Antiglioma Natural Products from the Marine-Associated Fungus Penicillium sp. ZZ1750

Marine-derived Penicillium fungi are one of the most important sources for the discovery of new bioactive natural products. This study characterized the isolation, structures, and antiglioma activities of twelve compounds, including three novel ones—penipyridinone B (1), 11S-(−)-penilloid A (2), and 11R,14E-(+)-penilloid A (3)—from the marine fungus Penicillium sp. ZZ1750. The structures of the novel compounds were determined via extensive nuclear magnetic resonance (NMR) spectroscopic analyses, high-resolution electrospray ionization mass spectroscopy (HRESIMS) data, Mosher’s method, optical rotation (OR) calculations, and electronic circular dichroism (ECD) calculations. Penipyridinone B represents the first example of its structural type and showed potent antiglioma activity, with IC50 values of 2.45 μM for U87MG cells and 11.40 μM for U251 cells. The known compounds of questiomycin A (9) and xanthocillin X (10) also showed antiproliferative activity against both U87MG and U251 cells, with IC50 values of 13.65 μM to 22.56 μM. The antiglioma activity of questiomycin A and xanthocillin X may be related to the promotion of reactive oxygen species (ROS) production, the reduction of mitochondrial membrane potential (MMP), and the enhancement of caspase-3 enzyme activity.


Introduction
Marine-associated fungi from the genus Penicillium have been proved to be one of the most important resources for the discovery of novel bioactive natural products [1][2][3]. From 1991 to 2014, 390 new compounds with diverse chemical structures were identified from the marine-derived Penicillium fungi, of which 58% showed antitumor, antiviral, antibacterial, and anti-inflammatory activities [1]. An updated catalog of this field from 2015 to 2020 identified 188 secondary metabolites with diverse bioactivities [3]. Recently, more and more novel compounds continued to be discovered from the marine-derived Penicillium fungi, such as poloncosidins A-F from the sea cold-seep-derived fungus Penicillium polonicum CS-252 [4], penitanzacids A-J from the deep-sea-derived fungus Penicillium sp. KWF32 [5], steckwaic acids A-H from the deep-sea coral-derived endozoic Penicillium steckii AS-324 [6,7], dicitrinones G-J from the starfish-derived symbiotic fungus Penicillium sp. GGF16-1-2 [8], oxalierpenes A and B from the marine Penicillium oxalicum HBU-208 [9], and pyrrospirones K-Q from the marine-derived fungus Penicillium sp. SCSIO 41512 [10].
indicated the 8E, 10E and 32E geometries, respectively; the 6E, 21E, 25Z, and 35Z geometries were determined based on the NOE correlations of H-7 with H-9; H-8 with H-10 and H3-16; H2-19 with H-21; H-26 with H3-27; and H-36 with H3-43, respectively. The absolute configuration of C-38 in 1 was established by the Mosher ester NMR method. Treatment of 1 with (R)-α-methoxy-α-(trifluoromethyl) phenylacetyl chloride (R-MTPA-Cl) or S-MTPA-Cl yielded its S-MTPA ester (1s) or R-MTPA ester (1r). The 1 H NMR chemical shift differences (ΔδS-R, Figure 3 and Table S6) between 1s and 1r in negative values for H-28, H-29, and H-39 and positive values for H-35, H-36, and H-43 were observed, indicating a 38S-configuration in 1. Finally, ECD calculations were used to determine the configurations of C-20, C-24, C-29, and C-30. Due to the relationship between the fused positions of the rings C, D, and E in the structure, C-20 and C-24 had only two possible configurations: 20S,24S and 20R,24R. In addition, the absence of NOE correlation of H-29 with H3-40 observed in the NOESY spectrum suggested only two possible configurations: 29S,30S and 29R,30R. Therefore, four model molecules of 2S,3S,4R,5S,20S,24S,29S,30S,38S-1 (1f), 2S,3S,4R,5S,20S,24S,29R,30R,38S-1 (1g), 2S,3S,4R,5S,20R,24R,29S,30S,38S-1 (1h), and 2S, 3S,4R,5S,20R,24R,29R,30R,38S-1 (1i) were applied for ECD calculations. The ECD calculated results ( Figure 4 and Tables S7-S14) showed that the experimental ECD spectrum of 1 was in agreement with the calculated ECD curve of the model molecule of 2S,3S,4R,5S, 20S,24S,29S,30S,38S-1 (1f). Therefore, the signed configurations for the partial structure (1a) were the same as those of restrictinol (1b) [29,30] and penipyridinone A (1e), a reported new compound previously isolated from the same fungal strain Penicillium sp. ZZ1750 [17], while the configurations of the partial structure (1c) were the same as those of the known meroterpenoid neosetophomone (1d) [31]. Based on the foregoing evidence, the structure of 1 was elucidated as a novel compound, named penipyridinone B. The 13 C and 1 H NMR data (Table 1) of penipyridinone B (1) were assigned by a combination of 1 H, 13 C, DEPT, HMQC, HMBC, and NOESY NMR spectroscopic analyses.   Compound 2 was obtained as a yellow amorphous solid and had a molecular formula C17H15N5O2 deduced from its HRESIMS ion peaks at m/z 322.1298 [M + H] and 344.1118 [M + Na], as well as from 13 C NMR data. Careful interpretation of its 13 C and 1 H NMR data (Table 2) as well as the COSY and HMBC correlations ( Figure 5) demonstrated that 2 had the same planar structure as that of a known indole alkaloid penilloid A [33]. However, the configuration at C-11 of penilloid A was not determined, although its positive optical rotation (OR) value (+218.2, c 0.06, MeOH) was reported, which was opposite to the negative OR value (−201.6, c 0.16, MeOH) of 2. Therefore, OR calculations [34,35] were used to assign the configuration at C-11, the only chiral carbon in 2. The OR calculated results (Tables S15-S18) showed a positive OR value (+697.42) for 11R and a  (Table 2) as well as the COSY and HMBC correlations ( Figure 5) demonstrated that 2 had the same planar structure as that of a known indole alkaloid penilloid A [33]. However, the configuration at C-11 of penilloid A was not determined, although its positive optical rotation (OR) value (+218.2, c 0.06, MeOH) was reported, which was opposite to the negative OR value (−201.6, c 0.16, MeOH) of 2. Therefore, OR calculations [34,35] were used to assign the configuration at C-11, the only chiral carbon in 2. The OR calculated results (Tables S15-S18) showed a positive OR value (+697.42) for 11R and a negative OR value (−697.60) for 11S. Accordingly, a 11S-configuration was assigned for 2 because of its negative OR value, and penilloid A should have a 11R-configuration. The 11S-configuration for 2 was further confirmed by ECD calculated results ( Figure 5 and Tables S19-S22), because the experimental ECD spectrum of 2 showed good agreement with the calculated ECD curve of the model molecule of 11S-2. The geometry of the C 14 -C 17 can be assigned based on the chemical shift difference (∆δ C14-C18 ) between C-14 and C-18 [20,33]. Usually, an ∆δ C14-C18 value of over 10 ppm indicated a 14Z-geometry; while an ∆δ C14-C18 value of less 5 ppm was suggestive of an E-geometry. As shown in Table 2, the ∆δ C14-C18 values in MeOH-d 4 and DMSO-d 6 were 13.0 and 11.6 ppm, respectively, indicating a 14Z-geometry for 2. Based on the foregoing evidence, the structure of 2 was elucidated as a new indole diketone piperazine alkaloid, named 11S-(−)-penilloid A.    Figure 5) demonstrated that 2 had the same planar structure as that of a known indole alkaloid penilloid A [33]. However, the configuration at C-11 of penilloid A was not determined, although its positive optical rotation (OR) value (+218.2, c 0.06, MeOH) was reported, which was opposite to the negative OR value (−201.6, c 0.16, MeOH) of 2. Therefore, OR calculations [34,35] were used to assign the configuration at C-11, the only chiral carbon in 2. The OR calculated results (Tables S15-S18) showed a positive OR value (+697.42) for 11R and a negative OR value (−697.60) for 11S. Accordingly, a 11S-configuration was assigned for 2 because of its negative OR value, and penilloid A should have a 11R-configuration. The 11S-configuration for 2 was further confirmed by ECD calculated results ( Figure 5 and Tables S19-S22), because the experimental ECD spectrum of 2 showed good agreement with the calculated ECD curve of the model molecule of 11S-2. The geometry of the C14-C17 can be assigned based on the chemical shift difference (ΔδC14-C18) between C-14 and C-18 [20,33]. Usually, an ΔδC14-C18 value of over 10 ppm indicated a 14Z-geometry; while an ΔδC14-C18 value of less 5 ppm was suggestive of an E-geometry. As shown in Table 2, the ΔδC14-C18 values in MeOH-d4 and DMSO-d6 were 13.0 and 11.6 ppm, respectively, indicating a 14Z-geometry for 2. Based on the foregoing evidence, the structure of 2 was elucidated as a new indole diketone piperazine alkaloid, named 11S-(−)-penilloid A.  Compound 3 had the same molecular formula, C 17 H 15 N 5 O 2 , and very similar UV absorptions as those of 2. Further detailed analyses of the 13 C and 1 H NMR data (Table 3), as well as the HMQC, COSY, and HMBC correlations ( Figure 6) of 3, led to the conclusion that both 3 and 2 were isomers of indole diketone piperazine alkaloids. As described for 2, the chemical shift difference (∆δ C14-C18 ) between C-14 and C-18 can be used to assign the geometry of the C 14 -C 17 double bond in 3 [20,33]. Therefore, the ∆δ C14-C18 values of 0.4 ppm in MeOH-d 4 Table 3, in DMSO-d 6 ) also supported the 14E-geometry for 3. Compared with the negative OR value (−201.6) of 2, compound 3 had a positive OR value (+203.8), suggesting an 11R-configuration for 3, which was the same as that of penilloid A. The 11R-configuration for 3 was further confirmed by the OR calculated results (Tables S23-S26) and the ECD calculated results ( Figure 6 and Tables S27-S30). Therefore, structure of 3 was elucidated as a new indole diketone piperazine alkaloid, named 11R,14E-(+)-penilloid A.

Antiglioma Activity Evaluation
A sulforhodamine B (SRB) assay [36] was applied to evaluate the activity of all of the isolated compounds (1-12) against the proliferation of glioma cells.
Doxorubicin (DOX) was used as a positive control. The results (Table 4) indicated that the

Antiglioma Activity Evaluation
A sulforhodamine B (SRB) assay [36] was applied to evaluate the activity of all of the isolated compounds (1-12) against the proliferation of glioma cells. Doxorubicin (DOX) was used as a positive control. The results (Table 4) indicated that the new penipyridinone B (1) had potent antiproliferative activity, with IC 50 values of 2.45 µM for U87MG cells and 11.40 µM for U251 cells, and the known compounds of questiomycin A (9) and xanthocillin X (10) also showed moderate antiproliferative activity against both U87MG and U251 cells, with IC 50 values of 13.65 µM to 22.56 µM, compared with the control drug DOX, with IC 50 values of 3.76 µM for U87MG cells and 9.85 µM for U251 cells. Other tested compounds were inactive at a concentration of 50 µM. The effects of questiomycin A (9) and xanthocillin X (10) on the reactive oxygen species (ROS) production, the mitochondrial membrane potential (MMP), and the caspase-3 activity in glioma U251 and U87MG cells were further investigated. The results indicated that both questiomycin A (9, 20 µM) and xanthocillin X (10, 20 µM) significantly increased the ROS production (p < 0.001, Figure 7) and reduced the MMP (p < 0.001, Figure 8) in both U251 and U87MG cells after treatment for 48 h, when compared with the blank control (CON). Questiomycin A (9) and xanthocillin X (10) also significantly increased the caspase-3 activity in glioma U251 and U87MG cells (** p < 0.01 or *** p < 0.001, Figure 9) after treatment for 24 h. The enhancement of caspase-3 activity induced by questiomycin A (9) in both U251 and U87MG cells was significantly reduced by the caspase-3 inhibitor Ac-DEVD-CHO ( # p < 0.05 or ### p < 0.001, Figure 9). However, Ac-DEVD-CHO only significantly reduced the enhancement of caspase-3 activity induced by xanthocillin X (10) in U251 cells ( # p < 0.05).  The effects of questiomycin A (9) and xanthocillin X (10) on the reactive oxygen species (ROS) production, the mitochondrial membrane potential (MMP), and the caspase-3 activity in glioma U251 and U87MG cells were further investigated. The results indicated that both questiomycin A (9, 20 μM) and xanthocillin X (10, 20 μM) significantly increased the ROS production (p < 0.001, Figure 7) and reduced the MMP (p < 0.001, Figure  8) in both U251 and U87MG cells after treatment for 48 h, when compared with the blank control (CON). Questiomycin A (9) and xanthocillin X (10) also significantly increased the caspase-3 activity in glioma U251 and U87MG cells (** p < 0.01 or *** p < 0.001, Figure 9) after treatment for 24 h. The enhancement of caspase-3 activity induced by questiomycin A (9) in both U251 and U87MG cells was significantly reduced by the caspase-3 inhibitor Ac-DEVD-CHO ( # p < 0.05 or ### p < 0.001, Figure 9). However, Ac-DEVD-CHO only significantly reduced the enhancement of caspase-3 activity induced by xanthocillin X (10) in U251 cells ( # p < 0.05).    (9) and xanthocillin X (10) on the ROS production in glioma U251 and U87MG cells. Glioma U251 and U87MG cells were treated with 9 (20 µM), 10 (20 µM), and H2O2 (6 µM) at different time points. Data are presented as the mean ± SD, n = 4; *** p < 0.001 (vs. CON).  ROS is known as one of the main upstream effectors in the regulation of apoptosis and excess cellular levels of ROS cause damage to cellular components, which can result in cell death [37]. Mitochondria plays an important role in tumorigenesis and apoptosis. Mitochondrial damage, characterized by the loss of MMP, can cause cells to enter an irreversible process of apoptosis [38]. Caspase-3 is a cysteine-aspartic acid protease and has been identified as an important mediator of apoptosis in cancer; its deficiency may disturb the apoptosis, resulting in carcinogenesis. Thereupon, the enhancement of caspase-3 activity can induce apoptosis of cancer cells, with clinical significance [39]. The current study demonstrated that questiomycin A (9) and xanthocillin X (10) significantly enhanced the ROS production and the caspase-3 activity and decreased the MMP in glioma U251 and U87MG cells. All data, taken together, indicated that the antiglioma activities of both questiomycin A (9) and xanthocillin X (10) were related to the promotion of ROS production, the reduction of MMP, and the enhancement of caspase-3 activity.
ROS is known as one of the main upstream effectors in the regulation of apoptosis and excess cellular levels of ROS cause damage to cellular components, which can result in cell death [37]. Mitochondria plays an important role in tumorigenesis and apoptosis. Mitochondrial damage, characterized by the loss of MMP, can cause cells to enter an irreversible process of apoptosis [38]. Caspase-3 is a cysteine-aspartic acid protease and has been identified as an important mediator of apoptosis in cancer; its deficiency may disturb the apoptosis, resulting in carcinogenesis. Thereupon, the enhancement of caspase-3 activity can induce apoptosis of cancer cells, with clinical significance [39]. The current study demonstrated that questiomycin A (9) and xanthocillin X (10) significantly enhanced the ROS production and the caspase-3 activity and decreased the MMP in glioma U251 and U87MG cells. All data, taken together, indicated that the antiglioma activities of both questiomycin A (9) and xanthocillin X (10) were related to the promotion of ROS production, the reduction of MMP, and the enhancement of caspase-3 activity.

Isolation and Taxonomic Identity of Penicillium sp. ZZ1750
The marine-derived fungus Penicillium sp. ZZ1750 was isolated from a sample of marine mud, which was collected from the Arabian Sea close to Karachi, Sindh, Pakistan, in January 2019. The detailed isolation and taxonomic identity (Table S1 and Figure S1) of Penicillium sp. ZZ1750 were reported in a previous publication [17].

Large Culture of Strain ZZ1750
The large culture of the strain ZZ1750 in rice medium for 30 days was described in a previous publication [17]. A scale-up culture of the strain ZZ1750 was also conducted in GA liquid medium ( Figure S2). Briefly, the pure colony of strain ZZ1750 from the PDA medium slant was inoculated into 500 mL Erlenmeyer flasks, each containing 200 mL of GA liquid medium, then incubated at 28 • C for 4 days on a rotary shaker (180 rpm) to produce seed broth. The seed broth (5 mL) was transferred into 200 mL GA liquid medium in a 500 mL Erlenmeyer flask, then incubated in static state at room temperature for 30 days. A total of 60 L culture (300 flasks) was prepared for this study.

Extraction and Isolation of Compounds 1-12
Compound 1 was isolated from the culture [17] of the strain ZZ1750 in rice medium. Briefly, the culture of the strain ZZ1750 in rice medium in each flask was extracted with EtOAc (250 mL) three times. The combined EtOAc extract was dried in vacuo to give an extract (100 g). This extract was fractionated on a column of silica gel (1600 g) eluting with a mixture of cyclohexane and EtOAc in different ratios (10:1, 5:1, 2:1, 1:1, and 1:2, each 1000 mL) to give four fractions of Frs.A-D based on the results of TLC analyses. Then, Fr. B was separated on a column of ODS (200 g) successively eluting with 60%, 70%, 80%, and 90% MeOH (each 1000 mL) to yield four subfractions of SFrs.B 1 -B 4 . SFr.B 4 was further separated by HPLC using a Zorbax SB-C 18 column (250 × 9.4 mm, 5 µm; mobile phase: MeOH/H 2 O, 87/13; flow rate: 1.0 mL/min; UV detection: 275 nm) to yield compounds 1e (3.8 mg, t R 41.0 min) and 1 (2.8 mg, t R 48.0 min).
Compounds 2-12 were isolated from the culture of the strain ZZ1750 in GA liquid medium. The 60 L culture of the strain ZZ1750 in GA liquid medium was centrifuged to give filtrate and mycelia. The mycelia were extracted with EtOAc three times to yield an EtOAc extract after removing the organic solvent. The filtrate was applied to a HP-20 column eluting with water, then 100% MeOH. The collected MeOH fraction was dried in vacuo to obtain a MeOH extract. The EtOAc extract and MeOH extract were combined (13.

Esterification of Penipyridinone B (1)
Penipyridinone B (1, 1.0 mg) and dimethylaminopyridine (1.1 mg) were dissolved in 0.5 mL anhydrous pyridine; then, either R-MTPA-Cl or S-MTPA-Cl (50 µL) was added. The mixture was stirred at room temperature for 48 h until 1 mL MeOH was added to terminate the reaction. The reaction mixture was dried under reduced pressure to provide a residue. This residue was separated by the Zorbax SB-C 18  the OD value of the blank control (0.3% DMSO in water). IC 50 values were calculated by using GraphPad software and presented as the mean ± SD (n = 5).

Reactive Oxygen Species (ROS) Measurement
The ROS level was measured by using the DCFH-DA kit (Yeasen Biotechnology, Shanghai, China) according to the manufacturer's instruction. Briefly, glioma U251 and U87MG cells in logarithmic growth phase were cultured in 96-well plates with a density of 3 × 10 3 cells per well for 24 h, then treated with tested compounds, positive control H 2 O 2 , or blank control DMSO. After the treatment of the set time points, the treated cells were exposed to a 50 µL DCFH-DA solution (10 µM) for 30 min. Excess DCFH-DA was removed by washing the cells twice with phosphate buffered saline (PBS). Fluorescence was immediately measured in a microplate reader (Synergy 2, Bio Tek) with an excitation/emission wavelength of 485/525 nm.

Mitochondrial Membrane Potential (MMP) Measurement
The MMP was determined using the specific MMP fluorescent probe JC-1 kit (Beyotime Biotechnology, Shanghai, China). Briefly, glioma U251 and U87MG cells (3 × 10 3 cells per well) in logarithmic growth phase were cultured in 96-well plates for 24 h, then treated with tested compounds, positive control bortezomib (BTZ), or blank control DMSO at the set time points. The treated cells were incubated for 20 min at 37 • C with a 50 µL JC-1 working solution, then washed twice with PBS. The stained cells were measured in a microplate reader (excitation/emission wavelength 485/528 nm for green and 500/590 nm for red). The MMP was indicated by the fluorescent ratio of red/green.

Caspase-3 Activity Determination
The Caspase-3 activity was determined by using the GreenNuc™ Caspase-3 Substrate kit (Beyotime Biotechnology). Briefly, glioma U251 and U87MG cells (3 × 10 3 cells per well) in logarithmic growth phase were cultured in 96-well black plates for 24 h, then treated with tested compounds or blank control DMSO for 48 h. The treated cells were incubated for 30 min at room temperature in dark with a 100 µL GreenNuc™ Caspase-3 Substrate solution (5 µM), then washed twice with PBS. The stained cells were measured in a microplate reader (excitation/emission wavelength 485/515).

Statistical Analysis
All data are presented as the mean ± SD. GraphPad Prism 7.0 was used for statistical analysis. Comparisons between two groups were carried out with a two-tailed Student's t-test. Variances among more than two groups were analyzed with one-way ANOVA. The p value < 0.05 was considered to indicate statistical significance.
Compounds 2-7 were six indole alkaloids, as the major metabolites of strain ZZ1750 in GA liquid medium, and showed no antiglioma activity. It is well known that marinederived indole alkaloids possess not only intriguing structures, but also diverse biological activities [40]. Therefore, other activities of these isolated indole alkaloids, especially the new 11S-(−)-penilloid A (2) and 11R,14E-(+)-penilloid A (3), need to be further evaluated.
The data from our previous publications and current study indicated that the fungal strain ZZ1750 produced different structural types of secondary metabolites in different culture conditions. The rare glycosylated alkylresorcinols of peniresorcinosides A-E were the only metabolites found in the rice solid medium for a 30 day culture and the polyhydroxanthones of ergochromes D-G were only found in the rice solid medium for a 90 day culture. In the GA liquid medium, the strain ZZ1750 produced the indole alkaloids, but no glycosylated alkylresorcinols or polyhydroxanthones.