Nanangenines: drimane sesquiterpenoids as the dominant metabolite cohort of a novel Australian fungus, Aspergillus nanangensis

Chemical investigation of an undescribed Australian fungus, Aspergillus nanangensis, led to the identification of the nanangenines – a family of seven new and three previously reported drimane sesquiterpenoids. The structures of the nanangenines were elucidated by detailed spectroscopic analysis supported by single crystal X-ray diffraction studies. The compounds were assayed for in vitro activity against bacteria, fungi, mammalian cells and plants. Bioinformatics analysis, including comparative analysis with other acyl drimenol-producing Aspergilli, led to the identification of a putative nanangenine biosynthetic gene cluster that corresponds to the proposed biosynthetic pathway for nanangenines.


Single crystal X-ray diffraction crystal structure determinations
In each case, a suitable specimen was selected and attached to a fibre or nylon loop with Exxon Paratone N, then quenched in a cold nitrogen gas stream from an Oxford Cryosystems Cryostream. Data were collected with a four-circle kappa goniometer and CCD detector equipped SuperNova Xray diffractometer, using mirror monochromated Cu K radiation generated from a microsource. Data processing was undertaken with CrysAlisPro [1] software and included a multi-scan absorption correction. Subsequent computations were carried out with the assistance of ShelXle [1], WinGX [2,3], PLATON [4,5] and XNPP [6]. The structures were obtained using ShelXT [7] and extended and refined with ShelXL [8]. In general non-hydrogen atoms in the asymmetric unit were modelled with anisotropic displacement parameters and a riding atom model was used for the hydrogen atoms. Images with displacement ellipsoid representations were generated with Olex2 [9]. Crystallographic data are summarised in Table S2.
General procedure for synthesis of 4-bromobenzoate derivatives: A solution of nanangenine to be derivatised (0.03 mmol), 4-bromobenzoyl chloride (6.6 mg, 0.03 mmol) and triethylamine (50 L) in MeCN (950 L) was heated at 70 °C for 4 h. The solution was washed with sodium bicarbonate solution (3 M, 50 mL) and extracted into ethyl acetate before being concentrated in vacuo. The residue was purified by preparative HPLC (Hypersil C18, isocratic 100% MeCN, 20 mL min −1 ). Crystals suitable for X-ray diffraction were grown by slow evaporation of methanolic solutions.

Nanangenine A (1):
A colourless blade-like crystal of 1 was used for the data collection and the structure was obtained and modelled in the space group P21(#4). The asymmetric unit was found to contain four crystallographically-independent molecules. The hydroxy hydrogens were located in final difference maps and in general modelled with isotropic displacement parameters. The displacement parameter for one of the model hydroxy hydrogen atoms (H3O_4) was tied to that of the associated oxygen atom. In addition to an intramolecular hydrogen bond, intermolecular hydrogen bonds link molecules in discrete layers on the (002) planes of the unit cell. A depiction of one the molecules with 50% displacement ellipsoids is provided in Figure S3. The Parsons intensity quotients based Flack parameter determined by ShelXL was 0.06 (3). The PLATON calculated Bayesian probability estimate that the assigned absolute structure is correct is 1.0. The probability estimate that the assignment is incorrect is 0.0, while that for an inversion twin is 8.0 × 10 -51 . Accordingly, the assigned absolute structure is confirmed. CCDC reference: 1940618.

9-O-(4-Bromobenzoyl)-nanangenine B (2b):
A colourless columnar crystal was used for the data collection and the structure was obtained in the space group P21 (#4). The asymmetric unit contains two crystallographically-independent molecules. The diffraction data are pseudo-orthorhombic and a structure can be obtained and modelled in P212121 (#19), however the pendant alkyl residue is poorly resolved. Relatively large displacement envelopes on the alkyl residue mask disorder. The hydroxy hydrogen sites were located in final difference maps and modelled with isotropic displacement parameters. An Olex2 depiction of one of the molecules, with 50% displacement ellipsoids, is provided in Figure S4. The Parsons intensity quotients based Flack parameter determined by ShelXL was −0.003 (7) and accordingly the assigned absolute structure is confirmed. CCDC reference: 1940620.

1-O-(4-Bromobenzoyl)-isonanangenine B (3b):
A pale orange prismatic fragment cut from a larger crystal was used for the data collection and the structure was obtained in the space group P21(#4). The asymmetric unit contains two crystallographically independent molecules. The asymmetric unit contains two crystallographically-independent molecules, one of which exhibits minor site disorder of S5 the bromobenzoate residue. Two orientations were modelled for the disordered residue, with respective occupancies refined and then fixed at 0.93 and 0.07. A rigid body was applied in modelling the minor occupancy bromobenzoate sites. Isotropic displacement parameters were used for the minor occupancy sites. The hydroxy hydrogen position was located in final difference maps and refined with isotropic displacement parameters. The two independent molecules are linked by a hydrogen bond between the hydroxy residue of one and the furanone carbonyl moiety of the second. An Olex2 depiction of one of the molecules, with 75% displacement ellipsoids, is provided in Figure S5. The Parsons intensity quotients based Flack parameter determined by ShelXL refined to -0.017 (5). The PLATON calculated Bayesian probability estimate that the assigned absolute structure is correct is 1.0. The probability estimate that the assignment is incorrect or is an inversion twin is 0.0. CCDC reference: 1941156.

Nanangenine C (4):
A colourless prismatic crystal of 4 was used for the data collection and the structure was obtained in the space group P212121 (#19). The asymmetric unit contains 4 linked by a hydrogen bond to a water molecule; the water hydrogen positions were inferred from final difference maps. A depiction of the molecule with 50% displacement ellipsoids is provided in Figure S6. The Parsons intensity quotients based Flack parameter determined by ShelXL was 0.10 (9) and that of XNPP refined to 0.12 (7). The PLATON calculated Bayesian probability estimate that the assigned absolute structure is correct is 1.0. The probability estimate that the assignment is incorrect is 1.0 × 10 -37 , while that for an inversion twin is 1.0 × 10 -6 . Accordingly, though not definitively determined, the assigned absolute structure appears likely to be correct. CCDC reference: 1940619.

Nanangenine G (9):
A colourless plate like crystal of 9 was used for the data collection and the structure was obtained in the space group P212121 (#19). The pedant alkyl residue of 9 is disordered about two orientations. Site occupancies were refined and then fixed at 0.6 and 0.4; isotropic displacement parameters were used for the minor occupancy sites. The hydroxy hydrogen positions for all but that of O3 were located in final difference maps and refined with isotropic displacement parameters. Adjacent molecules are linked by hydrogen bonds. A depiction of the molecule with 50% displacement ellipsoids is provided in Figure S7. The ShelXL and XNPP calculated Flack parameter refined to 0.10(10) and 0.08 (9) respectively. The PLATON calculated Bayesian probability estimate that the assigned absolute structure is correct is 1.0. The probability estimate that the assignment is incorrect is 4.0 × 10 -24 , while that for an inversion twin is 4.0 × 10 -5 . Accordingly, though not definitively determined, the assigned absolute structure is likely correct. CCDC reference: 1940621.

Bioassay of compounds
Purified metabolites were dissolved in DMSO to provide stock solutions of 10,000 g/mL. An aliquot of each stock solution was transferred to the first lane of Rows B to G in a 96-well microtitre plate and two-fold serially diluted with DMSO across the 12 lanes of the plate to provide a 2,048-fold concentration gradient. Bioassay medium was added to an aliquot of each test solution to provide a 100-fold dilution into the final bioassay, thus yielding a test range of 100 to 0.05 g/mL in 1% DMSO. Row A contained no test compound (as a reference for no inhibition) and row H was uninoculated (as a reference for complete inhibition).
ProTOX is a generic bioassay platform for antibiotic discovery. In the present study Bacillus subtilis (ATCC 6633) and Escherichia coli (ATCC 25922) were used as indicative species for Gram-positive and Gram-negative antibacterial activity. A bacterial suspension (50 mL in 250 mL flask) was prepared in nutrient media by cultivation for 24 h at 250 rpm, 28 °C. The suspension was diluted to an absorbance of 0.01 absorbance units per mL, and 10 L aliquots were added to the wells of a 96-well microtitre plate, which contained the test compounds dispersed in nutrient broth (Amyl) with resazurin (12.5 g mL -1 ). The plates were incubated at 28 °C for 24 h during which time the positive control wells change colour from a blue to light pink colour. MIC end points were determined visually. The absorbance was measured using Spectromax plate reader (Molecular Devices) at 605 nm and the IC50 values determined graphically.
EuTOX is a generic bioassay platform for antifungal discovery. In the present study, the yeast Candida albicans (ATCC 10231) was used as indicative species for antifungal activity. A yeast suspension (50 mL in 250 mL flask) was prepared in 1% malt extract broth by cultivation for 24 h at 250 rpm, 28 °C. The suspension was diluted to an absorbance of 0.005 absorbance units per mL for C. albicans. Aliquots (20 L and 30 L) of C. albicans were applied to the wells of a 96-well microtitre plate, which contained the test compounds dispersed in malt extract agar containing bromocresol green (50 g mL -1 ). The plates were incubated at 28 °C for 24 h during which time the positive control wells change colour from a blue to yellow colour. MIC end points were determined visually. The absorbance was measured using Spectromax plate reader (Molecular Devices) at 620 nm and the IC50 determined graphically.
Phytox is a generic bioassay platform for herbicidal discovery. In the present study, Eragrostis tef (teff) seed was used as indicative species for herbicidal discovery. Teff seeds (10 to 15) were dispensed using a LabTIE seed dispenser into the wells of a 96-well microtitre plate, which contained the test compounds dispersed in 200 L of agar (1% w/v) per well. The plates were placed in a tray wrapped with a semi-opaque bag, exposed to 1600 lux (within the tray) using Power-GLO (20 W) and Sun-GLO (20 W) tubes, and incubated for 72 h at 24 °C.

Genomic DNA extraction
A. nanangensis was grown in potato dextrose broth (BD Bioscience) at room temperature. Mycelium was collected from one week-old liquid culture and ground in liquid nitrogen. The frozen ground mycelium was extracted using a buffer with 10mM Tris, 10 mM EDTA, 1% SDS and incubated at 70 °C for 20 min. This is followed by precipitation of the SDS and cell debris with equal volume of 2.8 M potassium acetate and clean-up with chloroform/isoamyl alcohol (24:1) extraction. Finally, the genomic DNA was precipitated from the supernatant using 0.7 volume isopropanol and the DNA pellet was dissolved in TE buffer. The quality of the genomic DNA was checked by gel electrophoresis.

Bioinformatics analyses
Alignment of A. nanangensis MST-FP2251 ITS to an undescribed sequence: A. nanangensis ITS sequence was retrieved from genome sequence and is deposited on GenBank (MK979278). ITS sequences of A. janus and an uncultured fungus were obtained from NCBI at accessions EU021598.1 and GQ921753.1, respectively. Sequences were trimmed based on length of the shortest sequence, and the multiple sequence alignment was constructed using Clustal Omega v1.2.4 [10].

Alignment of A. nanangensis MST-FP2251 with other
Aspergilli: ITS sequences of Aspergillus species were obtained from NCBI (Table S13) and were aligned using MAFFT with the L-INS-i option. ML tree was constructed using RAxML-NG v0.6.0 [11] from 1000 bootstrap replicates. BI tree was constructed using MrBayes v3.2.6 [12] from 200000 generations. Both trees were constructed under the General Time Reversible (GTR) substitution model.

Alignment of AstC and homologs listed in Shinohara et al:
Though Shinohara et al. list the revised annotation of AstC as AORIB40_05408, its amino acid sequence of was retrieved from AspGD [13] under the locus tag AO090026000582. AORIB40_NS.05916 was obtained from AspGD at coordinates Chr3_A_oryzae_RIB40:3512882-3513472(-). All other sequences were retrieved from NCBI via their accession codes. A multiple sequence alignment was constructed using ClustalOmega v1.2.4 [10].
Alignment of AstC to tBLASTn hits: Local BLAST databases were constructed inside Geneious 10.2.6 and AstC homologs were identified using tBLASTn. A multiple sequence alignment of the AstC homologs was constructed using MUSCLE v3.8.31 [14] and a phylogenetic tree was constructed under the JTT model using FastTree 2.1.5 [15].
Custom Python scripts: Homologous gene clusters were identified using a custom Python script, named clusterblaster (https://github.com/gamcil/clusterblaster). Briefly, the script takes as input a set of query sequences in FASTA format (presumably clustered), searches a pre-constructed DIAMOND [16] database consisting of protein sequences obtained from NCBI, then leverages the 'identical protein group' functionality of NCBI Entrez to ascertain the original genomic context of hit proteins (i.e. start and end on an assembly scaffold). Hits are then grouped by scaffold and ordered by location; those that meet user-defined clustering thresholds (maximum intergenic distance, minimum number of proteins conserved) are reported.
Gene cluster homology was visualised using crosslinker (https://github.com/gamcil/crosslinker). Briefly, crosslinker takes as input GenBank files of gene clusters, performs pairwise alignments between all sets of protein sequences, then draws a to-scale SVG figure depicting gene cluster homology with shaded boxes linking homologous proteins, the darkness of which indicates the degree of amino acid identity.

Analysis Info
Acquisition Date 4/11/20163:51:14 PM D:\ Data\nick-2016    S61 Figure S49. Taxonomic distribution of drimane sesquiterpene production across the Aspergillus genus. Underlying phylogenetic tree is built from a concatenated multi-locus alignment consisting of internal transcribed spacer (ITS), beta-tubulin and calmodulin sequences.