Data acquired on the TFA salts. a,b Resonances with the same superscript overlap and assignments can be interchanged.

**Figure 5.** 2D NMR (methanol-*d*4) correlations for **5**–**10**.

**Figure 6.** X-ray crystal structure of noonindole A (**5**).


nd: Not detected. <sup>a</sup> Resonances with the same superscript within a column are overlapping. \* Obscured by residual solvent resonance. # Data for **8**–**9** acquired on the free bases (not TFA salts).

HRESIMS analysis of **9** revealed a molecular formula (C34H44N2O6, Δmmu +1.2) consistent with an oxidised analogue of **5**. Comparison of the NMR (methanol-*d*4) data for **9** (Tables 2 and S6, and Figures S37–S41) with **5** allowed the principle differences to be attributed to replacement of the sp<sup>3</sup> H-7/C- 7 methine in **5** (δ<sup>H</sup> 4.91, H-7; δ<sup>C</sup> 74.6, C-7) with an sp<sup>2</sup> quaternary C-7 and sp2 H-6/C-6 methine in **9** (δ<sup>H</sup> 5.74, m, H-6; δ<sup>C</sup> 113.5, C-6; δ<sup>C</sup>

146.4, C-7), consistent with incorporation of a Δ6,7. These conclusions were reinforced by 2D NMR correlations (Figure 5) which, together with biogenetic considerations, allowed assignment of the structure for noonindole E (**9**).

HRESIMS analysis of **10** revealed a molecular formula (C27H33NO5, Δmmu +1.1) consistent with a hydrolysed analogue of **5** lacking the *N*,*N*-dimethyl-valinyloxy ester side chain. This hypothesis was confirmed on comparison of the NMR (methanol-*d*4) data for **10** (Tables 2 and S7, and Figures S43–S48) with **5** which revealed the absence of resonances for the *N*,*N*-dimethyl-valinyloxy moiety and a significant shielding of the resonance for H-14 in **10** (δ<sup>H</sup> 4.16) compared with **5** (δ<sup>H</sup> 5.53). These conclusions were reinforced by 2D NMR correlations (Figure 5) which, together with biogenetic considerations, allowed assignment of a structure for noonindole F (**10**).

Co-isolation of noonindoles A–F (**5**–**10**) and the known IDTs **11**–**14** together with an X-ray crystal analysis of **5** supported a common absolute configuration across the hexacyclic indole terpene core, and established the configuration of the *N*,*N*-dimethyl-L-valinyloxy moiety in **5**. Although low yields combined with *N*-alkylation precluded hydrolysis and independent assignment of the amino acid residue absolute configuration in **6**–**9** (i.e., Marfey's analysis), an amino acid L configuration across **6**–**9** was proposed based on the likelihood of a common NRPS-*like* aminoacyl modifying enzyme in the noonindole biosynthetic gene cluster (BGC) (see below).

The metabolites **5**–**14** did not inhibit the growth of human colon (SW620) or lung (NCI-H460) carcinoma cells (IC50 > 30 μM) (Figure S81), or the fungus *Candida albicans* ATCC10231, the Gram-negative bacterium *Escherichia coli* ATCC11775, or the Gram-positive bacteria *Staphylococcus aureus* ATCC25923 or *Bacillus subtilis* ATCC6633 (IC50 > 30 μM), with the exception of noonindole A (**5**) which displayed modest antifungal activity (IC50~5 μM) (Figure S80). This lack of cellular toxicity bodes well for future (ongoing) evaluation of noonindole ion channel inhibitory pharmacology.

GNPS analysis of the EtOAc extract of the analytical scale D400 (MATRIX) culture of CMB-M0339 detected **5**–**9** along with associated nodes for a selection of putative minor analogues (Figure 7). The MS/MS spectra for **5**–**9** (Figures S72–S75) revealed three common fragmentations attributed to loss of water (Figure 8A), retro-Aldol loss of acetone (Figure 8B), and loss of the amino acid residue (Figure 8C). While low yields precluded isolation of the minor analogues **i**–**viii**, diagnostic MS/MS fragmentations and highresolution mass measurements (i.e., molecular formulae) permitted tentative assignments for **i**–**v** (Figures S74, S76–S79, Table S12) and on the basis of GNPS co-clustering and biosynthetic considerations to **vi**–**viii**, albeit with some allowance isomeric alternatives (Tables 3 and S12). The diversity of IDT amino acid conjugates produced by CMB-M0339 is in stark contrast to existing knowledge, which is limited to **3** from *Aspergillus nomius* [9] and **4** from *A. alliaceus* [10]. Unlike these earlier published accounts, it appears CMB-M0339 employs an NRPS-*like* aminoacylation enzyme with an adenylation domain tolerant of different amino acid substrates (i.e., Val, Leu, Ile, Pro, Ser, Thr, and homo-Ala).

A preliminary assessment of the noonindole biosynthetic gene cluster (BGC) suggests a biogenetic relationship linking **5**–**14** and inclusive of the minor co-metabolites **i**–**viii** starting with emindole SB (**14**) undergoing stereospecific epoxidation and ring closure to paspaline (**11**) followed by sequential oxidation to paspaline B (**12**) and the carboxylic acid **13**, followed by decarboxylation and oxidation to paxilline (**1**) (Figure 9). Oxidation of **1** could then yield noonindole F (**10**) with further oxidation and/or amino acid acylation returning noonindoles A–E (**5**–**9**) and co-clustering minor analogues (**i**–**viii**). Consistent with this hypothesis, close examination of the CMB-M0339 D400 extract GNPS and UPLC-DAD-MS data using single ion extraction (SIE) detected an ion with a molecular formula attributable to **1** (Figure S71).

**Figure 7.** GNPS cluster of the EtOAc extract of a D400 (MATRIX) culture of CMB-M0339 revealing **5**–**9** along with closely associated nodes for the minor analogues **i**–**viii**.

**Figure 8.** MS/MS fragmentations common to **1**–**5** and minor co-metabolites **i**–**viii**; (**A**) loss of water, (**B**) retro-Aldol loss of acetone and, (**C**) loss of the amino acid residue.



(a,b,c) Proposed hexacyclic scaffolds. Possible alternate isomers: <sup>A</sup> *N*-methyl-homoalanine, valine; <sup>B</sup> *N*-methylthreonine; <sup>C</sup> pipecolic acid; <sup>D</sup> *N*,*N*-dimethyl-*allo*-isoleucine.

**Figure 9.** Plausible biogenetic relationship linking **5**–**14** (and **1**) and inclusive of the minor cometabolites **i**–**viii**.

Our investigation into the marine-derived *Aspergillus noonimiae* CMB-M0339 led to the discovery of noonindoles A–F (**5**–**10**) and related minor analogues (**i**–**viii**) as new examples of a rare class of fungal indole diterpene amino acid conjugate. This discovery highlights the continued capacity of fungi to provide access to new chemical space and validates molecular networking (GNPS) as an effective platform to detect, dereplicate, and prioritize new over known chemistry, and cultivation profiling (MATRIX) as a means to optimise the production. Our discovery of the noonindoles suggests the CMB-M0339 features an NRPS-*like* aminoacyl modifying enzyme in the noonindole biosynthetic gene cluster (BGC) capable of accommodating and incorporating multiple lipophilic amino acids. Further studies into the structure, biosynthesis, and biology of these and other CMB-M0339 indole diterpenes are ongoing, and will be reported elsewhere.

#### **3. Materials and Methods**

#### *3.1. General Experimental Procedures*

Chemicals were purchased from Sigma-Aldrich or Merck unless otherwise specified. Solvent extractions were performed using analytical-grade solvents, while HPLC, UPLC, and HPLC-MS analyses employed HPLC-grade solvents 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 (Tewksbury, MA, USA). Microorganisms were manipulated under sterile conditions in a Laftech class II biological safety cabinet and incubated in either an MMM Friocell incubator (Lomb Scientific, NSW, Australia) or an Innova 42R incubator shaker (John Morris, NSW, Australia) at 26.5 ◦C. Semi-preparative and preparative HPLCs were performed using Agilent 1100 series HPLC instruments with corresponding detectors, fraction collectors, and software. Analytical UPLC chromatograms were obtained on an Agilent 1290 infinity UPLC instrument equipped with a diode array multiple wavelength detector (Zorbax C8 RRHD 1.8 μm, 50 × 2.1 mm column, gradient elution at 0.417 mL/min over 2.50 min from 90% H2O/MeCN to 100% MeCN with isocratic 0.01% TFA/MeCN modifier). UPLC-QTOF analyses were performed on an Agilent 6545 Q-TOF instrument incorporating an Agilent 1290 Infinity II UHPLC (Zorbax C8 RRHD 1.8 μm, 50 × 2.1 mm column, gradient elution

at 0.417 mL/min over 2.5 min from 90% H2O/MeCN to 100% MeCN with isocratic 0.1% formic acid/MeCN modifier). Chiroptical measurements ([α]D) were obtained on a JASCO P-1010 polarimeter in a 100 × 2 mm cell at specified temperatures. Nuclear magnetic resonance (NMR) spectra were acquired on a Bruker Avance 600 MHz spectrometer with either a 5 mm PASEL 1H/D-13C Z-Gradient probe or 5 mm CPTCI 1H/19F-13C/15N/DZ-Gradient cryoprobe, controlled by TopSpin 2.1 software, at 25 ◦C in either methanol-*d*4, CDCl3, or DMSO-*d*6, with referencing to residual 1H or 13C solvent resonances (methanol*d*4: δ<sup>H</sup> 3.31 and δ<sup>C</sup> 49.15; CDCl3: δ<sup>H</sup> 7.24 and δ<sup>C</sup> 77.23; DMSO-*d*6: δ<sup>H</sup> 2.50 and δ<sup>C</sup> 39.50). High-resolution ESIMS spectra were obtained on a Bruker micrOTOF mass spectrometer by direct injection in MeOH at 3 μL/min using sodium formate clusters as an internal calibrant. Structural assignments were made with additional information from gCOSY, gHSQC, and gHMBC experiments.

#### *3.2. Fungal Isolation and DNA Taxonomic Analysis*

A marine sediment collected in 2008 from a location off Perth, Western Australia, was used to inoculate an M1 agar plate (inclusive of 3.3% artificial sea salt) which was incubated at 27 ◦C for 10–14 days, after which colony selection yielded an array of isolates including fungus CMB-M0339. Genomic DNA was extracted from the mycelia of CMB-M0339 using the DNeasy Plant Mini Kit (Qiagen) as per the manufacturers protocol, and the 18s rRNA genes were amplified by PCR using the universal primers ITS-1 (5 -TCCGTAGGTGAACCTGCGG-3 ) and ITS-4 (5 -TCCTCCGCTTATTGATATGC-3 ) purchased from Sigma-Aldrich. The PCR mixture (50 μL) containing 1 μL of genomic DNA (20–40 ng), 200 μM of each deoxynucleoside triphosphate (dNTP), 1.5 mM MgCl2, 0.3 μM of each primer, 1 U of *Taq* DNA polymerase (Fisher Biotec), and 5 μL of PCR buffer was amplified using the following conditions: initial denaturation at 95 ◦C for 3 min, 30 cycles in series of 94 ◦C for 30 s (denaturation), 55 ◦C for 60 s (annealing), and 72 ◦C for 60 s (extension), followed by one cycle at 72 ◦C for 6 min. PCR products were purified with PCR purification kit (Qiagen, Victoria, Australia) and examined by agarose gel electrophoresis, with DNA sequencing performed by the Australian Genome Research Facility (AGRF) at The University of Queensland. A BLAST analysis (NCBI database) on the resulting CMB-M0339 ITS gene sequence (Figures S1–S3, GenBank accession no. OP132523) revealed 92.5% identity with the fungal strain *Aspergillus noonimiae.*

#### *3.3. Global Natural Product Social (GNPS) Molecular Networking*

Aliquots (1 μL) of CMB-M0339 cultivation extract (100 μg/mL in MeOH) were analysed on an Agilent 6545 Q-TOF LC/MS equipped with an Agilent 1290 Infinity II UPLC system (Zorbax C8, 0.21 μm, 1.8 × 50 mm column, gradient elution at 0.417 mL/min over 2.5 min from 90% H2O/MeCN to MeCN with an isocratic 0.1% formic acid/MeCN modifier). UPLC-QTOF-(+) MS/MS data acquired for all samples at a collision energy of 35 eV were converted from Agilent MassHunter data files (d) to mzXML file format using MSConvert software, and transferred to the GNPS server (gnps.ucsd.edu). Molecular networking was performed using the GNPS data analysis workflow [45] employing the spectral clustering algorithm with a cosine score of 0.5 and a minimum of 6 matched peaks. The resulting spectral network was imported into Cytoscape version 3.7.1 [47] and visualized using a ball-and-stick layout where nodes represent parent mass and cosine score was reflected by edge thickness. Moreover, group abundances were set as pie charts, which reflected the intensity of MS signals. MS/MS fragmentation analysis was performed on the same machine for ion detected in the full scan range at an intensity above 1000 counts at ten scans/s, with an isolation width of 4~*m*/*z* using fixed collision energy and a maximum of 3 selected precursors per cycle. General instrument parameters including gas temperature at 325 ◦C, drying gas 10 L/min, nebulizer 20 psig, sheath gas temperature 400 ◦C, fragmentation Volta 180eV, and skimmer 45 eV.

#### *3.4. MATRIX Cultivation Profiling*

The fungus CMB-M0339 was cultured in a 24-well plate microbioreactor under ×11 different media for 10–14 days in solid phase (27 ◦C), as well as in static (30 ◦C) and shaken broths (30 ◦C, 190 rpm) [46], with regular monitoring of growth (Figure S7). At this point, wells were individually extracted with EtOAc (2 mL), and the organic phase was centrifuged (13,000 rpm, 3 min) and dried under N2 at 40 ◦C to yield ×33 extracts. Individual extracts were redissolved in MeOH (30 μL) containing calibrant (2,4-dinitrophenoldecane ether, 50 μg/mL), and aliquots (1 μL) were subjected to: (i) UPLC-DAD analysis (Zorbax C8 1.8 μm, 2.1 × 50 mm column, gradient elution at 0.417 mL/min over 2.52 min from 90% H2O/MeCN to 100% MeCN followed by 0.83 min isocratic elution with MeCN, inclusive of an isocratic 0.01% TFA/MeCN modifier) (Figure S8); and (ii) GNPS analysis (Figure 3). This process identified solid phase D400 as the optimal culture conditions for producing targeted CMB-M0339 natural products.

#### *3.5. Scale Up Cultivation and Fractionation*

The fungus CMB-M0339 was cultivated on D400 agar (×300 plates) at 27 ◦C for 10–14 days after which the agar and fungal mycelia were harvested and extracted with EtOAc (2 × 5 L), and the combined organic phase was filtered and concentrated in vacuo at 40 ◦C to yield an extract (2.9 g). This extract was sequentially triturated with *n*-hexane (20 mL), CH2Cl2 (20 mL), MeOH (20 mL), and concentrated in vacuo to afford *n*-hexane (961.8 mg), CH2Cl2 (1783.8 mg), and MeOH (80.3 mg) soluble fractions. A portion of the CH2Cl2 soluble fraction (1363 mg) was subjected to preparative reversed-phase HPLC (Phenomenex Luna-C8 10 μm, 21.2 × 250 mm column, with gradient elution at 20 mL/min over 20 min from 90% H2O/MeCN to 100% MeCN with constant 0.1% TFA/MeCN modifier) to yield noonindole A (**5**) (Rf 15.7 min, 83.5 mg, 4.3%). The remaining mixed fractions were subjected to semi-preparative reversed-phase HPLC to yield noonindole B (**6**) (Rf 18.3 min, 2.0 mg, 0.1%) (semi-preparative HPLC (Zorbax C8 5mm column, 9.4 × 250 mm, 3 mL/min isocratic elution of 37% MeCN/H2O over 20 min with constant 0.1% TFA modifier)); noonindole F (**10**) (Rf 19.9 min, 0.7 mg, 0.03%) (semi-preparative HPLC (Agilent C8-Ep 5mm column, 9.4 × 250 mm, 3 mL/min isocratic elution of 50% MeCN/H2O over 25 min with constant 0.1% TFA modifier)); noonindole D (**8**) (Rf 28.6 min, 0.4 mg, 0.02%) (semi-preparative HPLC (Zorbax C18 5mm column, 9.4 × 250 mm, 3 mL/min isocratic elution of 40% MeCN/H2O over 30 min with constant 0.1% TFA modifier)); noonindole E (**9**) (Rf 20.9 min, 0.8 mg, 0.04%) (semi-preparative HPLC (Zorbax C18 5mm column, 9.4 × 250 mm, 3 mL/min isocratic elution of 40% MeCN/H2O over 30 min with constant 0.1% TFA modifier)); 12-demethylpaspaline-12-carboxylic acid (**13**) (Rf 9.8 min, 1.6 mg, 0.08%) (semi-preparative HPLC (Zorbax C18 5mm column, 9.4 × 250 mm, 3 mL/min isocratic elution of 85% MeCN/H2O over 15 min with constant 0.1% TFA modifier)); paspaline B (**12**) (Rf 26.1 min, 0.3 mg, 0.01%) (semi-preparative HPLC (Agilent C8-Ep 5mm column, 9.4 × 250 mm, 3 mL/min isocratic elution of 60% MeCN/H2O over 25 min with constant 0.1% TFA modifier)); paspaline (**11**) (Rf 17.8 min, 1.2 mg, 0.06%) (Semi-preparative HPLC (Agilent CN 5mm column, 9.4 × 250 mm, 3 mL/min isocratic elution of 60% MeCN/H2O over 20 min with constant 0.1% TFA modifier)); emindole SB (**14**) (Rf 19.7 min, 1.2 mg, 0.06%) (semi-preparative HPLC (Agilent CN 5mm column, 9.4 × 250 mm, 3 mL/min isocratic elution of 60% MeCN/H2O over 20 min with constant 0.1% TFA modifier)); and solidphase extraction (Sep-Pak (Agilent Bond Elut C18 cartridge, 5 g) gradient elution from 90% H2O/MeCN to 100% MeCN) and semi-preparative reversed-phase HPLC (Zorbax C8 5mm column, 9.4 × 250 mm, 3 mL/min isocratic elution of 40% MeCN/H2O over 20 min with constant 0.1% TFA modifier) to yield noonindole C (**7**) (Rf 21.9 min, 2.0 mg, 0.1%) (Figure S9). (Note: All % yields are weight to weight estimates based on unfractionated EtOAc extract).

#### *3.6. Characterization of Metabolites* **5**–**14**

*noonindole A* (**5**); pale yellow solid; [α]D21–18 (*c* 0.02, MeOH); NMR (600 MHz, methanol*d*4), see Table S2 and Figures S10–S15; HRMS (ESI) *m*/*z*: [M+H]<sup>+</sup> calcd for C34H47N2O6 579.3429; found 579.3456.

*noonindole B* (**6**); white solid; [α]D21–13 (*c* 0.08, MeOH); NMR (600 MHz, methanol-*d*4), see Table S3 and Figures S17–S22; HRMS (ESI) *m*/*z*: [M+H]<sup>+</sup> calcd for C33H45N2O6, 565.3272; found 565.3292.

*noonindole C* (**7**); white solid; [α]D21–9 (*c* 0.06, MeOH); NMR (600 MHz, methanol-*d*4), see Table S4 and Figures S24–S29; HRMS (ESI) *m*/*z* [M+H]<sup>+</sup> calcd for C35H49N2O6, 593.3585; found 593.3611.

*noonindole D* (**8**); white solid; [α]D23–30 (*c* 0.03, MeOH); NMR (600 MHz, methanol-*d*4), see Table S5 and Figures S31–S35; HRMS (ESI) *m*/*z* [M+H]<sup>+</sup> calcd for C33H45N2O6, 565.3272; found 565.3287.

*noonindole E* (**9**); white solid; [α]D23–13 (*c* 0.06, MeOH); NMR (600 MHz, methanol-*d*4), see Table S6 and Figures S37–S41; HRMS (ESI) *m*/*z* [M+H]<sup>+</sup> calcd for C34H45N2O6, 577.3272; found 577.3284.

*noonindole F* (**10**); white solid; [α]D21–18 (*c* 0.02, MeOH); NMR (600 MHz, methanol-*d*4), see Table S7 and Figures S43–S48; HRMS (ESI) *m*/*z* [M+Na]+ calcd for C27H33NO5Na, 474.2251; found 474.2262.

*paspaline* (**11**); white solid; [α]D22–21 (*c* 0.09, CHCl3); [24] NMR (600 MHz, DMSO-*d*6), see Table S8 and Figures S50–S53; [24] HRMS (ESI) *m*/*z* [M+H]+ calcd for C28H40NO2, 422.3054; found 422.3071.

*paspaline B* (**12**) white solid; [α]D22–24 (*c* 0.02, CHCl3); [18] NMR (600 MHz, CDCl3), see Table S9 and Figures S55–S58; [18] HRMS (ESI) *m*/*z* [M+Na]+ calcd for C28H37NO3Na, 458.2666; found 458.2680.

12-demethylpaspaline-12-carboxylic acid (**13**); white solid; [α]D<sup>22</sup> + 37 (*c* 0.01, CHCl3); [24] NMR (600 MHz, DMSO-*d*6), see Table S10 and Figures S60–S63; [24] HRMS (ESI) *m*/*z* [M+H]+ calcd for C28H38NO4,452.2795; found 452.2807.

*emindole SB* (**14**); white solid; [α]D21–18 (c 0.05, CHCl3); [24] NMR (600 MHz, DMSO-*d*6), see Table S11 and Figures S65–S68; [24] HRMS (ESI) *m*/*z* [M+H]+ calcd for C28H40NO, 406.3104; found 406.3126.

#### *3.7. Phylogenetic Comparison of CMB-M0339 with Fungi Reported to Produce Biosynthetically Related Indole Terpenes*

Phylogenetic tree obtained by PhyML Maximum Likelihood analysis was constructed using the top similar 18S rRNA sequences displayed after BLAST on Refseq RNA NCBI database using CMB-M0339 18S rRNA as queries (Figure S4). The JC69 model was used to infer phylogeny sequences [48]. Sequence alignments were produced with the MUSCLE program [49]. Phylogenetic tree was constructed using the UGENE program using the aforementioned models and visualized using Ugene's tree view [50].

#### *3.8. UPLC-QTOF-SIE Detection of* **5**–**14** *in CMB-M0339 Extract*

The EtOAc extract of a CMB-M0339 D400 agar culture was dissolved in MeOH and subjected to UPLC-QTOF analysis with single ion extraction (SIE) analysis (Figure S70).

#### *3.9. X-ray Crystallography*

Crystals of **5** were obtained by slow evaporation from 50% DCM/Hexane in the cold room (−4 ◦C). Crystallographic data (Cu Kα, 2θmax = 125◦) for **5** 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. The structure was solved with SHELXT and refined with SHELXL [51]. The thermal ellipsoid diagrams were generated with Mercury [52]. All calculations were carried out within the WinGX graphical user interface [53]. The disordered water molecules in the structure were modelled with SQUEEZE implemented in PLATON [54]. The crystal data for **5** in CIF format were deposited in the CCDC database (2206901) (Table S13).

#### *3.10. Antifungal Assay*

The fungus *Candida albicans* ATCC 10231 was streaked onto a LB (Luria–Bertani) agar plate and was incubated at 37 ◦C for 48 h, after which a colony was transferred to fresh LB broth (15 mL) and the cell density was adjusted to 104–105 CFU/mL. Test compounds were dissolved in DMSO and diluted with H2O to prepare 600 μM stock solutions (20% DMSO), which were serially diluted with 20% DMSO to provide concentrations from 600 μM to 0.2 μM in 20% DMSO. An aliquot (10 μL) of each dilution was transferred to a 96-well microtiter plate and freshly prepared fungal broth (190 μL) was added to prepare final concentrations of 30–0.01 μM in 1% DMSO. The plates were incubated at 27 ◦C for 48 h and the optical density of each well was measured spectrophotometrically at 600 nm using POLARstar Omega plate (BMG LABTECH, Offenburg, Germany). Amphotericin B was used as the positive control (40 μg/mL in 10% DMSO). The IC50 value was calculated as the concentration of the compound or antibiotic required for 50% inhibition of the bacterial cells using Prism 7.0 (GraphPad Software Inc., La Jolla, CA, USA). See Figure S80.

#### *3.11. Antibacterial Assay*

The bacterium to be tested was streaked onto an LB agar plate and was incubated at 37 ◦C for 24 h, after which a colony was transferred to fresh LB broth (15 mL) and the cell density was adjusted to 104–105 CFU/mL. Test compounds were dissolved in DMSO and diluted with H2O to give 600 μM stock solutions (20% DMSO), which were serially diluted with 20% DMSO to prepare concentrations from 600 μM to 0.2 μM in 20% DMSO. An aliquot (10 μL) of each dilution was transferred to a 96-well microtiter plate and freshly prepared microbial broth (190 μL) was added to provide final concentrations of 30–0.01 μM in 1% DMSO. The plates were incubated at 37 ◦C for 24 h and the optical density of each well was measured spectrophotometrically at 600 nm using POLARstar Omega plate (BMG LABTECH, Offenburg, Germany). Each test compound was screened against the Gram-negative bacterium *Escherichia coli* ATCC 11775 and the Gram-positive bacteria *Staphylococcus aureus* ATCC 25923 and *Bacillus subtilis* ATCC 6633. Rifampicin was used as the positive control (40 μg/mL in 10% DMSO) for Gram-positive bacteria and a mixture of rifampicin and ampicillin was used as the positive control for Gram-negative bacteria. The IC50 value was calculated as the concentration of the compound or antibiotic required for 50% inhibition of the bacterial cells using Prism 7.0 (GraphPad Software Inc., La Jolla, CA, USA). See Figure S80.

#### *3.12. Cytotoxicity Assays*

Human colorectal (SW620) and lung carcinoma (NCI-H460) cells were seeded evenly in a 96-well micro-plate (2000 cells/well in 180 μL of RPMI 1640 medium (Roswell Park Memorial Institute medium) supplemented with 10% FBS (Fetal Bovine Serum)) and the plate was incubated for 18 h (37 ◦C; 5% CO2) to allow cells to attach. Test compounds were dissolved in 5% DMSO (*v*/*v*) and dilutions were generated from 300 μM to 300 nM. Aliquots (20 μL) of each dilution (or 5% aqueous DMSO for negative control and 5% aqueous SDS for positive control) were added to the plate in duplicate. After 68 h of incubation (37 ◦C; 5% CO2), a solution of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma, USA) in PBS (Phosphate Buffered Saline) was added to each well to a final concentration of 0.4 mg/mL and plates were incubated for a further 4 h (37 ◦C; 5% CO2) after which the medium was carefully aspirated and precipitated formazan crystals were dissolved in DMSO (100 μL/well). The absorbance of each well at 580 nm was measured with a PowerWave XS Microplate Reader from Bio-Tek Instruments Inc. (Vinooski, VT) and IC50 values were calculated as the concentration of the compound required for 50% inhibition of the cancer cells using Prism 5.0 from GraphPad Software Inc. (La Jolla, CA, USA). See Figure S81.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/md20110698/s1, Figure S1: ITS gene sequence of CMB-M0339, Figure S2: NCBI-BLAST search of 18S rRNA sequence of CMB-M0339, Figure S3: Blast search for CMB-M0339, Figure S4: Phylogenetic tree, Figure S5: CMB-M0339 cultivated on SD agar, Figure S6: UPLC-DAD chromatograms of crude extract of CMB-M0339, Figure S7: CMB-M0339 cultivated under MATRIX conditions, Figure S8: UPLC-DAD chromatograms of MATRIX extracts, Figure S9: Isolation scheme for **5–14**, Figures S10–S69: Annotated 1D and 2D NMR spectra and HRMS spectra for **5–14**, Figure S70–S71: Single ion extraction of fresh crude extract from CMB-M0339 showing the presence of **5–14** and **1**, Figures S72–S79: MS/MS fragmentation pattern of **5–9** and **i**-**v**, Figures S80–S81: Antimicrobial and cytotoxic activities of **5–14**. Table S1: Composition of media used for cultivation profiling, Tables S2–S11: 1D and 2D NMR data for **5–14**, Table S12: Predicted molecular formulae generated for **5–9** and **i**-**viii** observed in GNPS cluster, Table S13: Bond lengths and angles for X-ray crystal structure of **5**. Refs. [18,24] are cited in the Supplementary Materials.

**Author Contributions:** R.J.C. conceptualized the research; S.K. carried out the isolation, spectroscopic characterization, crystallization, and antibacterial, antifungal and cytotoxicity assays; P.V.B. performed the X-ray analyses; S.K. and Z.G.K. performed the taxonomic identification of the fungal strain; assigned molecular structures, and constructed the supplementary material; R.J.C. reviewed all data and drafted the manuscript, with support from S.K. and Z.G.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded in part by the Institute for Molecular Bioscience, The University of Queensland.

**Institutional Review Board Statement:** Not applicable.

**Data Availability Statement:** Raw NMR data is available at https://npmrd-project.org/ (access on 26 September 2022).

**Acknowledgments:** We thank R. Ritesh for isolation of CMB-M0339. S. K. thanks The University of Queensland for an International Postgraduate Scholarship.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


### *Article* **Thiolactones and Δ8,9-Pregnene Steroids from the Marine-Derived Fungus** *Meira* **sp. 1210CH-42 and Their** *α***-Glucosidase Inhibitory Activity**

**Hee Jae Shin 1,2,\*, Min Ah Lee 1,3, Hwa-Sun Lee 1,3 and Chang-Su Heo 1,2**


**Abstract:** The fungal genus *Meira* was first reported in 2003 and has mostly been found on land. This is the first report of second metabolites from the marine-derived yeast-like fungus *Meira* sp. One new thiolactone (**1**), along with one revised thiolactone (**2**), two new Δ8,9-steroids (**4**, **5**), and one known Δ8,9-steroid (**3**), were isolated from the *Meira* sp. 1210CH-42. Their structures were elucidated based on the comprehensive spectroscopic data analysis of 1D, 2D NMR, HR-ESIMS, ECD calculations, and the pyridine-induced deshielding effect. The structure of **5** was confirmed by oxidation of **4** to semisynthetic **5**. In the *α*-glucosidase inhibition assay, compounds **2**–**4** showed potent in vitro inhibitory activity with IC50 values of 148.4, 279.7, and 86.0 μM, respectively. Compounds **2**–**4** exhibited superior activity as compared to acarbose (IC50 = 418.9 μM).

**Keywords:** marine fungus; natural product; *Meira* sp.; thiolactone; pregnene steroid; epimer; stereochemistry; *α*-glucosidase inhibitor

#### **1. Introduction**

Fungi constitute one of the largest groups of organisms. Fungal-derived natural products (NPs) are pharmaceutically abundant, with several important biological applications ranging from highly potent toxins to approved drugs [1]. In particular, secondary metabolites obtained from marine fungi have garnered significant interest due to their unique chemical structures and potential biomedical applications [1,2]. While the number of cultivable marine fungi is extremely low (1% or less) compared to their global biodiversity [1,3], more than 1000 molecules have been reported and characterized from marine fungi, including alkaloids, lipids, peptides, polyketides, prenylated polyketides, and terpenoids [4–7]. Most research on secondary metabolites produced by marine fungi has primarily focused on a few genera, including *Penicillium*, *Aspergillus*, *Fusarium*, and *Cladosporium* [8,9]. Research into natural products derived from marine fungi is continually expanding, and as a result, a broader range of genera is now being investigated, with a particular focus on those associated with unique substrates and previously unexplored habitats [10–12].

In 2003, the genus *Meira* was first reported, namely *M. geulakonigii* and *M. argovae,* as a novel basidiomycetous [13]. *M. geulakonigii* was isolated from the citrus rust mite on pummelo (*Citrus grandis*), and *M. argovae* originated from a carmine spider mite on the leaves of castor bean (*Ricinus communis*) [8]. These *Meira* species have a similar morphology to yeast-like fungi. Nonetheless, the phylogenetic analysis of rDNA sequence data has identified *Meira* as a member of the Brachybasidiaceae family within the Exobasidiales, which

**Citation:** Shin, H.J.; Lee, M.A.; Lee, H.-S.; Heo, C.-S. Thiolactones and Δ8,9-Pregnene Steroids from the Marine-Derived Fungus *Meira* sp. 1210CH-42 and Their *α*-Glucosidase Inhibitory Activity. *Mar. Drugs* **2023**, *21*, 246. https://doi.org/10.3390/ md21040246

Academic Editor: Xian-Wen Yang

Received: 5 April 2023 Revised: 13 April 2023 Accepted: 14 April 2023 Published: 16 April 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

is classified under the Ustilaginomycetes (Basidiomycota) in the Exobasidiomycetidae group [14]. *M. geulakonigii* has been used successfully as a biological control agent against citrus and other phytophagous mites, as well as powdery mildew fungi [13,15–17]. A potential biocontrol agent against five mite species has been demonstrated for *M. argovae* [18]. Recently, *M. nicotianae* came from the rhizosphere of tobacco root, and that strain has the capability to promote plant growth possible in similar ways as plant growth-promoting fungi and arbuscular mycorrhizal fungi [19].

In this study, we isolated a yeast-like fungal species from a seawater sample. Phylogenetic analysis of ITS rDNA indicated that strain 1210CH-42 is closely related to other *Meira* species: *Meira* sp. M40, *M. nashicola* CY-1, and *M. miltonrushii* NIOSN-SK46-S121. So far, there are only a few reports on the isolation of *Meira* strains, but natural products from the genus *Meira* have not been investigated. This is the first report on the secondary metabolites from the marine-derived yeast-like fungus *Meira*. Herein, we report the isolation, structure elucidation, *α*-glucosidase inhibitory activity of **1**–**5**, and the structure revision of **2** isolated from the *Meira* strain 1210CH-42 (Figure 1).

**Figure 1.** Structures of **1**–**5** from the marine fungus strain *Meira* sp. 1210CH-42.

#### **2. Results and Discussion**

#### *2.1. Structure Elucidation of New Compounds*

Compound **1** was obtained as a white amorphous powder, and its molecular formula was determined to be C7H11NO2S by HR-ESIMS, with three degrees of unsaturation. The 1H and 13C NMR data of **<sup>1</sup>** are summarized in Table 1. The 1H NMR spectrum of **<sup>1</sup>** in CD3OD revealed two methine protons (*δ*<sup>H</sup> 4.79 and 2.86), one methylene proton (*δ*<sup>H</sup> 3.64 and 3.10), and two methyl protons (*δ*<sup>H</sup> 2.03 and 1.04). The 13C NMR and HSQC spectra showed the presence of seven signals, including two carbonyl carbons (*δ*<sup>C</sup> 206.5 and 173.8), two methines (*δ*<sup>C</sup> 63.9 and 36.0), one methylene (*δ*<sup>C</sup> 35.9) and two methyl carbons (*δ*<sup>C</sup> 22.4 and 13.0). The planar structure of **1** was elucidated by analysis of 1H-1H COSY and HMBC correlations (Figure 2). The COSY correlations from H-2 (*δ*<sup>H</sup> 4.79)/H-3 (*δ*<sup>H</sup> 2.86), H-3 (*δ*<sup>H</sup> 2.86)/H-4 (*δ*<sup>H</sup> 3.64), and H-3 (*δ*<sup>H</sup> 2.86)/H-5 (*δ*<sup>H</sup> 1.04) were observed. In addition, the HMBC correlations from H-2 (*δ*<sup>H</sup> 4.79) to C-1 (*δ*<sup>C</sup> 206.5)/C-3 (*δ*<sup>C</sup> 36.0)/C-5 (*δ*<sup>C</sup> 13.0)/C-7 (*δ*<sup>C</sup> 173.8), H-4 (*δ*<sup>H</sup> 3.10 and 3.64) to C-1 (*δ*<sup>C</sup> 206.5)/C-2 (*δ*<sup>C</sup> 63.9)/C-5 (*δ*<sup>C</sup> 13.0) and H-8 (*δ*<sup>H</sup> 2.03) to C-7 (*δ*<sup>C</sup> 206.5) suggested that **1** has a ring system, and confirmed the planar structure of **1**.

Detailed analysis of <sup>3</sup>*J*H,H coupling constants and 1D NOESY data determined the relative configuration of **1**. The relative stereochemistry of C-2 could be established by the observation of strong selective 1D NOESY correlations between H-2 and H-3/H-4b, between H-4b and H-2/H-3, and between H-5 and H-4a (Figure 2). These correlations suggested that the relative configurations of C-2 and C-3 must be *cis* rather than *trans*configuration in **1**. Thus, the relative configuration of **1** could be assigned as 2*S*\*, 3*R*\*. To determine the absolute configuration of **1**, the theoretical electronic circular dichroism (ECD) spectra of **1** and its enantiomer were calculated. The experimental ECD spectrum of **1**

showed a good agreement with the calculated ECD spectrum of the 2*S*, 3*R*-isomer (Figure 3). Therefore, the structure of **1** was elucidated to be a 2*S*-acetamide-3*R*-methyl-thiolactone.


**Table 1.** 1H and 13C NMR data of **1** and **2** (600 MHz for 1H and 150 MHz for 13C, in CD3OD).

**Figure 2.** 1H-1H COSY and key 2D NMR correlations of **1** and **2**.

**Figure 3.** Experimental CD spectra and the calculated ECD spectra of **1** and **2**.

Compound **2** was isolated as a white amorphous powder. The molecular formula of **2** was the same as that of **1** (C7H11NO2S) based on the HR-ESIMS data. Furthermore, the 1D NMR data of **2** were also similar but not identical to those of **1** (Table 1). The planar structure of **2** was determined to be the same as **1** by analysis of 1H-1H COSY and HMBC data (Figure 2). However, the 1H and 13C chemical shifts of **2** were different from **1**, especially those for the chiral centers, suggesting that the stereochemistry of **2** might be different from **1**. The relative configuration of **2** was also determined by analysis of <sup>3</sup>*J*H,H coupling constants and selective 1D NOESY data. The relative stereochemistry of C-2 could be established through the observation of strong NOESY contacts between H-2 and H-4a/H-5, between H-4a and H-2/H-5, and between H-4b and H-3. A relatively large coupling constant was observed between H-2 and H-3 (3*J*H,H = 12.5 Hz). Thus, the relative configurations of H-2 and H-3 had a *trans*-configuration in **2** (Figure 2). The *J*-based configurational analysis and NOESY measurements could not discriminate the possible relative configurations for (2*S*\*, 3*S*\*) or (2*R*\*, 3*R*\*). To solve this issue and to determine the absolute configuration of **2**, the ECD spectra of **2** and its enantiomer were calculated. The experimental ECD spectrum of **2** showed a good agreement with the calculated ECD spectrum of the 2*R*, 3*R*-isomer (Figure 3). Therefore, the structure of **2** was elucidated as an epimer of **1** and to be a 2*R*-acetamide-3*R*-methyl-thiolactone.

Notably, the 1H and 13C NMR data in CDCl3 of **2** were almost the same as those of the previously reported thiolactone with 2*R*, 3*S*-configuration isolated from a *Penicillium chrysogenum* (Table S1 and Figure S15) [20]. The reported compound with 2*R*, 3*S*-configuration possesses the same planar structure as **2** in this study. In the original paper for the compound with 2*R*, 3*S*-configuration, by the NOE correlation between H-3 (*δ*<sup>H</sup> 2.24) and H-2 (*δ*<sup>H</sup> 4.45), the authors insisted that the two protons were oriented on the same side of the ring system. However, its 1D NOE spectrum for the reported compound showed signals from H-3 (*δ*<sup>H</sup> 2.24) to H-2/H-4/H-5/H-6 and NH, making it unclear to determine the orientation of H-3 to the same side of H-2 or not (Figures S15 and S16). Moreover, if the reported configuration is correct, H-2 and H-3 are in *syn* relation, and they should have a small scalar coupling constant, but H-2 in the reported thiolactone had a large coupling constant (12.5 Hz) as in the revised structure (Table S1). In this study, we carefully compared and checked the selective 1D NOESY data of **2** with those for the reported compound. As noted above, **2** exhibited strong NOE correlations from H-2 to H-5/ H-4a and from H-4b to H-3 but not from H-4b to H-2, suggesting that H-2 and H-5 are on the same face. Furthermore, the reported compound with 2*R*, 3*S*-configuration and **1** (2*S*, 3*R*-configuration) are enantiomers and should have the same but opposite-in-sign specific rotation values. However, the optical rotation values of the reported thiolactone and **1** were [α] 25 <sup>D</sup> +1.5 (*c* 0.1, MeOH) and [α] 25 <sup>D</sup> +60.0 (*c* 0.1, MeOH), respectively. Considering all these results, the structure of the reported compound (2*R*, 3*S*-configuration) should be revised to 2*R*-acetamide-3*R*-methyl-thiolactone (Figure 4).

**Figure 4.** Reported and revised structures of **2**.

Compound **3** was isolated as a white amorphous powder, and its molecular formula was determined to be C21H32O2. By the comparison of the 1H and 13C NMR (Table 2), HR-ESIMS, and optical rotation data of **3** with those reported previously in the literature, **3** was identified as a known compound, (+)-03219A, Δ8,9-3*β*-hydroxy-5*α*-17-acetyl steroid [21–23].


**Table 2.** 1H and 13C NMR data of **3**–**5** (600 MHz for 1H and 150 MHz for 13C, in CD3OD).

**<sup>1</sup>** Signals were overlapped with other signals.

Compound **4** was purified as a white amorphous powder, and its molecular formula was determined to be C21H32O2 by HR-ESIMS, which is identical to that of **3**, with 6 degrees of unsaturation. The 1H and 13C NMR data of **4** are summarized in Table 2. The 1H NMR data for **4** revealed the signals of three methyl groups (*δ*<sup>H</sup> 0.57, 0.94, and 2.13), one oxymethine (*δ*<sup>H</sup> 3.97), nine methylenes, and three *sp*<sup>3</sup> methines. The 13C NMR and HSQC data of **4** exhibited 21 carbon signals containing three methyls (*δ*<sup>C</sup> 13.2, 17.3, and 31.7), one oxymethine (*δ*<sup>C</sup> 67.2), nine methylenes, two olefinic quaternary carbons (*δ*<sup>C</sup> 129.0 and 137.2), two *sp*<sup>3</sup> quarternary carbons (*δ*<sup>C</sup> 37.6 and 45.1), and one ketone carbonyl carbon (*δ*<sup>C</sup> 212.5). The planar structure of **4** was elucidated by 1H-1H COSY and HMBC data (Figure 5). The 1H-1H COSY correlations suggested the presence of four 1H-1H spin systems: from H-1 to H-4, from H-5 to H-7, from H-11 to H-12, and from H-14 to H-17. The HMBC correlations from H-6/H-7/H-11/H-14/H-15 to C-8 (*δ*<sup>C</sup> 129.0) and from H-11/H-12/H-14/H-19 to C-9 (*δ*<sup>C</sup> 137.2) indicated a double bond was located at C-8 and C-9. Additionally, the HMBC correlations from H-21 to C-17 (*δ*<sup>C</sup> 63.5)/C-20 (*δ*<sup>C</sup> 212.5) supported the assignment of an acetyl moiety connected to C-17 of the five-membered ring. The planar structure of **4** was the same as that of **3**, (+)-03219A [23], except for the difference in the chemical shifts around the oxymethine (*δ*<sup>H</sup> 3.97 and *δ*<sup>C</sup> 67.2) at C-3, suggesting that the stereochemistry of C-3 might be different from **3** (Figure 1 and Table 2). The stereochemistry of **4** was determined by analysis of the ROESY spectrum, 1D NOESY data, coupling constants, and the pyridine-induced deshielding effect. The relative configuration of **4** was confirmed by the ROESY correlations from H-3 to H-2a/H-2b/H-4, from H-19 to H-2b/H-4/H-11/H-18, and from H-18 to H-15/H-21 (Figure 5). The selective 1D NOE correlations were observed

from H-3 to H-2a/H-2b/H-4/H-19 (Figure S27). Furthermore, the small coupling constant of H-3 at *δ*<sup>H</sup> 3.97 (t, *J* = 2.8) was indicative of the C-3 hydroxyl group being axial from an examination of the Dreiding model (Table 2 and Figure 5) [24]. The significant deshielded chemical shifts of Heq-3 (Δ*δ*<sup>H</sup> = +0.32) and Hax-5 (Δ*δ*<sup>H</sup> = +0.48) in pyridine-*d*<sup>5</sup> compared with those in CD3OD indicated that OH-3 and H-5 adopted *α*-orientation, supporting the identified orientation (Figure 6 and Figure S29) [25–28]. Consequently, the structure of **4** was determined as a new epimer of **3**, Δ8,9-3*α*-hydroxy-5*α*-17-acetyl steroid.

**Figure 5.** 1H-1H COSY and key 2D NMR correlations of **4**.

**Figure 6.** Pyridine-induced deshielding effects of **4** (Δ*δ* = *δ*<sup>H</sup> in C5D5N-*δ*<sup>H</sup> in CD3OD).

Compound **5** was obtained as a white amorphous powder. The NMR data of **5** were similar to those of **4**, except for the absence of signals for the oxymethine at C-3 (*δ*<sup>H</sup> 3.97 and *δ*<sup>C</sup> 67.2) in **4** and the appearance of a ketone signal at C-3 (*δ*<sup>C</sup> 214.6) in **5** (Table 2), revealing that **5** would be an oxidized form of **4**. The 1H and 13C NMR spectra, compared to those of **3** and **4**, showed the significantly deshielded chemical shifts of C-2 (*δ*<sup>H</sup> 2.31/2.53 and *δ*<sup>C</sup> 39.1) and C-4 (*δ*<sup>H</sup> 2.11/2.40 and *δ*<sup>C</sup> 45.7). Additionally, the HMBC correlations between H-2b (*δ*<sup>H</sup> 2.53)/H-4 (*δ*<sup>H</sup> 2.11/2.40) and C-3 (*δ*<sup>C</sup> 214.6) determined the position of the ketone at C-3 (Figure 7). To clearly confirm the structure of **5**, **4** was oxidized to obtain the semisynthetic **5**. Both **5** and semisynthetic **5** exhibited identical 1H NMR, HSQC, and HMBC spectra (Figures S35, S36 and S37). The molecular formula of semisynthetic **5** was determined to be C21H30O2 by HR-ESIMS (*m*/*z* 337.2134 [M + Na]+, calcd. for C21H30O2Na, 337.2138). Based on these results, the structure of **5** was determined as a 3-keto derivative of **4**, with 7 degrees of unsaturation. Therefore, the structures of **5** and semisynthetic **5** were designated as Δ8,9-5*α*-3,20-dione-17-acetyl steroids.

#### *2.2. α-Glucosidase Inhibitory Activities of Compounds*

Compounds **1**–**4** were evaluated for *α*-glucosidase inhibitory activities (Table 3). Compound **4** exhibited the most significant inhibitory effect with an IC50 value of 86.0 μM, while **2** and **3** showed moderate activities with IC50 values of 148.4 and 279.7 μM, respectively. Further, **1** exhibited weak inhibitory activity at a concentration of 400 μM. The change in the stereochemistry of the compounds remarkably altered the *α*-glucosidase inhibitory

activities. Compounds **1** and **2**, as well as **3** and **4**, are stereoisomers of each other. Compounds **2** and **4** showed stronger *α*-glucosidase inhibitory effects than **1** and **3**. It could be noted herein that the stereochemistry was important for *α*-glucosidase inhibitory activity.

**Figure 7.** 1H-1H COSY and key HMBC correlations of **5**.

**Table 3.** *α*-Glucosidase inhibitory activities of **1**–**4**.


<sup>1</sup> The 50% inhibitory concentration (μM). <sup>2</sup> Acarbose is used as a positive control.

#### **3. Materials and Methods**

#### *3.1. General Experimental Procedures and Reagents*

NMR spectra were acquired with a Bruker AVANCE III 600 spectrometer (Bruker BioSpin GmbH, Rheinstetten, Germany) with a 3 mm probe operating at 600 MHz (1H) and 150 MHz (13C). Chemical shifts were expressed in ppm with reference to the solvent peaks (*δ*<sup>H</sup> 3.31 and *δ*<sup>C</sup> 49.15 ppm for CD3OD, *δ*<sup>H</sup> 7.26 and *δ*<sup>C</sup> 77.26 ppm for CDCl3). UV spectra were recorded with a Shimadzu UV-1650PC spectrophotometer (Shimadzu Corporation, Kyoto, Japan). IR spectra were obtained on a JASCO FT/IR-4100 spectrophotometer (JASCO Corporation, Tokyo, Japan). Optical rotations were measured with a Rudolph analytical Autopol III S2 polarimeter (Rudolph Research Analytical, Hackettstown, NJ, USA). LR-ESIMS data were obtained with an ISQ EM mass spectrometer (Thermo Fisher Scientific Korea Ltd., Seoul, Republic of Korea). HR-ESIMS data were obtained with a Waters SYNPT G2 Q-TOF mass spectrometer (Waters Corporation, Milford, CT, USA) at Korea Basic Science Institute (KBSI) in Cheongju, Republic of Korea and a Sciex X500R Q-TOF spectrometer (Framingham, MA, USA). ECD spectra were recorded with a JASCO J-1500 polarimeter at the Center for Research Facilities, Changwon National University, Changwon, Republic of Korea. HPLC was performed using a BLS-Class pump (Teledyne SSI, Inc., State College, PA 16803, USA) with Shodex RI-201H refractive index detector (Shoko Scientific Co., Ltd., Yokohama, Japan). Columns for HPLC were YMC-ODS-A (250 mm × 10 mm, 5 μm; and 250 mm × 10 mm, 5 μm) and YMC-Triart (250 mm × 10 mm, 5 μm; and 250 mm × 10 mm 5 μm). C18-reversed-phase silica gel (YMC-Gel ODS-A, 12 nm, S-75 μm) was used for open-column chromatography. Organic solvents were purchased as HPLC grade, and ultrapure waters were obtained from the Milipore Mili-Q Direct 8 system (Milipore S.A.S. Molsheim, France). The reagents used in the bioassay were purchased

from Sigma-Aldrich (Merck Korea, Seoul, Republic of Korea) and Tokyo Chemical Industry (TCI Co., Ltd., Tokyo, Japan).

#### *3.2. Fungal Strain and Fermentation*

The strain 1210CH-42 was isolated from a seawater sample collected at Chuuk Islands, Federated States of Micronesia, in 2010. The seawater sample was filtered, concentrated, and diluted (10−<sup>1</sup> and 10−2) with sterile seawater under aseptic conditions. Then the diluted sample was spread on Bennett's agar plates (1% D-glucose, 0.2% tryptone, 0.1% yeast extract, 0.1% beef extract, 0.5% glycerol, 1.7% agar, sea salt 32 g/L, pH 7.0). The plates were incubated for 7 days at 28 ◦C, and the single colony of the strain 1210CH-42 was collected. The fungus was identified as *Meira* sp. (GenBank accession number OQ693946) by DNA amplification and sequencing of the ITS region of the rRNA gene. The used primers were ITS4 (TCCTCCGCTTATTGATATGC) and ITS5 (GGAAGTAAAAGTCGTAACAAG G). The cultures of the strain 1210CH-42 were performed in modified Bennett's broth medium (1% D-glucose, 0.2% tryptone, 0.1% yeast extract, 0.1% beef extract, 0.5% glycerol, sea salt 10 g/L, pH 7.0). A seed culture was prepared from a spore suspension of the strain 1210CH-42 by inoculating into 1 L flasks and incubating it at 28 ◦C for 5 days on a rotary shaker at 120 rpm. The seed culture was inoculated aseptically into 2 L flasks (total 32 flasks) containing 1.0 L of medium and a 20 L fermenter containing 18 L of sterilized culture medium (0.1% *v*/*v*), respectively. The large-scale fermentation was done under the same conditions as the seed culture for 8 days and then harvested.

#### *3.3. Extraction and Isolation of Compounds* **1***–***5**

The culture broth (total 50 L) of the strain 1210CH-42 was harvested by high-speed centrifugation (60,000 rpm), and then the supernatant was extracted two times with ethyl acetate (100 L). The EtOAc extract was evaporated to afford a crude extract (3.05 g). The crude extract was subjected to ODS open column chromatography (YMC Gel ODS-A, 12 nm, S75 μm) followed by stepwise gradient elution with MeOH/H2O (*v*/*v*) (20:80, 40:60, 60:40, 80:20, and 100:0) as eluent. The 20% MeOH fraction was purified by a reversedphase HPLC (YMC ODS-A column, 250 × 10 mm i.d., 5 μm; 10% MeOH in H2O; flow rate: 1.5 mL/min; detector: RI) to yield **1** (2.9 mg, *t*<sup>R</sup> 44.0 min). Peak 10 from the 20% MeOH fraction was further purified by a reversed-phase HPLC (YMC ODS-A column, 250 × 10 mm i.d., 5 μm; 5% MeOH in H2O; flow rate: 1.5 mL/min; detector: RI) to yield **2** (0.6 mg, *t*<sup>R</sup> 64.0 min). The 80% MeOH fraction was purified by a reversed-phase HPLC (YMC ODS-A column, 250 × 10 mm i.d., 5 μm; 70% MeOH in H2O; flow rate: 1.5 mL/min; detector: RI) to yield **3** (0.6 mg, *t*<sup>R</sup> 84.0 min), **4** (2.1 mg, *t*<sup>R</sup> 95.5 min), and **5** (0.3 mg, *t*<sup>R</sup> 79.0 min).

Compound **1**: White amorphous powder; [*α*] 25 <sup>D</sup> +60.0 (*c* 0.1, MeOH); UV (MeOH) *λ*max (log *<sup>ε</sup>*) 204 (3.64), 235 (3.33) nm; IR (MeOH) *<sup>ν</sup>*max 3296, 2940, 1667, 1548, 1448, 1021 cm−1; 1H and 13C NMR data (CD3OD), see Table 1; HR-ESIMS *<sup>m</sup>*/*<sup>z</sup>* 196.0408 [M + Na]+, calcd. for C7H11NO2NaS, 196.0408.

Compound **2**: White amorphous powder; [*α*] 25 <sup>D</sup> +10.0 (*c* 0.1, MeOH); UV (MeOH) *λ*max (log *ε*) 206 (3.88), 234 (3.69) nm; IR (MeOH) *ν*max 3275, 2933, 1700, 1650, 1548, 1448, 1021 cm−1; 1H and 13C NMR data (CD3OD), see Table 1; HR-ESIMS *m*/*z* 196.0406 [M + Na]+, calcd. for C7H11NO2NaS, 196.0408.

Compound **3**: White crystalline needles; [*α*] 25 <sup>D</sup> +86.0 (*c* 0.1, MeOH); UV (MeOH) *λ*max (log *ε*) 202 (4.10) nm; IR (MeOH) *ν*max 3371, 2925, 2855, 1703, 1452, 1357, 1032 cm−1; 1H and 13C NMR data (CD3OD), see Table 2; HR-ESIMS *m*/*z* 339.2297 [M + Na]+, calcd. for C21H32O2Na, 339.2300.

Compound **4**: White crystalline needles; [*α*] 25 <sup>D</sup> +97.3 (*c* 0.1, MeOH); UV (MeOH) *λ*max (log *ε*) 202 (3.96) nm; IR (MeOH) *ν*max 3286, 2925, 2870, 1703, 1452, 1353, 1025 cm−1; 1H and 13C NMR data (CD3OD), see Table 2; HR-ESIMS *m*/*z* 339.2301 [M + Na]+, calcd. for C21H32O2Na, 339.2300.

Compound **5**: White amorphous; [*α*] 25 <sup>D</sup> +63.3 (*c* 0.1, MeOH); UV (MeOH) *λ*max (log *ε*) 204 (3.86) nm; IR (MeOH) *ν*max 3378, 2933, 2866, 1707, 1456, 1367, 1036 cm−1; 1H and 13C NMR data (CD3OD), see Table 2.

Oxidation of **4**. To a compound **4** (2.0 mg, 6.32 μmol) in anhydrous CH2Cl2 (0.5 mL) was added Dess-Martin reagent (8.04 mg, 18.96 μmol) at 0 ◦C. The mixture was stirred at r.t. for 24 h under N2 gas. The solution was washed with 5% NaHCO3 and brine and concentrated under reduced pressure [29,30]. Then the reactant was partitioned with EtOAc and H2O. The EtOAc layer was concentrated, and subjected to a reversed-phase HPLC (YMC-Triart C18 column, 250 × 10 mm i.d., 5 μm; 70% MeOH in H2O; flow rate: 2.0 mL/min; detector: RI) to yield semisynthetic **5** (0.5 mg): white amorphous solid; 1H NMR (600 MHz, CD3OD, representative signals) *δ*<sup>H</sup> 2.71 (t, *J* = 8.7 Hz, 1H), 2.53–2.31 (m, 2H), 2.40–2.08 (t, *J* = 14.6, o. l, 2H), 2.30 (o. l, H), 2.23 (o. l, 2H), 2.14 (s, 3H), 2.21–1.71 (o. l, 2H), 2.07 (o. l, 2H), 1.81 (m, 2H), 1.70–1.47 (o. l, 2H), 1.56 (o. l, 2H), 1.45 (o. l, 2H), 1.18 (s, 3H), 0.60 (s, 3H); 13C NMR data from HMBC spectrum (CD3OD, representative signals) *δ*<sup>C</sup> 214.5, 212.3, 135.6, 130.4, 63.5, 53.2, 45.7, 45.0, 44.4, 39.1, 38.1, 37.2, 31.7, 28.4, 26.7, 25.3, 24.3, 24.1, 23.7, 17.5, 13.1; HR-ESIMS *m*/*z* 327.2134 [M + Na]+, calcd. for C21H30O2Na, 327.2138.

#### *3.4. Computational Analysis*

The initial geometry optimization and conformational searches were generated using the Conflex 8 (Rev. B, Conflex Corp., Tokyo, Japan). The optimization and calculation for ECD were carried out using the Gaussian 16 program (rev. B.01, Gaussian Corp., Wallingford, C.T., USA). Conformational searches were performed using MMFF94s force field calculations with a 10 kcal/mol search limit. The conformers were optimized using the ground state method at the B3LYP/6-311+G (d, p) level in MeOH with a default model for ECD. The theoretical calculations of ECD spectra were performed using TD-SCF at the B3LYP /6-311+G (d, p) level in the gas phase. The ECD spectra were simulated by SpecDis (v. 1.71) using *σ* = 0.30–0.50 eV. All calculated curves were shifted to +10 nm to simulate experimental spectra better.

#### *3.5. Measurement of α-Glucosidase Inhibitory Activity*

The evaluation of *α*-glucosidase inhibitory activity was performed with reference to previously reported literature [31,32]. All the assays were carried out under 0.1 M PBS buffer (pH 7.4, Sigma). The samples (10 mM) were dissolved with DMSO (Sigma) and diluted into gradient concentrations with PBS buffer. The pre-reaction mixture consisted of the 130 μL sample with 30 μL *α*-glucosidase solution (0.2 U/mL, Sigma) and shaken well, then added to a 96-well plate and placed at 37 ◦C for 10 min in an incubator. Subsequently, 40 μL of 5 mM *p*-nitrophenyl-*α*-D-glucopyranoside (*p*NPG, TCI) was added and further incubated at 37 ◦C for 20 min. Finally, the *α*-glucosidase inhibitory activity was determined by measuring the release of *p*NPG at 405 nm of the microplate reader. The negative control was prepared by adding PBS buffer instead of the sample in the same way as the test. The blank was prepared by adding PBS buffer instead of *p*NPG using the same method. Acarbose was used as the positive control, and experiments were carried out in triplicate.

#### **4. Conclusions**

In summary, one new thiolactone (**1**), along with one revised thiolactone (**2**), two new Δ8,9-steroids (**4**, **5**), and one known Δ8,9-steroid (**3**), were isolated from the marinederived fungus *Meira* sp. 1210CH-42. The absolute configurations of **1** and **2** were determined by analysis of the selective 1D NOESY and ECD data. Compounds **1** and **2** were identified as a pair of acetamide epimers at C-2. While compounds **3** and **4** were identified as epimers for the hydroxyl group at C-3, which was confirmed by analysis of 1H NMR, ROESY, 1D NOESY, coupling constants, and the pyridine-induced deshielding effect. In addition, the structure of **5** was obtained as the 3-keto derivative of **3**. Compounds **1**–**4** were screened for their *α*-glucosidase inhibitory activity preliminarily. Compound **4** exhibited intense activity with an IC50 value of 86.0 μM. Furthermore, compounds **2** (IC50 = 148.4 μM)

and **3** (IC50 = 279.7 μM) demonstrated superior activity as compared to acarbose (IC50 = 418.9 μM). To the best of our knowledge, this is the first report of new bioactive metabolites with potent *α*-glucosidase inhibitory activity from the yeast-like fungus *Meira*. These results show that *Meira* sp. 1210CH-42 produces unique and diverse metabolites which have the potential for an anti-diabetic agent. The genus *Meira* is mostly found on land, and secondary metabolites from the marine-derived genus have not yet been reported. Therefore, further research is needed for the marine-derived fungus *Meira* sp. 1210CH-42 to discover novel secondary metabolites and investigate their biological properties.

**Supplementary Materials:** The following are available online at: https://www.mdpi.com/article/ 10.3390/md21040246/s1, Figures S1–S14: 1H, 13C NMR, HSQC, COSY, HMBC, selective 1D NOESY, and HR-ESIMS data of **1** and **2**, Table S1 and Figures S15–S16: 1H, 13C NMR data, and 1D NOESY data of the reported compound, Figures S17–S20: 1H, 13C NMR, HSQC, and HR-ESIMS data of **3**, Figures S21–S28: 1H, 13C NMR, HSQC, COSY, HMBC, ROESY, 1D NOESY, and HR-ESIMS data of **4**, Figure S29: Comparison of 1H data of **4** in pyridine-*d*<sup>5</sup> and in CD3OD, Figures S30–S34: 1H, 13C NMR, HSQC, COSY, and HMBC data of **5**, Figures S35–S38: 1H MMR, HSQC, HMBC, and HR-ESIMS data of semisynthetic **5.**

**Author Contributions:** Conceptualization, H.J.S.; investigation, M.A.L., H.-S.L. and C.-S.H.; resources, M.A.L.; writing—original draft preparation, M.A.L.; writing—review and editing, H.J.S.; project administration, H.J.S.; funding acquisition, H.J.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by the Korea Institute of Marine Science & Technology Promotion (KIMST) grant funded by the Ministry of Oceans and Fisheries, Korea (Grant no. 20220027) and the Korea Institute of Ocean Science and Technology (PEA0121).

**Institutional Review Board Statement:** Not applicable.

**Data Availability Statement:** The data presented in the article are available in the Supplementary Materials.

**Acknowledgments:** The authors express gratitude to Jung Hoon Choi, Korea Basic Science Institute, Ochang, Korea, for providing mass data.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

### *Article* **Diketopiperazine Alkaloids and Bisabolene Sesquiterpenoids from** *Aspergillus versicolor* **AS-212, an Endozoic Fungus Associated with Deep-Sea Coral of Magellan Seamounts**

**Yu-Liang Dong 1,2,3, Xiao-Ming Li 1,2,4, Xiao-Shan Shi 1, Yi-Ran Wang 1,2,3, Bin-Gui Wang 1,2,3,4,\* and Ling-Hong Meng 1,2,3,4,\***


**Abstract:** Two new quinazolinone diketopiperazine alkaloids, including versicomide E (**2**) and cottoquinazoline H (**4**), together with ten known compounds (**1**, **3**, and **5**–**12**) were isolated and identified from *Aspergillus versicolor* AS-212, an endozoic fungus associated with the deep-sea coral *Hemicorallium* cf. *imperiale*, which was collected from the Magellan Seamounts. Their chemical structures were determined by an extensive interpretation of the spectroscopic and X-ray crystallographic data as well as specific rotation calculation, ECD calculation, and comparison of their ECD spectra. The absolute configurations of (−)-isoversicomide A (**1**) and cottoquinazoline A (**3**) were not assigned in the literature reports and were solved in the present work by single-crystal X-ray diffraction analysis. In the antibacterial assays, compound **3** exhibited antibacterial activity against aquatic pathogenic bacteria *Aeromonas hydrophilia* with an MIC value of 18.6 μM, while compounds **4** and **8** exhibited inhibitory effects against *Vibrio harveyi* and *V. parahaemolyticus* with MIC values ranging from 9.0 to 18.1 μM.

**Keywords:** diketopiperazine; *Aspergillus versicolor*; deep-sea coral; endophytic fungus; antimicrobial activity

#### **1. Introduction**

Marine-derived fungi living under extreme survival conditions are considered as abundant sources of structurally diverse and biologically active compounds [1,2]. In the deep-sea habitats, seamounts are regarded locations for a wide variety of current-topography interactions and biophysical coupling which have large biomass and higher biodiversity than their surrounding deep-sea floors [3,4]. Endozoic fungi surviving in deep-sea seamounts are a promising new source to mining bioactive secondary metabolites owing to their unique habitats. To date, only three papers investigating bioactive secondary metabolites of fungi derived from deep-sea seamounts have been published [5–7]. Therefore, a study on the chemical diversity of deep-sea seamount-derived endozoic fungi is warranted.

The species in the fungal genus *Aspergillus*, especially *A. versicolor*, is widely distributed in various habitats (marine, terrestrial, and symbiotic sources) and possesses the ability to produce diversified bioactive secondary metabolites such as diketopiperazine alkaloids [8,9], peptides [10], xanthones [9,11], and sesquiterpenes [12]. Most of these metabolites are described to exhibit a variety of bioactivities, including antifungal [9], antitumor [10,11], and neuroprotective activities [12].

**Citation:** Dong, Y.-L.; Li, X.-M.; Shi, X.-S.; Wang, Y.-R.; Wang, B.-G.; Meng, L.-H. Diketopiperazine Alkaloids and Bisabolene Sesquiterpenoids from *Aspergillus versicolor* AS-212, an Endozoic Fungus Associated with Deep-Sea Coral of Magellan Seamounts. *Mar. Drugs* **2023**, *21*, 293. https://doi.org/10.3390/ md21050293

Academic Editor: Hee Jae Shin

Received: 4 April 2023 Revised: 4 May 2023 Accepted: 8 May 2023 Published: 10 May 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

In our continuous efforts to explore bioactive metabolites from deep-sea seamountderived fungi [5–7], chemical investigation of the endozoic fungus *Aspergillus versicolor* AS-212 associated with the deep-sea coral, *Hemicorallium* cf. *imperiale*, which was collected from the Magellan Seamounts in the Western Pacific Ocean was carried out due to its unique HPLC profiles. As a result, two new quinazolinone diketopiperazine alkaloids, namely, versicomide E (**2**) and cottoquinazoline H (**4**), together with five known related analogs (**1**, **3**, **5**–**7**) as well as four known bisabolene derivatives (**8**–**11**) and a bisabolene dimer (**12**), have been isolated and identified. Herein, we report the isolation and structure elucidation as well as the antimicrobial activities of compounds **1**–**12** (Figure 1).

**Figure 1.** Structures of compounds **1**–**12**, versicomides A and B, and (−)-isoversicomide A.

#### **2. Results and Discussion**

#### *2.1. Structure Elucidation of the Isolated Compounds*

Compound **1** was isolated as colorless crystals, and the molecular formula was established as C19H25N3O3 by analysis of the HRESIMS data. The 1H and 13C NMR of **1** (DMSO-*d*6, Table 1) extremely resembled those of versicomide A, a quinazoline-containing compound isolated from the crab-derived fungus *Aspergillus versicolor* XZ-4 which was collected from hydrothermal vent [8]. Further analysis of the 2D NMR spectra (Figure 2) indicated the same planar structure of **1** as that of versicomide A (Figure 1). However, a strong NOE cross-peak of H-3/H-20 was in favor of the structure with the 3*S*\*- and 14*R*\*-relative configuration rather than a 3*S*\* and 14*S*\* configuration (Figure 3). Singlecrystal X-ray diffraction analysis with Cu Kα radiation further demonstrated its structure and absolute configurations (Figure 4). A Flack parameter of 0.0(2) enabled the definition of its absolute configuration as 3*S*, 14*R*, and 15*S*, indicating that **1** was the 14-epimer of versicomide A.


**Table 1.** 1H (500 MHz) and 13C (125 MHz) NMR Spectroscopic Data for Compounds **1** and **2**.

<sup>a</sup> Recorded in DMSO-*d*6. <sup>b</sup> Recorded in CDCl3.

**Figure 2.** Key COSY (bold lines) and HMBC (blue arrows) correlations for compounds **1**, **2**, and **4**.

**Figure 3.** Key NOE correlations for compounds **1**, **2**, and **4** (red lines: *β*-orientation; blue lines: *α*-orientation).

**Figure 4.** X-ray crystal structures of compounds **1** and **3**.

Compound **1** was initially treated as a new quinazoline alkaloid during the preparation of this manuscript, while Tasdemir and co-workers recently reported a new quinazolinecontaining diketopiperazine (−)-isoversicomide A from the deep-sea sediment-derived fungus *Aspergillus versicolor* PS108-62 [13]. Notably, compound **1** shared the same planar structure and virtually similar optical rotation value ([α] 25 <sup>D</sup> −30 vs. [α] 20 <sup>D</sup> −25) as that of (−)-isoversicomide A, in which the stereogenic centers at C-3 and C-14 showed the same relative configurations with that of compound **1**. However, the configuration at C-18 on the short flexible aliphatic chain and the absolute configuration of (−)-isoversicomide A were not assigned due to the limited sample available [13]. Considering their similar rotation values and same relative configuration at C-3 and C-14, we assumed that compound **1** and (−)-isoversicomide A are the same compound. As the reported evidence to determine the absolute configuration of versicomide A does not seem entirely solid and in view of the highly similar NMR data of those isomers with multi-chiral centers, it is necessary to clarify the absolute configuration of **1**. The results from the X-ray diffraction analysis of compound **1** unambiguously determined its absolute configuration as 3*S*, 14*R*, and 15*S*. This is likely the first time the configuration of isoleucine in a quinazoline-containing diketopiperazine skeleton with a Val-Ile cyclic dipeptide moiety was unambiguously defined by X-ray crystallography analysis.

Versicomide E (**2**) was obtained as a colorless amorphous solid with the molecular formula C19H23N3O3 based on the HRESIMS data. Its NMR data (CDCl3, Table 1) were similar to those of **1**, which indicated that **2** possessed the same quinazoline backbone as **1**. The obvious difference was the absence of signals for two methines at *δ*<sup>C</sup> 58.1/*δ*<sup>H</sup> 4.70 (CH-3) and *δ*<sup>C</sup> 36.1/*δ*<sup>H</sup> 2.62 (CH-15) in the NMR spectra of **1**, whereas additional resonances corresponding to a tetra-substituted double bond at *δ*<sup>C</sup> 121.3 (C-3) and *δ*<sup>C</sup> 135.0 (C-15) were found in that of **2** (CDCl3, Table 1), which were further confirmed by COSY and HMBC correlations (Figure 2). The geometry of the double bond between C-3 and C-15 was determined as Z-configuration by key NOE correlations from NH-2 (*δ*<sup>H</sup> 8.02) to H-16 (*δ*<sup>H</sup> 2.29) and H3-17 (*δ*<sup>H</sup> 1.15) (Figure 3). Compound **2** has the same planar structure as that of versicomide B (Figure 1), which was also isolated from hydrothermal vent crab-derived fungus *Aspergillus versicolor* XZ-4 by Wu and co-workers in 2017 [8], with the exception of the geometry of the double bond at C3(15) (*Z* in **2** vs. *E* in versicomide B) and the absolute configuration of C-14 (*R* in **2** vs. *S* in versicomide B) as well. To clarify the stereochemistry of compound **2**, calculations of specific rotation (SR) were carried out for 14*R*-**2** and 14*S*-**2**, and the calculated SR value for 14*R*-**2** (+59.8) at CAM-B3LYP/TZVP level was compatible with the experimental SR value [α] 25 <sup>D</sup> +112.0 (*c* 0.08, MeOH), contrary to that of versicomide B ([α] 20 <sup>D</sup> −23.4) [8], which allowed the assignment of absolute configuration of C-14 in **2** as 14*R* (Table S2). To further verify the absolute configuration of C-14 in **2**, the time-dependent density functional (TDDFT)-ECD calculation was performed on 14*R*-**2** and 14*S*-**2** at the CAM-B3LYP/TZVP level in Gaussian 09. The experimental curve matched that of the calculated ECD spectrum for 14*R*-**2** and also assigned the absolute configuration of C-14 in **2** as 14*R* (Figure S27).

Compound **3** was obtained as colorless prisms and was identified as cottoquinazoline A by comparing its NMR data (measured in DMSO-*d*6, Table S3) with those previously reported in the literature [10]. Cottoquinazoline A is a 16-nor analog of the known fumiquinazoline D and was first isolated from a marine-derived fungal strain of *A. versicolor* (MST-MF495) by Capon and co-workers in 2009, with a partial stereostructure assigned [10]. Considering the complexity of the structure of **3** and the presence of many stereoisomers, it is important to clarify the assignment of the absolute configurations of **3** [10,14]. Fortunately, a suitable crystal of **3** was picked out from DMSO–MeOH (1:1) and subjected to X-ray crystallographic analysis to assign its absolute configurations of the stereogenic centers in **3** as 3*S*, 14*S*, 16*R*, 17*S*, and 19*S* (Figure 4).

Cottoquinazoline H (**4**) was obtained as a colorless amorphous solid. Its molecular formula was established as C24H21N5O4 by HRESIMS, with one CH2 unit more than that of **3**. Discreet comparisons of the NMR data (DMSO-*d*6, Table 2) and UV absorptions with **3** suggested that they shared the same core scaffold. However, the methyl substitution at C-20 in **3** was replaced by an ethyl group in **4**, as evidenced by the appearance of an additional methylene group resonating at *δ*<sup>C</sup> 21.0 and *δ*<sup>H</sup> 1.90/1.99 (CH2-29) in the NMR spectra of **4** (DMSO-*d*6, Table 2). Additionally, the chemical shift of C-20 was deshielded downfield from *δ*<sup>C</sup> 63.2 in **3** to *δ*<sup>C</sup> 68.1 in **4**. The COSY and HMBC correlations verified the above deduction (Figure 2). The relative configuration of **4** was also deduced from the analysis of NOESY experiments. The NOE cross-peaks from H-20 and H-15α to H-18 revealed the cofacial orientation of these groups (Figure 3). Given that the stereochemistry of co-isolated compound **3** was determined by X-ray diffraction analysis as well as their similar NMR chemical shifts and virtually identical experimental ECD curves (Figure 5), the absolute configurations of all chiral carbons in **4** were established as 3*S*, 14*S*, 17*R*, 18*S*, and 20*S*.


**Table 2.** 1H (500 MHz) and 13C (125 MHz) NMR data for compound **4** (in DMSO-*d*6).

In addition to compounds **1**–**4**, three related quinazolinone diketopiperazine alkaloids, namely, versicoloids A and B (**5** and **6**) [9], and chrysopiperazine A (**7**) [15], as well as five known bisabolene derivatives (**8**–**12**) including sydonic acid (**8**) [16], (*S*)-(+)-11 dehydrosydonic acid (**9**) [17], (−)-10-hydroxysydonic acid (**10**) [18], hydroxysydonic acid (**11**) [16], and peniciaculin B (**12**) [18] were also identified and isolated from the fungus *A. versicolor* AS-212, which were determined by the comparison of their NMR data and those previously described in the literature. Structurally, (−)-isoversicomide A (**1**) might be a plausible biosynthetic precursor that undergoes the transformation of the benzene ring to the oxepine ring to generate versicoloid A (**5**) [19], which provides the basis for the biosynthetic origins of versicoloid A.

**Figure 5.** Comparison of the experimental ECD spectra of compounds **3** (in red) and **4** (in blue) in CH3OH.

#### *2.2. Antimicrobial Assays*

The antimicrobial activity evaluation of all the isolated compounds was performed against human pathogenic bacterium (*Escherichia coli*), marine-derived aquatic pathogenetic bacteria (*Aeromonas hydrophila*, *Edwardsiella ictarda*, *Micrococcus luteus*, *Pseudomonas aeruginosa*, *Vibrio harveyi*, *V. parahemolyticus*, *V. vulnificus*), and plant-pathogenic fungi (*Colletotrichum gloeosporioides*, *Curvularia spicifera*, *Epicoccum sorghinum*, *Fusarium oxysporum*, *F. proliferatum*, and *Penicillium digitatum*) (Table 3). In the antimicrobial screening, compounds **4** and **8** exhibited potent inhibitory activity against the aquatic pathogenic bacterium *V. parahaemolyticus* with MIC values of 9.0 and 15.0 μM, while compounds **8** and **9** showed inhibitory activity against the aquatic pathogenic bacterium *V. harveyi* with the MIC values of 15.0 and 15.2 μM. In addition, compounds **3** and **4** displayed a broad spectrum of antimicrobial activity against most of the tested strains, with the MIC values ranging from 9.0 to 74.6 μM. The bisabolene derivatives (**8**–**12**) mainly exhibited activities against *M. luteus*, *V. harveyi*, and *V. parahaemolyticus,* with MIC values ranging from 15.0 to 121.2 μM. However, neither the quinazoline-containing diketopiperazine derivatives (**1** and **2**) nor the oxepine-containing diketopiperazine analogs (**5**–**7**) showed any activity against all the tested pathogenic bacteria. These data suggested that the 16-*nor*-methyl fumiquinazoline alkaloids generally showed higher antimicrobial activity than that of quinazolinone alkaloids (**3** and **4** vs. **1** and **2**) and the oxepine congeners (**5**–**7**). A comparison of the antimicrobial results of **3** and **4** revealed that different substituent groups at C-20 could influence the inhibitory potency against the pathogenic bacteria. Concerning bisabolene derivatives, the antimicrobial results revealed that compound **12**, a dimeric bisabolene analog, showed weaker antimicrobial activities than that of the monomeric bisabolenes (**8**–**11**) against *M. luteus* and *V. harveyi*. In addition, hydroxylation at C-10 or C-11 likely decreased the activity against *V. harveyi, V. parahaemolyticus,* and *C. gloeosporioides* (**8** vs. **10** and **11**).

**Table 3.** The antimicrobial activities of compounds **1**–**12** (MIC, μM) a.


<sup>a</sup> (-) = MIC > 200 μM; Positive control: <sup>b</sup> Chloromycetin; <sup>c</sup> amphotericin B.

The above results showed that compounds **4**, **8**, and **9** were found to be efficient in suppressing the growth of aquatic pathogenic bacteria *V. parahaemolyticus* and *V. harveyi*. To a great degree, the endozoic fungus *A. versicolor* AS-212 which is associated with the deep-sea coral *Hemicorallium* cf. *imperiale* may provide a chemical defense to help its host to fight off the aquatic pathogenic bacteria by producing an array of antimicrobial secondary metabolites.

#### **3. Experimental Section**

#### *3.1. General Experimental Procedures*

The general experimental procedures, apparatus, and solvents/reagents used in this work were the same as those described in our previous reports [5–7].

#### *3.2. Fungal Material*

The endophytic fungus *Aspergillus versicolor* AS-212 associated with deep-sea coral, *Hemicorallium* cf. *imperial*, was collected from the Magellan Seamounts (depth 1420 m) in May 2018. By comparing its ITS region sequence with that of *A. versicolor* (accession no. MT582751.1) in the GenBank database, the sequence data of strain AS-212 were identical (100%) to those of *A. versicolor* and subsequently uploaded in GenBank with accession no. OP009765.1. The fungus AS-212 has been conserved at the Key Laboratory of Experimental Marine Biology, Institute of Oceanology, Chinese Academy of Sciences (IOCAS).

#### *3.3. Fermentation, Extraction, and Isolation*

The fungal strain AS-212 was cultivated on potato dextrose agar (PDA) plates at 28 ◦C for 7 days to generate spores. The fresh mycelia were transferred into 1 L Erlenmeyer flasks, each containing 300 mL potato-dextrose broth (PDB) medium, which was reported in our previous publication [5], and fermented under static conditions for 30 days at room temperature. After 30 days of incubation, a total of 33 L cultures were filtered and collected to separate the broth and mycelia. The broth was adequately extracted three times with EtOAc, while the mycelia were mechanically crushed and then extracted three times with 80% volume aqueous acetone. Acetone was removed in vacuo to afford an aqueous solution, which was successively extracted with EtOAc. Based on their virtually similar TLC and HPLC profiles (Figure S26), both EtOAc extracts from broth and mycelia were combined and evaporated under a vacuum to render the EtOAc extract (61 g).

The EtOAc extract was subjected to vacuum liquid chromatography (VLC) eluted with petroleum ether (PE)–EtOAc gradient (20:1 to 1:1, *v*/*v*) and then CH2Cl2–MeOH (20:1 to 1:1, *v*/*v*) to afford nine fractions (Frs. 1–9). Fr. 4 (2.3 g) was fractionated by reverse-phase column chromatography (CC) with a MeOH–H2O gradient (from 10:90 to 100:0) to afford nine subfractions (Frs.4.1–4.9). Fr. 4.2 was directly purified by semi-preparative HPLC (Elite ODS-BP, 5μm; 10 × 250 mm; 80% MeOH–H2O, 2.5 mL/min) to yield compound **6** (3.0 mg, *t*<sup>R</sup> = 17 min). Fr. 4.4 (16 mg) was further purified by prep. TLC (plate: 20 × 20 cm, developing solvents: PE–EtOAc, 2:1) and by Sephadex LH-20 (MeOH) column to afford **7** (3.1 mg). Fr. 4.5 was purified by CC over Sephadex LH-20 chromatography (MeOH) and then by semi-preparative HPLC (85% MeOH–H2O, 2.5 mL/min) to give compound **1** (6.6 mg, *t*<sup>R</sup> = 14 min). Fr. 4.6 (75 mg) was fractionated by CC on Sephadex LH-20 column (MeOH) to yield five subfractions Frs.4.6.1–4.6.5. Fr. 4.6.5 (20 mg) was further purified by prep. TLC (developing solvents: DCM–MeOH, 20:1) and by Sephadex LH-20 (MeOH) to afford compound **9** (7.7 mg). Fr. 4.7 (36 mg) was directly purified by prep. TLC (developing solvents: CH2Cl2–EtOAc, 3:1) and by Sephadex LH-20 (MeOH) to afford compound **8** (4.0 mg). Fr. 4.8 (67 mg) was purified by CC on silica gel eluting with CH2Cl2–MeOH gradient (from 200:1 to 50:1) to obtain compound **12** (4.3 mg). Fr. 5 (3.4 g) was separated by reversed-phase CC using step-gradient elution with MeOH–H2O (from 10:90 to 100:0) to yield seven subfractions (Frs. 5.1–5.7). Fr. 5.2 (241 mg) was fractionated by CC on silica gel eluting with CH2Cl2–MeOH gradient (from 150:1 to 20:1) and then purified on Sephadex LH-20 (MeOH) to afford compounds **10** (11.3 mg) and **11** (4.3 mg). Fr. 5.4 (126 mg) was fractionated by CC on Sephadex LH-20 (MeOH) and further purified by

semi-preparative HPLC (70% MeOH–H2O, 2.5 mL/min) to afford compound **5** (6.3 mg, *t*<sup>R</sup> = 22 min). Fr. 5.6 was chromatographed via a Sephadex LH-20 column (MeOH) and then by semi-preparative HPLC (78% MeOH-H2O, 2.5 mL/min) to afford compound **2** (6.7 mg, *t*<sup>R</sup> = 20 min). Fr. 7 (5.8 g) was fractionated by reverse-phase CC with a MeOH–H2O gradient (from 10:90 to 100:0) to yield five subfractions (Frs. 7.1–7.5). Fr. 7.5 (328 mg) was applied to silica gel CC eluted with CH2Cl2/MeOH to give nine subfractions (Frs. 7.5.1–7.5.9). Fr.7.5.7 (36 mg) was purified by semi-preparative HPLC (45% MeCN–H2O, 2.5 mL/min) to provide compounds **3** (4.3 mg, *t*<sup>R</sup> = 9 min) and **4** (3.2 mg, *t*<sup>R</sup> = 12 min).

(−)-Isoversicomide A (**1**): colorless crystals; mp 197−199 ◦C; [α] 25 <sup>D</sup> −30 (*c* 0.10, MeOH); UV (MeOH) *λ*max (log *ε*) 227 (3.47), 277 (3.00), 326 (2.58) nm; ECD (0.52 mM, MeOH) *λ*max (Δ*ε*) 210 (−4.87), 233 (−15.03), 277 (+1.88), 328 (−1.02) nm; for 1H and 13C NMR data, see Table 1; HRESIMS *m/z* 344.1960 [M + H]+ (calcd for C19H26N3O3, 344.1969).

Versicomide E (**2**): colorless amorphous solid; [α] 25 <sup>D</sup> +112 (*c* 0.08, MeOH); UV (MeOH) *λ*max (log *ε*) 221 (3.26), 309 (2.88) nm; ECD (0.59 mM, MeOH) *λ*max (Δ*ε*) 207 (+6.69), 233 (−3.87), 255 (+6.18), 300 (+1.41), 343 (−1.15) nm; for 1H and 13C NMR data, see Table 1; HRESIMS *m/z* 340.1663 [M − H]<sup>−</sup> (calcd for C19H22N3O3, 340.1667).

Cottoquinazoline A (**3**): colorless prisms (MeOH-DMSO 1:1); mp 215−217 ◦C; [α] 25 D +160 (*c* 0.10, MeOH); UV (MeOH) *λ*max (log *ε*) 205 (3.39), 227 (3.17), 256 (2.81), 268 (2.74), 280 (2.68), 305 (2.26), 315 (2.14) nm; ECD (0.58 mM, MeOH) *λ*max (Δ*ε*) 211 (−13.16), 230 (+12.22), 308 (+3.60) nm; for 1H and 13C NMR data, see Table S2.

Cottoquinazoline H (**4**): colorless amorphous solid; [α] 25 <sup>D</sup> +150 (*c* 0.10, MeOH); UV (MeOH) *λ*max (log *ε*) 205 (3.60), 227 (3.43), 257 (3.11), 270 (3.03), 279 (2.96), 304 (2.52), 317 (2.40) nm; ECD (0.56 mM, MeOH) *<sup>λ</sup>*max (Δ*ε*) 212 (−22.91), 231 (+21.65), 308 (+7.16) nm; for 1H and 13C NMR data, see Table 2; HRESIMS *m/z* 444.1663 [M + H]+ (calcd for C24H22N5O4, 444.1666).

#### *3.4. X-ray Crystallographic Analysis of Compounds* **1** *and* **3**

Suitable crystals were picked out to obtain crystallographic data using a Bruker Smart-1000 or Bruker D8 VENTURE CCD diffractometer with Cu Kα radiation (λ = 1.54178 Å). Absorption correction was applied using the program SADABS [20]. The structures were solved by direct methods with the SHELXTL software package [21,22]. All non-hydrogen atoms were refined anisotropically. The absolute structures were determined by refinement of the Flack parameter [23]. The structures were optimized by full-matrix least-squares techniques. Crystallographic data have been deposited with the Cambridge Crystallographic Data Centre with deposition numbers CCDCs 2192654 and 2192653 for **1** and **3**, respectively. Crystal data and structure refinements for **1** and **3** are listed in Table S1.

*Crystal data for compound 1:* C19H25N3O3, F.W. = 343.2, space group P2(1)2(1)2(1), unit cell dimensions a = 13.2384(3) Å, b = 19.9347(4) Å, c = 6.8103(2) Å, V = 1797.26(8) Å3, <sup>α</sup> <sup>=</sup> <sup>β</sup> <sup>=</sup> <sup>γ</sup> = 90◦, Z = 4, dcalcd = 1.269 g/cm3, crystal dimensions 0.350 × 0.330 × 0.300 mm, μ = 0.702 mm–1, F(000) = 736. The 4007 measurements yielded 2826 independent reflections after equivalent data were averaged. The final refinement gave R1 = 0.0379 and wR2 = 0.0989 [I > 2σ(I)]. Flack parameter = 0.0(2).

*Crystal data for compound 3:* 2(C23H18N5O4)·C2OS2, F.W. = 960.99, orthorhombic space group C2221, unit cell dimensions a = 9.4022(11) Å, b = 25.878(4) Å, c = 19.112(2) Å, V = 4650.2(11) Å3, α = β = γ = 90◦, Z = 4, dcalcd = 1.373 g/cm3, crystal dimensions 0.200 × 0.180 × 0.150 mm, <sup>μ</sup> = 1.612 mm–1, F(000) = 1992. The 21,138 measurements yielded 4267 independent reflections after equivalent data were averaged. The final refinement gave R1 = 0.0951 and wR2 = 0.2851 [I > 2σ(I)]. Flack parameter = 0.145(12).

#### *3.5. Antimicrobial Assay*

A two-fold serial dilution method using 96-well microtiter plates was applied to evaluating the antimicrobial activities against a panel of aquatic pathogenic bacteria (*Aeromonas hydrophilia* QDIO-1, *Edwardsiella ictarda* QDIO-9, *Micrococcus luteus* QDIO-3, *Pseudomonas*

*aeruginosa* QDIO-4, *Vibrio harveyi* QDIO-7, *V. parahaemolyticus* QDIO-8, and *V. vulnificus* QDIO-10), one human pathogenic bacterium (*Escherichia coli* EMBLC-1), and six plantpathogenic fungi (*Penicillium digitatum* QDAU-3, *Colletotrichum gloeosporioides* QA-29, *Fusarium oxysporum* QDAU-8, *Curvularia spicifera* QA-26, *Epicoccum sorghinum* QA-20, and *F. proliferatum* QA-28) [24]. The aquatic pathogenic strains and human pathogenic bacterium were provided by IOCAS, while the plant-pathogenic fungi were provided by IOCAS and Qingdao Agricultural University. To assay the antimicrobial activities, DMSO was added to dissolve all isolated compounds and positive control (chloramphenicol and amphotericin B) to prepare a stock solution with a specific concentration.

#### *3.6. Specific Rotation and ECD Calculations*

General computational procedures were consistent with our previous reports [5,25].

#### **4. Conclusions**

In conclusion, two new quinazolinone derivatives, versicomide E (**2**) and cottoquinazoline H (**4**), along with ten known compounds (**1**, **3**, and **5**–**12**), were isolated and identified from the deep-sea coral-derived *Aspergillus versicolor* AS-212. This marks the first time that the absolute configurations of all the stereogenic centers in (−)-isoversicomide A (**1**) and cottoquinazoline A (**3**), which were not assigned in the previous literature, were accurately solved in the present work by X-ray crystallographic analysis. Compound **3** exhibited activity against aquatic pathogenic bacteria *A. hydrophilia* with an MIC value of 18.6 μM, while compounds **4** and **8**–**10** exhibited inhibitory effects against *V. harveyi* with MIC values ranging from 15.0 to 28.4 μM. In addition, compounds **4** and **8** exhibited potent inhibitory effects against *V. parahaemolyticus* with MIC values of 9.0 and 15.0 μM, which might have the potential to be developed as leading compounds in discovering aquatic antibiotics.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/md21050293/s1, Figures S1–S27: The analyzed data of MS, 1D and 2D NMR spectra of compounds **1**–**4**, crystal packing of compounds **1** and **3**, HPLC analysis of mycelia extract, broth extract, and compounds **1**–**12** of *Aspergillus versicolor* AS-212, and experimental and calculated ECD spectra of compound **2** at the CAM-B3LYP/TZVP level. Table S1: Crystal data and structure refinement for compounds **1** and **3**. Table S2: Calculated specific rotation values at 589.44 nm for the enantiomers 14*R*-**2** and 14*S*-**2** at the CAM-B3LYP/TZVP level. Table S3: 1H and 13C NMR spectroscopic data for compound **3**.

**Author Contributions:** Y.-L.D. performed the experiments for the extraction, isolation, structure elucidation, and bioactivity evaluation and prepared the manuscript; X.-M.L. performed the 1D and 2D NMR experiments; X.-S.S. and Y.-R.W. contributed to the isolation, identification, and small-scale screening of the fungus AS-212. L.-H.M. contributed to NMR analysis and structure elucidation; B.-G.W. supervised the research work and revised the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was financially supported by the National Natural Science Foundation of China (U2006203 and 41976090), the Senior User Project of RV *KEXUE* (KEXUE2020GZ02), and the Shandong Provincial Natural Science Foundation (ZR2021ZD28 and ZR2019ZD18).

**Data Availability Statement:** Not applicable.

**Acknowledgments:** B.-G.W. acknowledges the support of the Research Vessel *KEXUE* of the National Major Science and Technology Infrastructure from the Chinese Academy of Sciences (for sampling).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

### **Anthraquinone Derivatives and Other Aromatic Compounds from Marine Fungus** *Asteromyces cruciatus* **KMM 4696 and Their Effects against** *Staphylococcus aureus*

**Olesya I. Zhuravleva 1,2,†, Ekaterina A. Chingizova 1,†, Galina K. Oleinikova 1, Sofya S. Starnovskaya 1, Alexandr S. Antonov 1, Natalia N. Kirichuk 1, Alexander S. Menshov 1, Roman S. Popov 1, Natalya Yu. Kim 1, Dmitrii V. Berdyshev 1, Artur R. Chingizov 1, Alexandra S. Kuzmich 1, Irina V. Guzhova 3, Anton N. Yurchenko 1,\* and Ekaterina A. Yurchenko 1,\***


**Abstract:** New anthraquinone derivatives acruciquinones A–C (**1**–**3**), together with ten known metabolites, were isolated from the obligate marine fungus *Asteromyces cruciatus* KMM 4696. Acruciquinone C is the first member of anthraquinone derivatives with a 6/6/5 backbone. The structures of isolated compounds were established based on NMR and MS data. The absolute stereoconfigurations of new acruciquinones A–C were determined using ECD and quantum chemical calculations (TDDFT approach). A plausible biosynthetic pathway of the novel acruciquinone C was proposed. Compounds **1**–**4** and **6**–**13** showed a significant antimicrobial effects against *Staphylococcus aureus* growth, and acruciquinone A (**1**), dendryol B (**4**), coniothyrinone B (**7**), and ω-hydroxypachybasin (**9**) reduced the activity of a key staphylococcal enzyme, sortase A. Moreover, the compounds, excluding **4**, inhibited urease activity. We studied the effects of anthraquinones **1**, **4**, **7**, and **9** and coniothyrinone D (**6**) in an in vitro model of skin infection when HaCaT keratinocytes were cocultivated with *S. aureus*. Anthraquinones significantly reduce the negative impact of *S. aureus* on the viability, migration, and proliferation of infected HaCaT keratinocytes, and acruciquinone A (**1**) revealed the most pronounced effect.

**Keywords:** marine-derived fungus; secondary metabolites; anthraquinones; antibiotics; skin infection; HaCaT; sortase A; urease; migration

#### **1. Introduction**

Anthraquinones are usual metabolites for marine fungi. A recent review by Hafez Ghoran and coauthors described 296 specialized metabolites belonging to the anthraquinone class, which were isolated from 28 marine fungal strains from 2000 to 2021 [1]. They are acetate-derivative metabolites originating from a polyketide containing eight C2 units, which generates, in turn, with three aldol condensations, the carbon skeleton of anthraquinones, except for the two carbonyl oxygens of the central ring. The presence in their structure of many different functional groups makes them very active in interaction with various molecular targets and exhibit wide spectrum of biological activities, including anticancer and antibacterial effects [2].

One of the five main causative agents of nosocomial infections, which are united by the abbreviation ESKAPE, is *Staphylococcus aureus* [3]. A decrease in the protective

**Citation:** Zhuravleva, O.I.; Chingizova, E.A.; Oleinikova, G.K.; Starnovskaya, S.S.; Antonov, A.S.; Kirichuk, N.N.; Menshov, A.S.; Popov, R.S.; Kim, N.Y.; Berdyshev, D.V.; et al. Anthraquinone Derivatives and Other Aromatic Compounds from Marine Fungus *Asteromyces cruciatus* KMM 4696 and Their Effects against *Staphylococcus aureus*. *Mar. Drugs* **2023**, *21*, 431. https://doi.org/10.3390/md21080431

Academic Editor: Hee Jae Shin

Received: 22 June 2023 Revised: 26 July 2023 Accepted: 26 July 2023 Published: 29 July 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

properties of the skin and the body in hospital patients leads to damage to keratinocytes under the influence of *S. aureus* lytic toxins, the destruction of the protective barrier, and the penetration of *S. aureus* into the bloodstream [4]. The global prevalence of bacterial skin diseases in 2019, according to the Global Burden of Disease project, was 14,684.3 cases per 100,000 population [5]. These diseases have rarely been fatal (0.9 cases per 100,000), but the slightest infection can lead to sepsis if the course is unfavorable. There were an estimated 48.9 million cases of sepsis and 11.0 million sepsis-related deaths worldwide in 2017, accounting for 19.7% of all deaths worldwide [6], and the Gram-positive bacterium *S. aureus* is one of the main reasons for this.

Recently, chlorine-containing compounds acrucipentyns A–F were isolated by us from *Asteromyces cruciatus* KMM4696 fungus associated with brown alga *Sargassum pallidum*, and these compounds showed significant antibacterial activity against *Staphylococcus aureus* [7]. The detailed separation of the non-polar part of this fungal extract resulted in the isolation of a number of new and known anthraquinone derivatives. Thus, in this work, we describe the isolation and determination of the structure of these compounds, as well as the study of their antimicrobial properties, including their effects against *Staphylococcus aureus*-infected human HaCaT keratinocytes.

#### **2. Results**

#### *2.1. Isolated Compounds from Asteromyces cruciatus*

As a result of chromatographic separation of the ethyl acetate extract of the culture of the fungus *Asteromyces cruciatus* KMM 4696, new acruciquinones A–C (**1**–**3**), as well as known dendryol B (**4**) [8], pleosporone (**5**) [9], coniothyrinone D (**6**) [10], coniothyrinone B (**7**) [10], rubrumol (**8**) [11], ω-hydroxypachybasin (**9**) [12,13], trans-3,4-dihydroxy-3,4-dihydroanofinic acid (**10**) [14], quadricinctapyran A **(11**) [15], 7-hydroxymethyl-1,2 naphthalenediol (**12**) [16], and gliovictin (**13**) [17] (Figure 1), were isolated. The known compounds (**4**–**13**) were characterized by 1H, 13C NMR, and HR ESI MS data and identified by comparison with literature data.

**Figure 1.** Isolated compounds from *Asteromyces cruciatus*.

#### *2.2. Structural Characterization of New Compounds*

The molecular formula of **1** was determined as C15H16O5 based on the analysis of the (+)-HRESIMS spectrum (Figure S87) containing the peak of the cationized molecule [M + Na]+ (*m*/*z* 299.0887) and was confirmed by the 13C NMR data. In the 1H and 13C NMR spectra of compound **1** (Table 1, Figures S7–S12) there were signals of a tetrasubstituted benzene ring; an olefinic proton; a methyl; a methylene, and four methine groups, three of which were oxygenated; five *sp*2-hybridized quaternary carbons; and one unsaturated keto group.


**Table 1.** 1H NMR spectroscopic data (acetone-d6, *δ* in ppm, *J* in Hz) for **1** and **2**.

<sup>a</sup> Chemical shifts were measured at 500.13 MHz. <sup>b</sup> Chemical shifts were measured at 700.13 MHz.

The HMBC correlations from H-5 to C-7, C-10, C-12, and C-15; from H-7 to C-5, C-8, and C-12; from H3-15 to C-5, C-6, and C-7; and from H-9 to C-8, C-11, C-12, and C-13 (Figures 2a and S11) established the structure of rings A and B and determined the position of the methyl and hydroxyl groups in the tetrasubstituted benzene ring and the hydroxyl and keto groups in ring B. Observed 1H-1H-COSY interactions (H-9/H-13/H-1/H-2/H2-3/ H-4) and HMBC correlations from H-1 to C-2, C-3, C-9, and C-13; from H-3α to C-1, C-2, C-4, and C-14; and from H-4 to C-10 determined the structure of ring C, its fusion with ring B at C-13/C-14, the position of hydroxyl groups at C-1 and C-2, and the Δ4,14 position of the trisubstituted double bond.

**Figure 2.** Key 1H–13C HMBC and 1H-1H-COSY correlations (**a**) and ROESY correlations (**b**) of **1**.

The vicinal coupling constant values (Table 1), as well as the ROESY correlations (Figures 2b and S12) of the H-1/H-3β, H-9, and H-2/H-13 correlations, show that the first three protons in **1** are on the same side of the molecule, while H-2 and H-13 are oriented in the opposite direction.

The molecular formula of compound **2** was determined as C15H16O6 based on the analysis of the (+)-HRESIMS spectrum data containing the peak of the cationized molecule [M + Na]<sup>+</sup> (*m*/*z* 315.0830) and was confirmed by the 13C NMR data. The 1H and 13C NMR spectra of compound **2** (Table 1, Figures S14–S18) were very similar to those for **1**, with the exception of proton and carbon signals at C-1, C-2, C-3, C-4, and C-14 of the cyclohexene ring. Downfield chemical shifts at C-3 and the presence of an additional methine group in **2** instead of a methylene group in 1 suggested the structure of **2** as a 3-hydroxy derivative of **1**. Observed 1H-1H COSY interactions (H-13/H-1/H-2/H-3/H-4) proved the position of the hydroxyl groups in compound **2** at C-1, C-2, and C-3 (Figure S17).

The coupling constant values (Table 1), as well as the ROESY correlations (Figure 3) between H-1, H-3, and H-9 and between H-2 and H-13 showed that the relative structure of **2** was the same as that of **1**.

**Figure 3.** Key ROESY correlations in **2**.

An analysis of the literature data showed that the NMR spectra of compounds **1** and **2** were close to those for the known anthraquinones, dendryols A and D [8]. However, the values of chemical shifts and coupling constants of vicinal protons at C-1 and C-2 in the spectra of **1** and **2** significantly differed from those for known dendryols. Thus, compounds **1** and **2** are new stereoisomers of known dendryols A and D, respectively, and were named acruciquinones A (**1**) and B (**2**), respectively.

The molecular formula of compound **3** was determined as C15H18O5 based on the analysis of the (+)-HRESIMS spectrum data containing the peak of the cationized molecule [M + Na]<sup>+</sup> (*m*/*z* 301.1042) and was confirmed by the 13C NMR data. The 1H and 13C NMR spectra of **3** (Table 2, Figures S19–S24) contain signals of a tetrasubstituted benzene ring; a methyl ring; two methylene groups, one of which is oxygenated; five methine groups, two of which are bonded to oxygen; four quaternary *sp*2-carbons; and one unsaturated ketone group.

HMBC correlations from H-5 (δ<sup>H</sup> 7.25) to C-6 (δ<sup>C</sup> 139.7), C-7 (δ<sup>C</sup> 122.7), C-10 (δ<sup>C</sup> 197.8), C-12 (δ<sup>C</sup> 127.8), and C-15 (δ<sup>C</sup> 21.0); from H-7 (δ<sup>H</sup> 6.86) to C-5 (δ<sup>C</sup> 118.9), C-8 (δ<sup>C</sup> 158.3), C-12, and C-15; from H3-15 (δ<sup>H</sup> 2.31) to C-5, C-6, and C-7; and from H-9 (δ<sup>H</sup> 5.07) to C-8, C-11 (δ<sup>C</sup> 134.7), C-12, C-13 (δ<sup>C</sup> 54.7), and C-14 (δ<sup>C</sup> 51.4) (Figures 4a and S23) establish that the structure of rings A and B are the same as those for compounds **1** and **2**.


**Table 2.** 1H and 13C NMR spectroscopic data (δ in ppm, 700.13 /125.75 MHz, acetone-d6) for **3**.

**Figure 4.** Key 1H–13C HMBC and 1H–1H COSY correlations (**a**) and ROESY correlations (**b**) of **3**.

The observed 1H-1H COSY correlations (H-9/H-13/H-1(H2-2)/H-3/H2-4/H-14) and HMBC correlations from H-1 (δ<sup>H</sup> 2.14) to C-2 (δ<sup>C</sup> 62.9), C-3 (δ<sup>C</sup> 73.1), C-9 (δ<sup>C</sup> 74.5), and C-13; from H-4α (δ<sup>H</sup> 1.98) to C-1, C-3, C-10, and C-14; and from H-14 to C-4, C-10, and C-13 revealed the structure of ring C, its fusion with ring B, and the position of hydroxymethyl and hydroxyl groups at C-1 and C-3, respectively (Figure 4a).

The relative configurations of **3** were established based on the ROESY correlations (Figures 4b and S24): H-1/H-9, H-14, and H-13/H-2a, H-3.

Thus, the structure of compound **3** was determined and named acruciquinone C. It should be noted that acruciquinone C is the first and only representative of anthraquinone derivatives with a 6/6/5 framework.

The absolute configurations of **1**–**3** were determined using an approach based on a comparison of the ECD spectra recorded for these compounds (Figures S2, S4, and S6) with the theoretical spectra calculated for them using the B3LYP exchange–correlation functional and cc-pvTz basis set implemented in GAUSSIAN 16 software (Figure 5) [18].

**Figure 5.** Calculated (red) and experimental (black) CD spectra for compounds **1** (**a**), **2** (**b**), and **3** (**c**). The green color ECD curves were calculated for enantiomers of compounds **1**–**3**.

The best agreement between Δεexp and Δεcalc is achieved for **1**–**3** in the case of configurations 1*S*,2*S*,9*R*,13*S*, 1*S*,2*S*,3*R*,9*R*,13*S*, and 1*R*,3*S*,9*R*,13*S*,14*S*, respectively.

#### *2.3. Bioactivity of Isolated Compounds*

The effects of isolated anthraquinones on *Staphylococcus aureus* growth and the activity of its some enzymes were experimentally investigated. Moreover, the influence of antibacterial compounds on viability, migration, and proliferation of *S. aureus*-treated HaCaT keratinocytes was investigated. Compound **5** was isolated in an insufficient amount (1.0 mg) and was not investigated in any bioactivity tests. Compounds **2**, **3**, and **11** were isolated in very small amounts (0.9 mg, 1.5 mg, and 1.1 mg, respectively), so, only their influence on *S. aureus* growth was investigated.

#### 2.3.1. Antimicrobial Activity

The antimicrobial activity of compounds **1**–**4** and **6**–**13** against *Staphylococcus aureus* is presented in Figure 6.

**Figure 6.** Antimicrobial activity against *Staphylococcus aureus* of compounds **1**–**4** and **6**–**13**. All experiments were carried out in triplicate. The data are presented as a mean ± standard error of mean (SEM).

Acruciquinone B (**2**) did not show any influence on *S. aureus* growth up to a concentration of 100 μM. Acruciquinone A (**1**) inhibited *S. aureus* growth by 38.4 ± 1.5% and 40.5 ± 3.3% at concentrations of 50 and 100 μM, respectively. Compounds **4**, **6**, **7**, and **11** showed inhibition of *S. aureus* growth near 30% at concentrations up to 100 μM.

The half-maximal concentration (IC50) of antistaphylococcal action was calculated for compounds **3**, **8**–**10**, **12**, and **13** (Table 3).

**Table 3.** The calculated half-maximal (IC50) effect of compounds on *S. aureus* growth.


Compounds **8**-**10** showed the best effect on *S. aureus* growth, with calculated IC50 values of 35.4, 45.3, and 49.7 μM, respectively. Compounds **12** and **13** were less effective, with IC50 values of 52.1 and 58.2 μM, respectively. Acruciquinone C (**3**) had an IC50 near 100 μM.

Antimicrobial compounds can influence various aspects of bacterial life, including modification of environmental conditions via urease enzymes [19] or sortase A processing of the bacterial cell wall [20].

We investigated the effect of compounds **1**, **4**, **6**–**10**, **12**, and **13** on the activity of urease and sortase A from *S. aureus* in cell-free assays.

#### 2.3.2. Influence of Some Isolated Compounds on Urease Activity

Compounds **1**, **4**, **6**–**10**, **12**, and **13** were investigated as urease inhibitors, and only dendryol B (4) did not inhibit urease activity (Figure 7). The most significant effect was observed for compounds **8**, **10**, and **13**, which, at 100 μM, decreased the urease activity by 39.2%, 38.5%, and 38.3%, respectively, and, at a concentration of 50 μM, inhibited urease activity by 15.3%, 21.9%, and 21.2%, respectively. Compounds **12** and **9**, at 50 μM, decreased the urease activity by 20.3% and 13.5% and, at 100 μM, decreased urease activity by 31.6% and 21.8%, respectively. New acruciquinone A (**1**) decreased the urease activity by 10.7%

and 14.6% at concentrations of 50 μM and 100 μM, respectively, and compounds **6** and **7** showed similar effects.

**Figure 7.** Effects of compounds **1**, **4**, **6**–**10**, and **12**–**13** on urease activity. Thiourea was used as a control. All experiments were carried out in triplicate. The data are presented as a mean ± standard error of mean (SEM). \* Indicates significant differences between the control (DMSO) and compounds (*p* value ≤ 0.05).

2.3.3. Influence of Some Isolated Compounds on Sortase A Activity

The effects of the investigated compounds on sortase A activity are presented in Figure 8a. Dendryol B (**4**) showed the most significant inhibitory effect on sortase A activity. It inhibited sortase A activity at concentrations of 50 μM and 80 μM by 27.6% and 32.1%, respectively, and its effect was stable during all periods of observation (Figure 8b).

**Figure 8.** The effects of compounds **1**, **4**, **6**–**10**, and **12**–**13** on sortase A activity after 10 min of incubation (**a**) and time-dependent graph of inhibitory effect of dendryol B (**4**) (**b**). 4-(Hydroxymercuri)benzoic acid (PCMB) was used as a control. All experiments were carried out in triplicate. The data are presented as a mean ± standard error of mean (SEM). \* Indicates significant differences between the control (DMSO 0.8%) and compounds (*p* value ≤ 0.05).

Compounds **1**, **7**, and **9** had similar effects on sortase A activity, with significant inhibition of 14.7%, 6.3%, and 14.7%, respectively, at a concentration of 80 μM. The minimal effects of **10** and **13**, as well as those of **6** and **8**, were not statistically significant.

To detect the key structural moieties of anthraquinone derivatives for their inhibitory effect on sortase A, the molecular docking of compounds **1**, **4**, **5**, and **7**–**9** with sortase A was evaluated using fast online service SwissDock.

In the apo structure of sortase A (PDB ID 1T2P), a V-shaped pocket is formed by the β4, β7, and β8 strands on one side of the β barrel, together with three surrounding loops. The left side of the pocket is a hydrophobic tunnel formed by Ala92, Ala104, Ala118, Val161, Pro163, Val166, Val 168, Ile182, Val193, Trp194, Ile199, and Val201, along with two putative catalytic residues: Cys184 and Arg197 [21]. The right side of the pocket consists of several polar residues: Glu105, Asn114, Ser116, and Thr180. Earlier, the anthraquinone dimer skyrin N1287 was found as a sortase A inhibitor, and its complex with sortase A (PDB ID 1T2P) was investigated by molecular docking features. It was reported that skyrin, similar to curcumin, forms a hydrogen-bonding interaction or salt bridge with the guanidinium moiety of Arg197. N1287 and curcumin form extensive interactions with residues in the hydrophobic tunnel. In particular, the aromatic moiety from N1287 forms a cation-π interaction with Arg197. N1287 also forms hydrogen-bonding interactions with polar residues on the right side of the pocket, such as Asn114 and Ser116 [22].

In our calculations, the most active sortase A inhibitor, dendryol B (**4**), can form a pose (ΔG −6.8640547 kkal/mol) with the hydrogen-bonding interactions between Arg197 and its 9-OH, Glu105 and 3-OH, Asn114 and keto-group C-10, and Gly167 and 8-OH. Moreover, hydrophobic interactions between **4** and Val168, Ile199, and Leu169 were detected. In the other side, the stable pose (ΔG −6.800687) forms the hydrogen-binding interaction between Glu105 and 1-OH and the hydrophobic interactions between Cys184 and Me-15; Ala92 and C-7, H-7, Trp194, and Me-15; Ile182 and keto group C-10; and Ala118 and C-15 of **4**.

Acruciquinone A (**1**) can form complex ΔG −6.5611596 with hydrogen-bonding interactions between Arg197 and 2-OH of **1** and Ala92 and 1-OH, as well as with the hydrophobic interactions with Ala92, Gly192 (C-2, H-2), Ala104 (keto-group C-10), Ile182 (C-7 and H-7), Thr93 (1-OH), and Trp194 (H-1). Another pose was calculated (ΔG −6.747401) with the hydrogen-bonding interactions with Arg197 (keto-group C-10) and the hydrophobic interactions with Cys184 (Me-15), Trp194 (Me-15), Ala104 (C-5 and C-6), and Leu169 (2-OH).

Therefore, we can assume that the main differences in the structures of compounds **4** and **1** that influence their complexes with sortase A are the stereochemistry of the 9-OH group: the β orientation of 9-OH provides the opportunity to form a maximum number of interactions if both 9-OH and C(=O)-10 with sortase A (Figure 9).

Coniothyrinone B (**7**) can form complex ΔG −7.046784 with the hydrogen-bonding interactions with Arg197 and Gly192 and the hydrophobic interactions with Ile182, Trp194, Tyr187, Ala104, Gly192, and Thr93. Another complex (ΔG −6.411958) has hydrogenbonding interactions with Glu105 and hydrophobic interactions with Cys184, Trp194, Ala92, Leu97, and Ile182.

ω-Hydroxypachybasin (**9**) can form complex ΔG −6.769604 with the hydrogen-bonding interactions with Arg197, Asn114, and Gly167 and the hydrophobic interactions with Val168, Thr180, Ile199, Val166, Val201, and Gly167. The complex consists of hydrophobic interactions with Cys184 (as well as Ala92, Trp194, Thr93, and Ala104), whereas ΔG -6.2245307 does not have hydrogen-bonding interactions.

Rubrumol (**8**), which did not inhibit sortase A activity, can form complex ΔG −6.9308696 with the hydrogen-bonding interactions with Ala92 and the hydrophobic interactions with Cys184, Ala92, Trp194, and Ile182. Another complex (ΔG −6.237157) has hydrogen-bonding interactions with Glu105 and Ala92 and hydrophobic interactions with Cys184, Ala92, and Ile182.

Therefore, compounds **1**, **4**, **7**, and **9**, which inhibited the activity of sortase A in a SensoLyte 520 Sortase A Activity Assay, may form the interactions with Arg197. No poses with the interactions with Arg197 were predicted for compound **8**. This observation confirms the conclusion about the significance of building with Arg197 for inhibition of sortase A' activity by anthraquinones.

Pleosporone (**5**), which was not investigated in a SensoLyte 520 Sortase A Activity Assay, can form complex ΔG −6.9511595 with the hydrogen-bonding interactions with Arg197 (keto-group C-9), Asn114 (keto-group C-10), and Gly167 (8-OH) and the hydrophobic interactions with Val 166 (H-7), Val168 (aromatic ring A, C-9, C-10), Val201 (Me-15), Ile199 (C-6, C-7), Thr180 (C-5, C-10, C-11), and Gly167 (8-OH). Another calculated complex (ΔG −6.2182164) has hydrogen-bonding interactions with Ala92 (1-OH, 2-OH) and hydrophobic interactions with Cys184 (3-OH), Ile182 (3-OH, H-4b, C-4), Ala92 (1-OH), and Trp194 (2-OH).

A comparison of these poses with complexes of **4** allows us to assume that pleosporone (**5**) may also act as an inhibitor of sortase A activity.

(**a**) (**b**) (**c**)

(**d**) (**e**) (**f**)

**Figure 9. Figure 9**. The molecular docking poses of some anthraquinones with sortase A (PDB ID 1T2P): **1** (**a**), **4** (**b**), **5** (**c**), **7** (**d**), **8** (**e**), and **9** (**f**).

2.3.4. Effects of Compounds on HaCaT Keratinocytes Infected with *Staphylococcus aureus*

Thus, the investigated secondary metabolites of *Asteromyces cruciatus* KMM4696 fungus can inhibit sortase A, especially urease enzyme activities, and affect *S. aureus* growth. However, it is advisable to study the effects of these anthraquinone derivatives in a coculture of *S. aureus* with human cells before confidently talking about their real antibacterial potential. Therefore, the protective influence of compounds **1**, **4**, **7**, and **9** at a concentration of 10 μM on human HaCaT keratinocyte cells infected with *S. aureus* was experimentally investigated. Compound **6** did not show a significant effect on sortase A activity and had

a small influence on urease activity and *S. aureus* growth; therefore, it was selected for in vitro investigation for comparison of its effect with that of **7**.

*S. aureus* produces a number of lysing molecules causing the disruption of mammalian cells, so the release of lactate dehydrogenase (LDH) is used for detection of infected cell viability [23]. The effect of compounds **1**, **4**, **6**–**10**, **12**, and **13** on the LDH release from *S. aureus*-infected HaCaT cells is presented in Figure 10.

**Figure 10.** The effects of compounds **1**, **4**, **6**, **7**, and **9** on LDH release from HaCaT cells after infection with *S. aureus* (Sa) for 48 h. All compounds were used at a concentration of 10 μM. The experiments were carried out in triplicate. \* Indicates statistically significant differences between S. aureus-infected cells and *S. aureus*-infected cells treated with compounds (*p* < 0.05).

The incubation of HaCaT cells with *S. aureus* induced an increase in LDH release of 64.4%. All compounds investigated at a concentration of 10 μM showed significant effects on LDH release from staphylococci-infected HaCaT cells. After 48 h of coincubation, compounds **1**, **4**, **6**, **7**, and **9** caused statistically significant diminishments in LDH release from these cells of 29.4%, 23.8%, 18.3%, 18.4%, and 12.3%, respectively.

The effects of compounds **1**, **4**, **6**, **7**, and **9** on the proliferation of *S. aureus*-infected HaCaT cells were investigated using CFDA SE vital fluorescent dye and the flow cytometry technique described in [24]. The CFDA SE covalent builds with intracellular cytoplasm components, and its quantity (and intensity of fluorescence, respectively) in the cell decreases equivalent to the number of past divisions.

The analysis of obtained flow cytometry data resulted in the detection of four HaCaT cell subpopulations (Figure 11a), and *S. aureus* infection significantly changed the ratio between them (Figure 11b). The percentage of each subpopulation is presented in Table 4.

**Figure 11.** The proliferative profiles of non-treated HaCaT cells (**a**), *S. aureus*-infected HaCaT cells (**b**), and infected cells treated with compounds **1** (**c**), **4** (**d**), **6** (**e**), **7** (**f**), and **9** (**g**). All compounds were used at a concentration of 10 μM. The experiments were carried out in triplicate. The most representative picture for each case is presented.

The most noticeable change as a result of a staphylococcal infection was a change in the ratio between division 1 and division 2, which indicates a slowdown in HaCaT proliferation. Compounds **4**, **6**, and **7** did not show any observed changes in the picture (Figure 11d–f). Compound **9** induced a significant decrease in the amount of the cells in division 1 and an increase in the amount of cells in division 3. The most significant influence on infected HaCaT cells was observed for compound **1** (Figure 11c), which greatly increased the number of the cells in division 3 in comparison with infected and non-infected HaCaT cells.


**Table 4.** The effects of compounds on proliferation of *S. aureus*-infected HaCaT cells.

<sup>1</sup> All compounds were used at a concentration of 10 μM. The experiments were carried out in triplicate, and the percentage of each HaCaT cell subpopulation is presented as mean ± standard error of mean.

Finally, the effects of compounds **1**, **4**, **6**, **7**, and **9** on the migration of *S. aureus*-infected HaCaT cells were investigated (Figure 12). Manufacturing devices from Ibidi®were used for the creation of a cell-free zone in a monolayer of HaCaT cells stained with CFDA SE fluorescent dye, after which the *S. aureus* suspension and compounds were added and the cell migration to this cell-free zone was monitored by a fluorescent microscope for 24 h.

**Figure 12.** The effects of compounds **1**, **4**, **6**, **7**, and **9** on migration of *S. aureus* (Sa)-infected HaCaT cells. All compounds were used at a concentration of 10 μM. The experiments were carried out in triplicate. The most representative picture for each case is presented.

The first differences in cell position were detected after 8 h of observation, and full fusion of the cell-free zone in the non-infected cell layer was observed after 24 h. *S. aureus* infection inhibits fusion of this cell-free area, which was observed after 24 h. All investigated compounds improved migration of the *S. aureus*-infected cells in a cell-free zone. Complete confluence, similar to control cells, was observed for compound **1**, and compounds **6**, **9**, and especially **7** caused almost complete cell overgrowth of the cell-free zone. Compound **4** surprisingly showed the most incomplete fusion of the cell-free zone, but its positive effect was noticeable.

#### **3. Discussion**

#### *3.1. Secondary Metabolites of Asteromyces cruciatus KMM4696*

A biogenesis pathway for the framework of the novel acruciquinone C (**3**) has been proposed (Figure 13). It is obvious that the first steps of acruciquinone C biosynthesis are common to most fungal anthraquinones originating from the octaketide precursor [25]. The dehydration and tautomerization of intermediate *i2* result in anthrone *i3*, which is a plausible direct precursor of compounds **4**, **5**, and **7**–**9**. *i3* can also be sequentially oxidized and reduced to *i4*, from which compounds **1**, **2**, and **6** are most likely formed. Moreover, *i4* probably undergoes several reductions and tautomerizations, which, via diketone *i5*, lead to intermediate *i6* with monoene ring C [26]. Further oxidative cleavage of the double bond and tautomerization lead to *i7*, which, as a result of aldol condensation, turns into a direct precursor of acruciquinone C (**3**) with a 6/6/5 skeleton. Compound 3 is formed as a result of the reduction of aldehyde in *i8*.

**Figure 13.** Plausible biogenetic pathway of acruciquinone C (**3**).

Naphthalene derivative 12 was previously reported only as a synthetic compound [16]. This compound is undoubtedly a cyclization product of the linear hexaketide precursor.

Gliovictin (**13**), a diketopiperazine isolated from terrestrial fungi of the genera *Helminthosporium* and *Penicillium*, has been isolated from culture broths of the marine deuteromycete *Asteromyces cruciatus* [17].

It was previously shown that strain *A. cruciatus* KMM 4696 can produce the first chlorine-contained monocyclic cyclohexanols containing a 3-methylbutenynyl unit that obviously originated from a tetraketide precursor [7]. Benzopyranes **10** and **11** obviously originated from the same precursor. Thus, the *A. cruciatus* KMM 4696 fungal strain is a promising producer of structurally unique polyketides.

#### *3.2. Biological Activity of Isolated Anthraquinone Derivatives*

In our work, known dendryol B (**4**), rubrumol (**8**), trans-3,4-dihydroxy-3,4-dihydroanofinic acid (**10**), quadricinctapyran A (**11**), and gliovictin (**13**) were found as agents against *S. aureus* for the first time.

Dendryol B (**4**) was previously isolated from a weed pathogenic fungus, *Dendryphiella* sp., and caused necrotic events on barnyardgrass leaves [8]. Rubrumol (**8**) was assessed for cytotoxic activities against A549, MDA-MB-231, PANC-1, and HepG2 human cancer cell lines but displayed no significant cytotoxic activities. However, the authors showed the significant effect of **8** on the relaxation activity of topoisomerase I [11]. *Trans*-3,4 dihydroxy-3,4-dihydroanofinic acid (**10**) exhibited potent acetylcholinesterase-inhibitory activity [27]. The antimicrobial activity for quadricinctapyran A (**11**), which was not previously detected up to a concentration of 256 μg/mL [15], but the inhibition of *S. aureus* bacterial growth in microplates was estimated by visual observation only. The activity of gliovictin (**13**) against *Escherichia coli* and *Bacillus megaterium* was not observed [28], but it was tested in agar diffusion assays, which are subject to some limitations. In the present work, the antistaphylococcal activity of the compounds was tested using liquid broth titration with spectrophotometric detection, which can be crucial for detection of the effects of compounds.

Coniothyrinone D (**6**) and coniothyrinone B (**7**) were previously isolated from the culture of an endophytic fungus. *Coniothyrium* sp. They were studied as antimicrobials by the diffusion agar method and, their effects against Gram-positive *B. megaterium* were greater than their effects against Gram-negative *E. coli* [10]. The hydroxylated derivatives of coniothyrinone B (**7**), 8-hydroxyconiothyrinone B, 8,11-dihydroxyconiothyrinone B, 4R,8 dihydroxyconiothyrinone B, and 4S,8-dihydroxyconiothyrinone B, from marine algicolous fungus *Talaromyces islandicus* EN-501 showed pronounced activity against *S. aureus* EMBLC-2 growth [29]. Antistaphylococcal activity of ω-hydroxypachybasin (**9**) was reported when this compound was isolated from the plant *Ceratotheca triloba* [30].

In our work, we not only studied the influence of coniothyrinones B (**6**) and D (**7**) and ω-hydroxypachybasin (**9**) on *S. aureus* growth in detail but also their effects on sortase A and urease activity, as well as their potential for skin infection treatment for the first time.

#### *3.3. Perspectives of Isolated Anthraquinones for the Treatment of Skin Infections*

HaCaT keratinocytes cocultured with *S. aureus* are widely used in vitro models for antibiotic discovery, despite some limitations [23]. Our previously reported results showed that *S. aureus* infection caused HaCaT keratinocyte damage and cell cycle arrest in the G0/G1 phase [31] and resulted in inhibition of cell proliferation and migration, as observed in this work. The studied anthraquinones protect HaCaT cells from *S. aureus*-caused damage because a decrease in the LDH release from treated cells was detected. Moreover, one of the significant anthraquinones changes the proliferation profile and migration of *S. aureus*-infected HaCaT cells.

The various aspects of bacteria's vital activity are the targets for antibiotics. Bactericidal antibiotics were targeted at a diverse set of biomolecules for inhibition to achieve cell death, including DNA topoisomerases (quinolones ciprofloxacin, levofloxacin, and gemifloxacin), RNA polymerase (rifamycin), penicillin-binding proteins, transglycosylases and peptidoglycan building blocks (*β*-lactam penicillins, carbapenems, cephalosporins, glycopeptides, vancomycin, fosfomycin, and daptomycin), and ribosomes (macrolides, lincosamides, streptogramins, and others) [32]. But these strategies have led to the emergence of resistant bacterial strains, which has become one of the major global public health problems [33].

Therefore, new strategies including inhibition of bacterial sortase A or urease activities have led to the discovery of new drugs to which developing resistance will be less possible. The sortase A enzyme was named an "ideal target" for the development of new anti-infective drugs [34] because it plays a significant role in the pathogenesis of Gram-positive bacteria. Sortase A is a bacterial cell membrane enzyme that anchors crucial virulence factors to the cell wall surface [35], and numerous studies have aimed to find new sortase A inhibitors [22,36]. The urease enzyme is able to do so by virtue of its ability to catalyze the conversion of urea into ammonia, thereby allowing bacterial colonies to live in acidic conditions. To date, according to Hameed and coauthors, only one commercial urease inhibitor, Lithostat (acetohydroxamic acid), is available, but it has a number of limitations [37]. Currently, urease inhibitors are considered mainly as potential leaders in urinary tract infections. However, a number of works indicate the promise of this approach for skin staphylococcal infections [19].

Our data point to the great importance of the structure of anthraquinones for the inhibition of sortase A activity. β-Orientation of the 9-OH group in the structure of dendryol B (**4**) makes its interaction with residues in the binding site the most effective.

In the case of urease inhibition, the differences between the action of all the studied anthraquinones are insignificant, which does not allow us to discuss their structure–activity relationship. The highest activity was found for an alkaloid, i.e., gliovictin (**13**). Recently, a large number of sulfur- and nitrogen-containing compounds have been described as urease inhibitors [38]. Obviously, it is precisely the thiodiketopiperazine moiety of gliovictin that makes it interesting for further study against *Helicobacter pylori* and other urease-producing bacteria.

However, the effect on bacterial growth or enzyme activities does not yet mean that substances will be active in real infections, since an infection model is a more complex and multicomponent system. In this regard, the study of the effects of promising compounds in in vitro infection models can lead to unexpected results, as we see here. In our experiments, dendryol B (**4**) exhibited the greatest inhibition of sortase A activity, with a weak effect on *S. aureus* growth, but its effects in coculture experiments were not so great. In contrast, acruciquinone A (**1**) showed a weak (yet noticeable) inhibition of sortase A and urease activity and a moderate effect on *S. aureus*, but this new metabolite from *Asteromyces cruciatus* was the most effective against *S. aureus*-caused HaCaT cell damage and in a skin wound model. ω-Hydroxypachybasin (**9**) exhibited the most significant effect against *S. aureus* growth and a weak inhibition of urease and sortase A activities but showed the least pronounced protection against HaCaT damage, as well as coniothyrinones D (**6**) and B (**7**).

Thus, the protection of *S. aureus*-infected HaCaT keratinocytes by acruciciquinone A (**1**) is due to both its direct antibacterial action and the effect on the keratinocytes themselves.

#### **4. Materials and Methods**

#### *4.1. General Experimental Procedures*

Optical rotations were measured on a Perkin-Elmer 343 polarimeter (Perkin Elmer, Waltham, MA, USA). UV spectra were recorded on a Shimadzu UV-1601PC spectrometer (Shimadzu Corporation, Kyoto, Japan) in methanol. CD spectra were measured with a Chirascan-Plus CD spectrometer (Leatherhead, UK) in methanol. NMR spectra were recorded in CDCl3, acetone-*d6*, and DMSO-*d6* on a Bruker DPX-300 (Bruker BioSpin GmbH, Rheinstetten, Germany), a Bruker Avance III-500 (Bruker BioSpin GmbH, Rheinstetten, Germany), and a Bruker Avance III-700 (Bruker BioSpin GmbH, Rheinstetten, Germany) spectrometer. A calibration of NMR spectra was carried out using the residual solvent signals (7.26/77.16 for CDCl3 and 2.05/29.84 for acetone-d6 according to [39]). HRESIMS spectra were measured on a Maxis impact mass spectrometer (Bruker Daltonics GmbH, Rheinstetten, Germany). Microscopic examination and photography of fungal cultures were performed with an Olympus CX41 microscope equipped with an Olympus SC30 digital camera. Detailed examination of the ornamentation of the fungal conidia was performed using an EVO 40 scanning electron microscope (SEM).

Low-pressure liquid column chromatography was performed using silica gel (50/100 μm, Imid Ltd., Krasnodar, Russia) and Gel ODS-A (12 nm, S—75 um, YMC Co., Ishikawa, Japan). Plates precoated with silica gel (5–17 μm, 4.5 cm × 6.0 cm, Imid Ltd., Russia) and silica gel 60 RP-18 F254S (20 cm × 20 cm, Merck KGaA, Darmstadt, Germany) were used for thin-layer chromatography. Preparative HPLC was carried out on an Agilent 1100 chromatograph (Agilent Technologies, Santa Clara, CA, USA) with an Agilent 1100 refractometer (Agilent Technologies, Santa Clara, CA, USA) and a Shimadzu LC-20 chromatograph (Shimadzu USA Manufacturing, Canby, OR, USA) with a Shimadzu RID-20A refractometer (Shimadzu Corporation, Kyoto, Japan) using YMC ODS-AM (YMC Co., Ishikawa, Japan) (5 μm, 10 mm × 250 mm), YMC ODS-AM (YMC Co., Ishikawa, Japan) (5 μm, 4.6 mm × 250 mm), and Fusion Hydro-RP (Phenomenex, Torrance, CA, USA) (4 μm, 250 mm × 10 mm) columns.

#### *4.2. Fungal Strain*

The strain of the obligate marine fungus *Asteromyces cruciatus* KMM 4696 was isolated from the surface of the thallus of the brown alga *Sargassum pallidum* (Sea of Japan) and identified using morphological and molecular genetic features [7]. The fungal strain is stored in the Collection of Marine Microorganisms (KMM) of PIBOC FEB RAS (Vladivostok, Russia).

#### *4.3. Cultivation of Fungus*

*A. cruciatus* fungus was cultured on a rice medium at 22 ◦C for three weeks in 100 Erlenmeyer flasks (500 mL), each containing 20 g of rice, 20 mg of yeast extract, 10 mg of KH2PO4, and 40 mL of natural seawater from the Marine Experimental Station of PIBOC FEB RAS, Troitsa (Trinity) Bay, Sea of Japan.

#### *4.4. Extraction and Isolation*

At the end of the incubation period, the mycelium of the *Asteromyces cruciatus* KMM 4696 fungus, together with the medium, was homogenized and extracted with EtOAc (2 L). The obtained extract was concentrated to dryness. The dry residue (7.9 g) was dissolved in a H2O−EtOH (4:1) system (200 mL) and extracted successively with *n*-hexane (3 × 0.2 L) and EtOAc (3 × 0.2 L). The ethyl acetate extract was evaporated to dryness (5.3 g) and chromatographed on a silica gel column (3 × 14 cm), which was first eluted with *n*-hexane (200 mL), then with a stepwise gradient of 5% to 50% EtOAc in *n*-hexane (total volume 20 L). Fractions of 250 ml were collected and combined based on TLC data.

The fractions eluted with *n*-hexane-EtOAc (95:5, 80 mg) and *n*-hexane-EtOAc (90:10, 200 mg) were combined and separated on a YMC ODS-A reverse-phase column (1.5 × 5.5 cm), which was eluted with a step gradient from 40% to 80% MeOH in H2O (total volume: 1 L) to afford subfractions I and II. Subfraction I (40% MeOH, 146 mg) was separated by reversephase HPLC on a YMC ODS-A column, eluting first with MeOH–H2O (90:10) to afford two subfractions: I-1 and I-2. Subfraction I-1 was rechromatographed on a YMC ODS-A column eluting with MeOH–H2O (55:45) to **13** (4.8 mg). Subfraction I-2 was rechromatographed on an Ultrasfera Si column eluting with *n*-hexane–ethyl acetate (60:40) to **11** (1.1 mg). Subfraction II (60% MeOH, 110 mg) was separated by reverse-phase HPLC on a YMC ODS-AM column eluting with MeOH–H2O (80:20), then with MeOH–H2O (55:45) to **9** (4 mg) and **12** (21.6 mg).

The fraction of *n*-hexane-EtOAc (80:20, 470 mg) was separated on a Gel ODS-A column (1.5 × 5.5 cm), which was eluted with a step gradient from 40% to 80% MeOH in H2O (total volume 1 L) to afford subfraction III. Subfraction III (40% MeOH, 250 mg) was separated by reverse-phase HPLC on a YMC ODS-A column, eluting with MeOH–H2O (90:10), then with MeOH–H2O (60:40) and MeCN–H2O (60:40) to **1** (4.8 mg), **5** (1.0 mg), **8** (2.2 mg), and **10** (6.4 mg).

The *n*-hexane-EtOAc fraction (70:30, 580 mg) was separated on a column with a reverse-phase sorbent YMC ODS-A (1.5 × 5.5 cm), which was eluted with a step gradient from 40% to 100% MeOH in H2O (total volume 1 L) to subfractions IV and V. Subfraction IV (40% MeOH, 390 mg) was separated by reverse-phase HPLC on a YMC ODS-A column eluting with MeOH–H2O (95:5), then with MeOH–H2O (70:30) to afford compounds **3** (1.5 mg), **6** (4.6 mg), and **7** (3 mg). Subfraction V (100% MeOH, 40 mg) was separated by reverse-phase HPLC on a YMC ODS-A column eluting with MeOH-H2O (55:45), then with CH3CN-H2O (25:75) to **2** (0.9 mg) and **4** (2.7 mg).

#### *4.5. Spectral Data*

Acruciquinone A (**1**): amorphous solids; [α]D<sup>20</sup> −92.0 (*<sup>c</sup>* 0.1 MeOH); UV (MeOH) *λ*max (log *ε*) 335 (3.14), 285 (3.88), 198 (4.24) nm (see Supplementary Figure S1); CD (*c* 9.6 × <sup>10</sup>−4, MeOH), <sup>λ</sup>max (Δε) 202 (−16.07), 232 (0.35), 269 (1.94), 351 (−1.30) nm (see Supplementary Figure S2); for 1H and 13C NMR data, see Table 1 and Supplementary Figures S7–S12; HRESIMS *m*/*z* 275.0914 [M − H]<sup>−</sup> (calcd. for C15H15O5, 275.0925), 299.0887 [M + Na]+ (calcd. for C15H16O5Na, 299.0890) (see Figure S87).

Acruciquinone B (**2**): amorphous solids; [α]D<sup>20</sup> −121.4 (*<sup>c</sup>* 0.07 MeOH); UV (MeOH) *λ*max (log *ε*) 334 ( ), 283 (3.76), 198 (4.09) nm (see Supplementary Figure S3); CD (*<sup>c</sup>* 9.6 × <sup>10</sup><sup>−</sup>4, MeOH), <sup>λ</sup>max (Δε) 203 (−10.73), 256 (−0.56), 288 (−0.21), 366 (−0.91) nm (see Supplementary Figure S4); for 1H and 13C NMR data, see Table 1 and Supplementary Figures S13–S18; HRESIMS *m*/*z* 291.0882 [M − H]<sup>−</sup> (calcd. for C15H15O6, 291.0874), 315.0830 [M + Na]+ (calcd. for C15H16O6Na, 315.0839) (see Figure S88).

Acruciquinone C (**3**): amorphous solids; [α]D<sup>20</sup> −58.0 (*<sup>c</sup>* 0.10 MeOH); UV (MeOH) *λ*max (log *ε*) 315 (3.55), 261 (3.86), 213 (4.96) nm (see Supplementary Figure S5); CD (*c* 9.6 × <sup>10</sup>−4, MeOH), <sup>λ</sup>max (Δε) 215 (−10.42), 257 (−3.62), 306 (6.04), 339 (−0.95) nm (see Supplementary Figure S6); for 1H and 13C NMR data, see Table 2 and Supplementary Figures S19–S24; HRESIMS *m*/*z* 277.1087 [M − H]<sup>−</sup> (calcd. for C15H17O5, 277.1081), 301.1042 [M + Na]<sup>+</sup> (calcd. for C15H18O5Na, 301.1046) (see Figure S89).

#### *4.6. Quantum Chemical Calculations*

The quantum chemical calculations for compounds **1**–**3** in methanol were performed using exchange–correlation functional B3LYP, the polarization continuum model (PCM), and the cc-pvTz basis set implemented in the Gaussian 16 package of programs [18]. Conformations with relative Gibbs free energies in the range of Δ*Gim* ≤ 5 kcal/mol were chosen for calculations of the UV and ECD spectra at the B3LYP/cc-pvTz\_PCM//B3LYP/ccpvTz\_PCM level of theory. The statistical weights of conformations are:

$$g\_{im} = e^{-\frac{\Lambda G\_{im}}{RT}} \Big/ \sum\_{i} e^{-\frac{\Lambda G\_{im}}{RT}}$$

where *T* = 298.15 K, and the subscript "*m*" denotes conformation, for which *G* is minimal.

Each individual transition from electronic ground state to the *i*-th calculated excited electronic state (1 ≤ *i* ≤ 55) was simulated as a Gauss-type function. The bandwidths taken at 1/e peak heights were chosen to be *σ* = 0.34 eV for **1** and **3** and 0.24 eV for **2**. The UV shifts taken for simulations of spectra are Δ*λ* = 0 nm for **1** and **2** and Δ*λ* = −7 nm for **3**.

The scaled theoretical and experimental ECD spectra were obtained according to the following equation:

$$
\Delta \varepsilon\_{\rm scaled}(\lambda) = \Delta \varepsilon(\lambda) \Big/ \left| \Delta \varepsilon \left( \lambda\_{\rm peak} \right) \right| \Big|
$$

where the denominator (|Δ*ε*(*λpeak*)|) is a modulo of the peak value for the chosen characteristic negative band in corresponding ECD spectra (200 ≤ *λpeak* ≤ 220 nm).

#### *4.7. Sortase A Activity Inhibition Assay*

The enzymatic activity of sortase A from *Staphylococcus aureus* was determined using a SensoLyte 520 Sortase A Activity Assay Kit \* Fluorimetric \* (AnaSpec AS-72229, AnaSpec, San Jose, CA, USA) in accordance with the manufacturer's instructions. The compounds were dissolved in DMSO and diluted with reaction buffer to obtain a final concentration of 0.8% DMSO, which did not affect enzyme activity. DMSO at a concentration of 0.8% was used as a control. 4-(Hydroxymercuri)benzoic acid (PCMB) was used as sortase A enzyme activity inhibitor. Fluorescence was measured with a PHERAStar FS plate reader (BMG Labtech, Offenburg, Germany) for 60 min, with a time interval of 5 min. The data were processed with MARS Data Analysis v. 3.01R2 (BMG Labtech, Offenburg, Germany). The results are presented as relative fluorescent units (RFUs) and percentage of the control data and were calculated using STATISTICA 10.0 software.

#### *4.8. Urease Inhibition Assay*

A reaction mixture consisting of 25 μL enzyme solution (urease from *Canavalia ensiformis*, Sigma, 1U final concentration) and 5 μL of test compounds dissolved in water (10–300.0 μM final concentration) was preincubated at 37 ◦C for 60 min in 96-well plates. Then, 55 μL of phosphate buffer solution with 100 μM urea was added to each well and incubated at 37 ◦C for 10 min. The urease-inhibitory activity was estimated by determining ammonia production using the indophenol method. Briefly, 45 μL of phenol reagent (1% *w*/*v* phenol and 0.005% *w*/*v* sodium nitroprusside) and 70 μL of alkali reagent (0.5% *w*/*v* NaOH and 0.1% active chloride NaClO) were added to each well. The absorbance was measured after 50 min at 630 nm using a MultiskanFS microplate reader (Thermo Scientific Inc., Beverly, MA, USA). All reactions were performed in triplicate in a final volume of 200 μL. The pH was maintained at 7.3–7.5 in all assays. DMSO 5% was used as a positive control.

#### *4.9. Antimicrobial Activity*

The bacterial culture of *Staphylococcus aureus* ATCC 21027 (Collection of Marine Microorganisms PIBOC FEBRAS) was cultured in a Petri dish at 37 ◦C for 24 h on solid Mueller Hinton broth medium with agar (16.0 g/L).

The assays were performed in 96-well microplates in appropriate Mueller Hinton broth. Each well contained 90 μL of bacterial suspension (10<sup>9</sup> CFU/mL). Then, 10 μL of a compound diluted at concentrations from 1.5 μM to 100.0 μM using twofold dilution was added (DMSO concentration < 1%). Culture plates were incubated overnight at 37 ◦C, and the OD620 was measured using a MultiskanFS spectrophotometer (Thermo Scientific Inc., Beverly, MA, USA). Gentamicin was used as a positive control at a concentration of 1 mg/mL, and 1% DMSO solution in PBS was used as a negative control. The results were calculated as a percentage of the control data by SigmaPlot 14.0 software.

#### *4.10. Cell Line and Culture Conditions*

The human HaCaT keratinocyte cell line was kindly provided by Prof. N. Fusenig (Cancer Research Centre, Heidelberg, Germany). All cells had a passage number ≤ 30. The cells were incubated in humidified 5% CO2 at 37 ◦C in DMEM medium (BioloT, St. Petersburg, Russia) containing 10% FBS and 1% penicillin/streptomycin (BioloT, St. Petersburg, Russia).

#### *4.11. Cocultivation of HaCaT Cells with S. aureus and Lactate Dehydrigenase Release Test*

HaCaT cells at a concentration of 1.5 × <sup>10</sup><sup>4</sup> cells per well were seeded in 96-well plates for 24 h. Then, a culture medium in each well was changed with *S. aureus* suspension (102 CFU/mL) in full DMEM medium. Fresh DMEM medium without *S. aureus* suspension was added to other wells as needed. The compounds at a concentration of 10 μM were added to wells after 1 h, and HaCaT cells and *S. aureus* were cultured at 37 ◦C in a humidified atmosphere with 5% (*v*/*v*) CO2 for 48 h.

After incubation, the plate was centrifuged at 250× *g* for 10 min, and 50 μL of supernatant from each well was transferred into the corresponding wells of an optically clear 96-well plate. An equal volume of the reaction mixture (50 μL) from an LDH Cytotoxicity Assay Kit (Abcam, Cambridge, UK) was added to each well and incubated for up to 30 min at room temperature. The absorbance of all samples was measured at λ = 450 nm using a Multiskan FC microplate photometer (Thermo Scientific, Waltham, MA, USA) and expressed in optical units (o.u.).

#### *4.12. Migration of HaCaT Cells Cocultivated with S. aureus*

The silicon 2-well inserts (Ibidi®, Gräfelfing, Germany) were placed in the center of wells in a 24-well plate, and HaCaT cell suspension was added to each well for 24 h. After adhesion, the inserts were removed, and the cells were labeled with (5,6)-carboxyfluorescein succinimidyl ester (CFDA SE) dye (LumiTrace CFDA SE kit, Lumiprobe, Moscow, Russia). CFDA SE stock solution at 5 mM in DMSO was dissolved in PBS for preparation of a 10 μM solution. The cell culture medium was replaced with this CFDA SE solution for 5 min at 37 ◦C. Then, the cells were washed twice with PBS, and *S. aureus* suspension (10<sup>2</sup> CFU/mL) in full DMEM medium was added to each well as necessary. The medium without bacteria was added to control wells. The compounds at a concentration of 10 μM were added to wells after 1 h, and HaCaT cells and *S. aureus* were cultured at 37 ◦C in a humidified atmosphere with 5% (*v*/*v*) CO2.

The silicon 2-well inserts from Ibidi®formed cell-free zones and migration of Ha-CaT cells in these zones were observed using an MBF-10 fluorescent microscope (Lomo Microsystems, St.-Peterburg, RF, Russia) during 30 h of incubation.

#### *4.13. Proliferation of HaCaT Cells Cocultivated with S. aureus*

The HaCaT cells at a concentration of 1.5 × <sup>10</sup><sup>4</sup> were seeded in a 12-well plate for 24 h. After adhesion, the cells were strained with (5,6)-carboxyfluorescein succinimidyl ester (CFDA SE) dye (LumiTrace CFDA SE kit, Lumiprobe, Moscow, Russia). CFDA SE stock solution at 5 mM in DMSO was dissolved in PBS for preparation of a 10 μM solution. The cell culture medium was replaced with this CFDA SE solution for 5 min at 37 ◦C. Then, the cell layer was washed with PBS twice, an *S. aureus* suspension (102 CFU/mL) in full DMEM medium was added to each need well, and after 1 h, the compound at a concentration of 10 μM was added to the wells. The medium without bacterial suspension was added to the control well.

After 48 h of incubation, the cells were washed with PBS twice, scrabbed, and collected in 1.5 mL tubes. The intensity of CFDA fluorescence was analyzed with a NovoCyte flow cytometer (Agilent, Austin, TX, USA).

#### *4.14. Molecular Docking*

The pdb file of sortase A (PDB ID 1T2P) was obtained from the RCSB Protein Data Bank (https://www.rcsb.org accessed on 25 July 2023) and prepared for docking by the PrepDock package of UCFS Chimera 1.16 software. The chemical structures of ligands were prepared for docking by ChemOffice and checked by the PrepDock package of UCFS Chimera 1.16 software.

Docking was conducted on the SwissDock online server (http://www.swissdock.ch accessed on 25 July 2023) based on EADock DSS docking software [40]. The algorithm implies the generation of many binding modes in the vicinity of all target cavities (blind docking) and estimation of their CHARMM energies via the Chemistry at HARvard Macromolecular Mechanics (CHARMM) algorithm [41] for evaluation of the binding modes with the most favorable energies with FACTS (Fast Analytical Continuum Treatment of Solvation) [42] and, finally, clustering of these binding modes [43].

The predicted building models for each target/ligand pair were visualized and analyzed by UCFS Chimera 1.16 software. Docking parameters such as Gibb's free energy

(ΔG, kcal/mol), FullFitness score (FF, kcal/mol), and hydrogen-bonding (H-bond) and hydrophobic interactions were used for analysis of target/ligand complexes.

#### *4.15. Statistical Data Evaluation*

All data were obtained in three independent replicates, and calculated values are expressed as a mean ± standard error mean (SEM). Student's *t*-test was performed using SigmaPlot 14.0 (Systat Software Inc., San Jose, CA, USA) to determine statistical significance. Differences were considered statistically significant at *p* < 0.05.

#### **5. Conclusions**

The *Asteromyces cruciatus* KMM4696 fungal strain is a promising producer of structurally unique and antibacterial polyketides. New acruciquinone C (**3**) possessed an unprecedented 6/6/5 anthraquinone-derived skeleton. The effect of new acruciquinone A (**1**) and known dendryol B (**4**) on sortase A activity and their weak antimicrobial effects indicate their potential antivirulence properties, with a reduced risk of antimicrobial resistance, made both these compounds very interesting as antivirulence agents. Their effects against *Staphylococcus aureus* in coculture with human HaCaT keratinocytes conditioned inhibition of sortase A and urease activity but did not limit inhibition, which ensures their positive effect on migration and proliferation of infected keratinocytes.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/ 10.3390/md21080431/s1, Figures S1–S6: ECD and UV spectra of compounds **1**–**3**; Figures S7–S90: NMR spectra of compounds **1**–**13**; Figures S91–S104: HR ESI MS spectra of compounds **1**–**3** and **6**–**13**.

**Author Contributions:** Conceptualization, O.I.Z. and E.A.Y.; Data curation, N.N.K. and A.N.Y.; Formal analysis, O.I.Z., E.A.C. and S.S.S.; Funding acquisition, E.A.C.; Investigation, O.I.Z., E.A.C., G.K.O., A.S.A., N.N.K., A.S.M., R.S.P., N.Y.K., D.V.B., A.R.C., A.S.K. and E.A.Y.; Methodology, E.A.C. and I.V.G.; Resources, O.I.Z., I.V.G. and A.N.Y.; Software, D.V.B.; Supervision, I.V.G.; Validation, A.N.Y.; Visualization, E.A.C., S.S.S., D.V.B., A.S.K. and E.A.Y.; Writing—original draft, O.I.Z., E.A.C. and S.S.S.; Writing—review and editing, A.N.Y. and E.A.Y. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Russian Science Foundation (grant number 23-24-00471).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The original data presented in the study are included in the article/Supplementary Material; further inquiries can be directed to the corresponding author.

**Acknowledgments:** This study was carried out using the equipment of the Collective Facilities Center "The Far Eastern Center for Structural Molecular Research (NMR/MS) PIBOC FEB RAS" and the Collective Facilities Center "Collection of Marine Microorganisms PIBOC FEB RAS".

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

MDPI St. Alban-Anlage 66 4052 Basel Switzerland www.mdpi.com

*Marine Drugs* Editorial Office E-mail: marinedrugs@mdpi.com www.mdpi.com/journal/marinedrugs

Disclaimer/Publisher's Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Academic Open Access Publishing

mdpi.com ISBN 978-3-0365-9087-5