaffold for 7a, with ROESY correlations between the terminal N-acetyl-valine and associated N-methyl amide moiety establishing 7a as the E rotamer about the valine amide bond. Comparable analysis of the minor isomer 7b confirmed the same planar structure with ROESY correlations between the phenylalanine and valine, a methines consistent with the Z rotamer about the valine amide bond. A Marfey's analysis performed on 7 confirmed incorporation of L-valine and N-methyl-L-phenylalanine residues ( Figure S4), permitting assignment of the structure for aspergillamide E (7) as shown.  (Tables 1 and 2, Tables S1 and S2 and Figure 4 and Figures S5-S10) revealed an isomeric mixture (ratio 1:0.6). Analysis of NMR resonances attributed to the major isomer 7a revealed a mono substituted benzene, a C-2 mono substituted indole, an E 1,2-disubstituted double bond (J 8, 9 15 Hz), and three carbonyl carbons accounting for all DBE. Diagnostic 2D NMR correlations allowed assembly of the tripeptide scaffold for 7a, with ROESY correlations between the terminal N-acetyl-valine and associated N-methyl amide moiety establishing 7a as the E rotamer about the valine amide bond. Comparable analysis of the minor isomer 7b confirmed the same planar structure with ROESY correlations between the phenylalanine and valine, a methines consistent with the Z rotamer about the valine amide bond. A Marfey's analysis performed on 7 confirmed incorporation of L-valine and N-methyl-L-phenylalanine residues ( Figure S4), permitting assignment of the structure for aspergillamide E (7) as shown. Molecules 2022, 27, x FOR PEER REVIEW HRESI(+)MS measurement on 8 returned a molecular formula (C27H32N4O3 +2.5) isomeric with 7. Comparison of the 1D and 2D NMR (DMSO-d6) data for 8 (T and 2 and S3 andS4 and Figures 4 and S12-S17) with 7 revealed the same equil major E (8a) and minor Z (8b) amide rotamers with key differences between 8 a tributed to a Z enamino configuration (J8, 9 9.5 Hz). A Marfey's analysis perform confirmed the incorporation of L-valine and N-methyl-L-phenylalanine residues S4), permitting assignment of the structure for aspergillamide F (8) as shown. Ale potential for chemically reactive natural products to form handling artifacts [20] termined that the Z geometric isomer 8 isomerizes on standing, under either a neutral conditions, to the E isomer 7 and that this process is accelerated on heatin ertheless, chemical analysis of a fresh extract detected both 7 and 8, confirming the as natural products.   HRESI(+)MS measurement on 8 returned a molecular formula (C 27 H 32 N 4 O 3 , ∆mmu +2.5) isomeric with 7. Comparison of the 1D and 2D NMR (DMSO-d 6 ) data for 8 (Tables 1  and 2, Tables S3 and S4 and Figure 4 and Figures S12-S17) with 7 revealed the same equilibrating major E (8a) and minor Z (8b) amide rotamers with key differences between 8 and 7 attributed to a Z enamino configuration (J 8,9 9.5 Hz). A Marfey's analysis performed on 8 confirmed the incorporation of L-valine and N-methyl-L-phenylalanine residues ( Figure S4), permitting assignment of the structure for aspergillamide F (8) as shown. Alert to the potential for chemically reactive natural products to form handling artifacts [20], we determined that the Z geometric isomer 8 isomerizes on standing, under either acidic or neutral conditions, to the E isomer 7 and that this process is accelerated on heating. Nevertheless, chemical analysis of a fresh extract detected both 7 and 8, confirming their status as natural products.
Tryptamine enamino tripeptides have appeared in the natural products literature on several occasions over the last 25 years, starting in 1997 with terpeptin (19) from a soilderived Aspergillus terreus 95F-1 collected near Naha City, Okinawa Prefecture, Japan [21]; and progressing through 1998 with aspergillamides A-B (20-21) from a saline lake sedimentderived Aspergillus sp. CNC-120 collected in the Bahamas [22]; 2002 with miyakamides A 1 , A 2 , B 1, and B 2 (22-25) from a fallen leaf-derived Aspergillus flavus Link var. columnaris FKI-0739 collected off Miyakojima Island, Japan [23]; 2008 with terpeptins A-B (26-27) from the mangrove plant (Acanthus ilicifolius) epiphyte Aspergillus sp. W-6 collected in Dongzhai Gang, China [24]; 2010 with JBIR-81 (28) and JBIR-82 (29) from the seaweed (Sargassum sp.) derived Aspergillus sp. SpD081030G1f1 collected off Ishigaki Island, Okinawa Prefecture, Japan [25]; and 2019 with aspergillamides C-D (30-31) from a marine sponge-derived Aspergillus terreus SCSIO 41008 collected off Guangdong Province, China [26] (Figure 5). It is noteworthy that all these reports were from fungi of the genus Aspergillus, albeit isolated from geographically diverse terrestrial and marine substrates. Encouraged by the relatively small number of reported members of this class of natural product, we returned our attention to the GNPS molecular network analysis and the MS/MS fragmentations that defined the aspergillamide molecular family. With diagnostic losses of 158 and 301 amu revealing the sodiated ion for the N-terminal amino acid, it was possible to detect nodes consistent with the known aspergillamides A-D (20-21, 30-31), and for putative new very minor homologues, homo-Ala (i) and Ala (ii) (Figure 6 and Figures S58-S61).       (Table 3 and Table S7, Figures S26-S31) revealed resonances for twelve sp 2 carbons, of which eight could be attributed to two aromatic rings and four to an E 1,2-disubstituted double bond, an ester/amide, and a ketone carbonyl, accounting eleven DBE and requiring that 11 incorporate two additional rings. Diagnostic 2D NMR correlations permitted assembly of a planar structure (Figure 7) with a single crystal X-ray crystallographic analysis (Figure 8) unambiguously establishing the structure and absolute configuration of aflaquinolone H (11). revealed resonances for twelve sp 2 carbons, of which eight could be attributed to two aromatic rings and four to an E 1,2-disubstituted double bond, an ester/amide, and a ketone carbonyl, accounting eleven DBE and requiring that 11 incorporate two additional rings. Diagnostic 2D NMR correlations permitted assembly of a planar structure (Figure 7) with a single crystal X-ray crystallographic analysis (Figure 8) unambiguously establishing the structure and absolute configuration of aflaquinolone H (11).   Remarkably, the crystal structure of 11 comprises six independent molecules in the asymmetric unit; molecule A is shown in Figure 8. All molecules adopt similar conformations with the only significant difference being the orientation of the 4-methoxy group on the phenyl ring which is found in one of two conformers related by a 180 • rotation. An overlay of the six molecules is shown in the Supplementary Materials. A strong intramolec-ular H-bond (O5A-H . . . O4A 1.89 Å, 144.9 • ) corresponding to hydroxyl groups in the 4and 6-positions of the 3, 4-dihydroquinolinone ring system is apparent in Figure 8, and this is present in all six molecules comprising the structure. This H-bond restrains the 4-hydroxy group (O4A) in an equatorial position, and this locks the confirmation of the bicyclic ring system leaving the p-methoxyphenyl substituent in an axial position. The absolute configuration of 11 was established by the Bijvoet analysis of Hooft and co-workers; P2 = 1.000, Hooft parameter y = −0.06(9) [27]. Remarkably, the crystal structure of 11 comprises six independent molecules in the asymmetric unit; molecule A is shown in Figure 8. All molecules adopt similar conformations with the only significant difference being the orientation of the 4-methoxy group on the phenyl ring which is found in one of two conformers related by a 180° rotation. An overlay of the six molecules is shown in the Supplementary Materials. A strong intramolecular H-bond (O5A-H…O4A 1.89 Å, 144.9°) corresponding to hydroxyl groups in the 4and 6-positions of the 3, 4-dihydroquinolinone ring system is apparent in Figure 8, and this is present in all six molecules comprising the structure. This H-bond restrains the 4hydroxy group (O4A) in an equatorial position, and this locks the confirmation of the bicyclic ring system leaving the p-methoxyphenyl substituent in an axial position. The absolute configuration of 11 was established by the Bijvoet analysis of Hooft and co-workers; P2 = 1.000, Hooft parameter y = −0.06(9) [27].
HRESI(+)MS measurement on 12 revealed a molecular formula (C27H33NO6, Δmmu -1.0), consistent with a hydrogenated analogue of 11. Comparison of the NMR (CDCl3) data for 12 (Tables 3 and S8 (Table 3 and Table S8 (Table 4 and Table S10, Figures S43-S46) with the 1D NMR (acetone-d 6 ) data for 16 ( Figure S52, Note-16 proved insoluble in methanol) revealed a high level of similarity, with key differences attributed to replacement of the isoprene D 8 ,9 moiety in 16 (d H 5.36, m, H-8 ; δ C 123.3, C-8 ; δ C 133.1, C-9 ) with a dioxygenated moiety in 14 (δ H 3.82, dd, H-8 ; δ C 69.4, C-8 ; δ C 78.3, C-9 ). The latter resonances could in principle be attributed to three possible substructures (Figure 10) consisting of a dihydrobenzofuran, a dihydrobenzopyran, or an epoxide. The dihydrobenzofuran moiety (Figure 10a) is prominent among known aspulvinones ( Figure S62) and features a consistent range of 13 C NMR (methanol-d 4 ) chemical shifts for C-8 (δ C 89.9-91.4) and C-9 (δ C 71.1-73.1) [16,31]. While the dihydrobenzopyran moiety (Figure 10b) is less common among known aspulvinones ( Figure S62), there are nevertheless reports of the 13 C NMR (methanol-d 4 ) chemical shifts for C-8 (δ C 68.7-71.0) and C-9 (δ C 77.2-77.5) [16]. Finally, although there are no known accounts of aspulvinones featuring an epoxy moiety (Figure 10c), there are reports of 13 C NMR chemical shifts that are relevant to both C-8 (δ C 60-65) and C-9 (δ C 57-59) [32]. These spectroscopic comparisons, together with diagnostic 2D NMR correlations (Figure 11), and biogenetic considerations, permitted assignment of the structure for aspulvinone Y (14) as shown. The sample of 14 proved unstable and decomposed during NMR data acquisition, precluding measurement of an optical rotation and consideration of absolute configuration.  Figure S62) and features a consistent range of 13 C NMR (methanol-d4) chemical shifts for C-8″ (δC 89.9-91.4) and C-9″ (δC 71.1-73.1) [16,31]. While the dihydrobenzopyran moiety (Figure 10b) is less common among known aspulvinones ( Figure S62), there are nevertheless reports of the 13 C NMR (methanol-d4) chemical shifts for C-8″ (δC 68.7-71.0) and C-9″ (δC 77.2-77.5) [16]. Finally, although there are no known accounts of aspulvinones featuring an epoxy moiety ( Figure  10c), there are reports of 13 C NMR chemical shifts that are relevant to both C-8″ (δC 60-65) and C-9″ (δC 57-59) [32]. These spectroscopic comparisons, together with diagnostic 2D NMR correlations (Figure 11), and biogenetic considerations, permitted assignment of the structure for aspulvinone Y (14) as shown. The sample of 14 proved unstable and decomposed during NMR data acquisition, precluding measurement of an optical rotation and consideration of absolute configuration.   although there are no known accounts of aspulvinones featuring an epoxy moiety ( Figure  10c), there are reports of 13 C NMR chemical shifts that are relevant to both C-8″ (δC 60-65) and C-9″ (δC 57-59) [32]. These spectroscopic comparisons, together with diagnostic 2D NMR correlations (Figure 11), and biogenetic considerations, permitted assignment of the structure for aspulvinone Y (14) as shown. The sample of 14 proved unstable and decomposed during NMR data acquisition, precluding measurement of an optical rotation and consideration of absolute configuration.   A review of the scientific literature revealed a rich history of fungal natural products belonging to the aspulvinone structure class. While these originated with 1973 [17] and 1975 [18] accounts of the pulvinones from Aspergillus terreus IAM 2054, the trivial nomenclature and structures were subsequently revised in 1976 [19] and 1979 [33] as aspulvinones A-I. These early discoveries were followed by reports of new members of this A review of the scientific literature revealed a rich history of fungal natural products belonging to the aspulvinone structure class. While these originated with 1973 [17] and 1975 [18] accounts of the pulvinones from Aspergillus terreus IAM 2054, the trivial nomenclature and structures were subsequently revised in 1976 [19] and 1979 [33] as aspulvinones A-I. These early discoveries were followed by reports of new members of this structure class, including a 2011 account of aspulvinones I-CR to N-CR from a Costa Rican marine sediment-derived Aspergillus sp. 05545 [16]; a 2013 account of isoaspulvinone E from a Fijian marine mangrove rhizosphere soil-derived Aspergillus terreus Gwq-48 [34]; a 2013 account of aspulvinone B1 from Aspergillus terreus NIH 2624 [35]; a 2015 account of aspulvinone O from a Chinese marine red alga ephiphyte Paecilomyces variotii EN-291 [36] (curiously the trivial nomenclature "aspulvinones J-N" do not appear to have been utilised); a 2016 account of aspulvinone P-Q from a Chinese wetland mud-derived Aspergillus flavipes PJ03-11 [37]; a 2017 account of aspulvinone R from a Chinese medicinal plant-derived Aspergillus sp. CPCC 400735 [38]; a 2021 account of aspulvinones R-U from a Thai marine sponge-derived Aspergillus flavipes KUFA1152 [39] (note-the trivial nomenclature "aspulvinone R" had been prior assigned to a different fungal metabolite in 2017 [38]); a 2021 account of aspulvinones V-X from a marine-derived Aspergillus terreus ASM-1 [31]; a 2022 account of aspulvins A-H from the Chinese medicinal plant-derived Cladosporium sp. 7951 [40]; and a 2022 account of aspulvinones S-V from the Taiwanese marine green alga-derived Aspergillus terreus NTU243 [41] (note-the trivial nomenclature "aspulvinones S-V" had been prior assigned to different fungal metabolites in 2021 [31,39]). Given this history, we attributed the trivial nomenclature aspulvinone Y to 14.
Aspulvinone D (17) may be considered as the biosynthetic precursor of other aspulvinone analogs via a putative epoxide intermediate, which can be further modified (either enzymatically or non-enzymatically) [20] by one of three pathways: (i) intramolecular S N 2 addition of the C-4 phenol moiety to C-8 with concomitant opening of the epoxide to yield aspulvinone F, (ii) intramolecular S N 2 addition of the C-4 phenol moiety to C-9 with concomitant opening of the epoxide to yield aspulvinone N-CR (15), or (iii) S N 2 addition of H 2 O to C-9 (or C-8 ) with concomitant opening of the epoxide to yield aspulvinone L-CR ( Figure 12). This same set of transformations could in principle be applied to other (as yet undetected) aspulvinone/aspulvin epoxides as a pathway to all known heterocyclic members of this structure class. Molecules 2022, 27, x FOR PEER REVIEW 13 of 18 Figure 12. A plausible biogenetic relationship linking prenylated to hydroxylated and heterocyclic aspulvinones (i-iii three different pathways to yield various aspulvinone analogs). Highlighted (bold, yellow, blue, red, green) indicate biogenetically related residues.

General Experimental Procedure
Chiroptical measurements ([α]D) were obtained on a JASCO P-1010 polarimeter (JASCO International Co. Ltd., Tokyo, Japan) in a 100 × 2 mm cell at specified temperatures. Nuclear magnetic resonance (NMR) spectra were acquired on a Bruker Avance 600 MHz spectrometer (Bruker Pty. Ltd., Preston, VIC, Australia) with either a 5 mm PASEL 1 H/D-13 C Z-Gradient probe or 5 mm CPTCI 1 H/ 19 F-13 C/ 15 N/DZ-Gradient cryoprobe. In all cases, spectra were acquired at 25 °C in DMSO-d6 or CDCl3 or methanol-d4 or acetone-d6 with referencing to residual 1 H or 13 C signals (DMSO-d6, δH 2.50 and δC 39.5; CDCl3 δH 7.26 and δC 77.1; methanol-d4, δH 3.31 and δC 49.0; acetone-d6, δH 2.05). High-resolution ESIMS spectra were obtained on a Bruker micrOTOF mass spectrometer (Bruker Pty. Ltd., Preston, VIC, Australia) by direct injection in MeOH at 3 μL/min using sodium formate clusters as an internal calibrant. Liquid chromatography diode array mass spectrometry In summary, an integrated workflow of chemical (GNPS) and cultivation (MATRIX) profiling has proved effective at exploring the chemical diversity encoded within the genome of the Queensland sheep pasture soil-derived fungus Aspergillus sp. CMB-MRF324. In addition to meroterpene millmeranones A-F (1-6), this latest study has demonstrated that CMB-MRF324 also produces new tryptamine enamino tripeptides, aspergillamide E-F (7)(8), new dihydroquinoline-2-one aflaquinolones H-I (11)(12), and a new prenylated phenylbutyrolactone aspulvinone Y (14), along with an array of known metabolites. The majority of the new natural products documented in this study proved chemically unstable once purified, further demonstrating the importance of employing effective and sensitive analytical methods to monitor and ensure chemical integrity during purification, characterisation, handling, and storage.

General Experimental Procedure
Chiroptical measurements ([α] D ) were obtained on a JASCO P-1010 polarimeter (JASCO International Co. Ltd., Tokyo, Japan) in a 100 × 2 mm cell at specified temperatures. Nuclear magnetic resonance (NMR) spectra were acquired on a Bruker Avance 600 MHz spectrometer (Bruker Pty. Ltd., Preston, VIC, Australia) with either a 5 mm PASEL 1 H/D-13 C Z-Gradient probe or 5 mm CPTCI 1 H/ 19 F-13 C/ 15 N/DZ-Gradient cryoprobe. In all cases, spectra were acquired at 25 • C in DMSO-d 6 or CDCl 3 or methanol-d 4 or acetone-d 6 with referencing to residual 1 H or 13  . MS/MS analysis was performed on the same instrument for ions detected in the full scan at an intensity above 1000 counts at 10 scans/s, with an isolation width of 4 m/z using a fixed collision energy and a maximum of 3 selected precursors per cycle. Chemicals were purchased from Sigma-Aldrich or Merck unless otherwise specified. Analytical grade solvents were used for solvent extractions. Chromatography solvents were of HPLC grade supplied by Labscan or Sigma-Aldrich and filtered/degassed through 0.45 µm polytetrafluoroethylene (PTFE) membrane prior to use. Deuterated solvents were purchased from Cambridge Isotopes. Microorganisms were manipulated under sterile conditions using a Laftech class II biological safety cabinet and incubated in either MMM Friocell incubators (Lomb Scientific Pty. Ltd., Taren Point, NSW, Australia) or an Innova 42R incubator shaker (John Morris Scientific Pty. Ltd., Chatswood, NSW, Australia).

Collection and Taxonomy of CMB-MRF324
The fungus CMB-MRF324 was isolated from the soil sample collected from a sheep pasture near Millmerran, Queensland, Australia. Following 7 days cultivation on an SD agar plate at 30 • C., Genomic DNA was extracted from the mycelia using the DNeasy Plant Mini Kit (Qiagen, Clayton, VIC, Australia), the 18S rRNA genes were amplified, and BLAST analysis (NCBI database) confirmed the closest homology with Aspergillus terreus (Genbank accession number: MZ823609) as previously reported [8].

Isomerisation of Aspergillamide F (8) to E (7)
Three samples A-C of 8 (10 µg in 100 µL MeOH each) were prepared. Sample A was stirred at room temperature for 24 h. Sample B was treated with 0.1% TFA at room temperature for 24 h. Sample C was heated at 60 • C for 24 h. After 24 h aliquots (5 µL) of each were analysed by UPLC-DAD (Zorbax C 8 RRHD 1.8 µm, 50 × 2.1 mm column, gradient elution over 2.50 min at 0.417 mL/min from 90% H 2 O/MeCN to 100% MeCN with an isocratic 0.01% TFA modifier) to reveal that 8 transformed to 7 slowly at room temperature even in the presence of acid, while heating at 60 • C accelerated this conversion ( Figure S3).

X-ray Crystallography
Crystals of 11 were obtained by slow evaporation from 50% MeOH/CH 2 Cl 2 at room temperature. Crystallographic data (CuKα radiation 1.54184 Å, 2θmax = 125 • ) were collected on an Oxford Diffraction Gemini S Ultra CCD diffractometer with the crystal cooled to 190 K with an Oxford Cryosystems Desktop Cooler. Data reduction and empirical absorption corrections were carried out with the CrysAlisPro program (Oxford Diffraction vers. 171.38.46). The structure was solved by direct methods with SHELXT and refined with SHELXL [44]. The thermal ellipsoid diagrams were generated with Mercury [45]. All crystallographic calculations were carried out within the WinGX graphical user interface [46]. The crystal structures data in CIF format have been deposited in the Cambridge Structural Database (CCDC 2217840) (see Figure 8 and Supplementary Materials).