New Bioactive Polyketides from the Mangrove-Derived Fungus Penicillium sp. SCSIO 41411

Abstract

Three new polyketides, including three ester derivatives (1, 3, and 5) and a new natural product, which was a benzoquinone derivative, embelin A (4), together with nine known ones (2 and 6–13), were isolated from the mangrove-derived fungus Penicillium sp. SCSIO 41411. Their structures were determined by detailed NMR and MS spectroscopic analyses. The X-ray single-crystal diffraction analysis of 4 was described for the first time. Compound 9 displayed obvious inhibition against PDE4 with an inhibitory ratio of 40.78% at 10 μM. Compound 12 showed DPPH radical scavenging activity, with an EC₅₀ of 16.21 µg/mL, compared to the positive control (ascorbic acid, EC₅₀, 11.22 µg/mL). Furthermore, compound 4 exhibited cytotoxicity against PC-3 and LNCaP with IC₅₀ values of 18.69 and 31.62 µM, respectively.

1. Introduction

Mangroves are a unique intertidal ecosystem located in tropical and subtropical regions, with nearly 60–70% of the world’s tropical and subtropical coastlines covered by mangroves, widely regarded as one of the most productive ecosystems [1,2]. One of the key populations in this ecosystem is microbial diversity [2]. Fungi and bacteria make up 91% of the microbial biomass in tropical mangroves, which produce abundant marine natural products. As of December 2020, scientists have discovered at least 1387 new structures from mangrove-derived fungi, numerous of which demonstrated a variety of pharmacological activities [1,2,3,4].

Penicillium sp., as a representative of marine fungi, generates a wide range of bioactive secondary metabolites, mainly including polyketides, alkaloids, peptides, etc. [5,6,7,8]. Auroglaucin, isolated from an aciduric fungus strain, Penicillium oxalicum OUCMDZ-5207, exhibited strong selective inhibition on A549 cells, with an IC₅₀ value of 5.67 μM [9]. Aspterric acid, which was generated from the fungus Penicillium polonicum H175, displayed a noteworthy hypoglycemic impact comparable to that of the positive drug rosiglitazone (RSG) at 10 μM [10].

In our study, three new ester derivatives (1, 3, and 5) and one new natural product (4), together with nine known compounds (2 and 6–13) (Figure 1), were isolated from a mangrove sediment-derived fungus, Penicillium sp. SCSIO 41411. Herein, the specifics of the isolation, structural elucidation, and bioactive assessments of isolated compounds were reported.

2. Results and Discussion

2.1. Structural Determination

Compound 1 was isolated as a brown oil, and its molecular formula of C₁₁H₁₂O₆ was determined by HRESIMS data at m/z 241.0709 [M+H]⁺. The 1D NMR (

Table 1. The ¹H and ¹³C NMR data of compounds 1 and 5.

Pos.	1 a		5 b
	δC Type	δH (J in Hz)	δC Type	δH (J in Hz)
1	168.3, C		173.3, C
2			67.0, CH	5.27, s
3	81.5, CH	5.26, br s	126.6, C
4	134.3, C		146.9, C
5	157.3, C		115.8, CH	6.60, d (8.5)
6	104.2, CH	6.41, s	115.2, CH	6.50, dd (8.5, 3.0)
7	153.7, C		149.7, C
8	103.4, C		114.2, CH	6.67, d (3.0)
9	141.5, C		51.5, CH3	3.57, s
10	59.7, CH3	3.69, s
1′	65.1, CH	4.18, q (5.9)
2′	20.6, CH3	1.28, d (6.5)

^{a 1}H (500 MHz) and ¹³C (125 MHz) measured in DMSO-d₆. ^{b 1}H (700 MHz) and ¹³C (175 MHz) measured in DMSO-d₆.

) and HSQC spectra of 1 showed signals of one carbonyl carbon (δ_C 168.3), five unsaturated carbon signals (δ_C 157.3, 153.7, 134.3, 141.5, 103.4), one aromatic methine (δ_H/C 6.41/104.2), two oxygen-containing saturated methylenes (δ_H/C 5.26/81.5 and 4.18/65.1), and two methyls (δ_H/C 3.69/59.7 and 1.28/20.6) (one of them was methoxyl). These aforementioned data were similar to the 1D NMR data of the known compound embeurekol C (2), a polyketide derivative obtained from the fungus Embellisia eureka [11]. HMBC and COSY correlations (Figure 2) confirmed that the planar structures of 1 and 2 were consistent. To determine the absolute configuration of the secondary alcohol at C-1′ in 1 and 2, we performed the Mosher ester analysis [12]. The (R)-MPA and (S)-MPA esters were prepared. Analysis of ¹H NMR data yielded Δδ_RS values (δ_R–δ_S), confirming that 1 and 2 had the same S configuration at C-1′ (Figure 3). Therefore, two theoretical configurations containing (3S, 1′S)-1 and (3R, 1′S)-1 were speculated for ECD calculation. Combined with the trend of the experimental curve of 1, the absolute configuration of 1 was ultimately inferred to be 3S, 1′S (Figure 4). However, the experimental ECD spectra of 2 showed opposite trends with 1, so it was speculated that the absolute configuration of C-3 in 2 was opposite to that of 1, which was 3R. The specific optical rotation of 1 (${\left[\alpha \right]}_{\mathrm{D}}^{25}$ + 49.0 (c 0.1, CH₃OH)) and embeurekol C (2) (${\left[\alpha \right]}_{{20}_{\mathrm{D}}}$ − 17.0 (c 0.05, CH₃OH)) [11] further confirmed their opposite configurations at C-3. Consequently, compound 2 was identical to embeurekol C, and 1 was named embeurekol D.

Compound 3 was obtained as a red-brown solid. The molecular formula was determined as C₁₄H₂₀O₅ by HRESIMS data at m/z 269.1389 [M+H]⁺. The one dimensional (1D) NMR (

Table 2. The ¹H and ¹³C NMR data of compounds 3 and 4.

Pos.	3 a		4 b
	δC Type	δH (J in Hz)	δC Type	δH (J in Hz)
1	172.7, C		183.6, C
2	144.2, C		161.9, C
3	115.6, CH	6.96, s	104.0, CH	5.88, s
4	160.0, C		184.5, C
5	150.2, C		156.1, C
6			120.1, C
7	153.0, C		23.3, CH2	2.40, t (7.7)
8	27.5, CH2	2.68, t (7.5)	29.2, CH2	1.43, m
9	26.1, CH2	1.61, m	30.2, CH2	1.32, m
10	28.3, CH2	1.30, m	30.6, CH2	1.29, m
11	28.4, CH2	1.30, m	33.0, CH2	1.29, m
12	31.1, CH2	1.24, m	23.7, CH2	1.32, m
13	22.0, CH2	1.24, m	14.4, CH3	0.90, t (7.1)
14	13.9, CH3	0.85, t (6.8)	57.2, CH3	3.83, s
15	53.3, CH3	3.87, s

^{a 1}H (500 MHz) and ¹³C (125 MHz) measured in DMSO-d₆. ^{b 1}H (700 MHz) and ¹³C (175 MHz) measured in CD₃OD.

) and HSQC spectrum of 3 showed signals of two carbonyls (δ_C 172.7, 150.2), three nonprotonated sp² carbons (δ_C 160.0, 153.0, 144.2), one alkene methine (δ_H/C 6.96/115.6), six methylenes (δ_H/C 1.24/31.1, 1.30/28.4, 1.30/28.3, 2.68/27.5, 1.61/26.1, 1.24/22.0), and two methyls (δ_H/C 3.87/53.3, 0.85/13.9) (one of them was methoxyl). According to the HMBC correlation (Figure 2) of H-3/C-1, C-2, C-4, and C-5, compound 3 was an unsaturated pentolactone derivative with a carboxyl group connected to C-2. The HMBC correlations of H₃-15/C-4 revealed that 15-OCH₃ was located at C-4. According to the molecular formula, the COSY-related signal indicated a spin-coupled system H₃-14/H₂-13/H₂-12/H₂-11/H₂-10/H₂-9/H₂-8, indicating the presence of heptane. The HMBC correlations of H₂-8/C-7, 2, and H₂-9/C-7 revealed that CH₂-8 was located at C-7. Compound 3 was unambiguously characterized, as shown in Figure 1. The structure of 3 was similar to that of dothydeopyron B [13], with the main difference being that 3 had a carboxyl group while dothydeopyron B had a methoxy group, and there was a hydroxyl group substituted at the branch. As a newly discovered natural product, the novelty of compound 3 lay in its presence of an aliphatic chain, which made it potentially more hydrophilic compared to compounds with longer chains. Finally, compound 3 was named 7-heptyl-4-methoxy-6-oxo-3H-pyran-2-carboxylic acid.

Compound 4 was obtained as a yellow solid and was determined to have the molecular formula C₁₄H₂₀O₄ from the HRESIMS data at m/z 253.1442 [M+H]⁺. The 1D NMR (Table 2) and HSQC spectra of 4 showed signals of two carbonyl carbons (δ_C 184.5, 183.6), three alkene carbons (δ_C 161.9, 156.1, 120.1), one alkene methine (δ_H/C 5.88/104.0), six methylenes carbons (δ_H/C 1.29/33.0, 1.29/30.6, 1.32/30.2, 1.43/29.2, 1.32/23.7, 2.40/23.3), and two methyls (δ_H/C 3.83/57.2, 0.90/14.4) (one of them was methoxyl). The above data suggested that 4 had a benzoquinone skeleton and a fatty acid chain. The COSY correlations (Figure 2) of H₂-9/H₂-8/H₂-7 and H₂-12/H₃-13 and the HMBC correlations of H₃-13/CH₂-11 revealed that C-7–C-13 was a fatty acid chain. The HMBC correlations of H₂-7/C-1, C-5, C-6, and H₂-8/C-6 revealed that CH₂-7 was located at C-6. The HMBC correlations of H₃-14/C-2 indicated that the 14-OCH₃ was located at C-2. Furthermore, the X-ray crystal structure of 4 (CCDC 2363995, Figure 5) obtained by slow evaporation in CH₃OH at room temperature further confirmed the above elucidation of the planar structure. Compound 4 has been previously identified as a synthetic product, which was synthesized to illustrate the structure-activity relationship against α-glucosidase of the embelin derivatives [14]. Here we reported for the first time that this compound has been explored from nature and confirmed as a new natural product. Therefore, the compound was named embelin A.

Compound 5 was isolated as a brown oil and had the molecular formula C₉H₁₀O₅ as determined by HRESIMS data at m/z 221.0429 [M+Na]⁺. The ¹H NMR of compound 5 shows a typical 1,2,4-trisubstituted benzene ring [δ_H 6.67 (1H, d, J = 3.0 Hz); 6.50 (1H, dd, J = 8.5, 3.0 Hz); 6.60 (1H, d, J = 8.5 Hz)]. The one-dimensional (1D) NMR (Table 1) and HSQC spectra of 5 showed signals of one carbonyl carbon (δ_C 173.3), three aromatic carbons (δ_C 149.7, 146.9, 126.6), three aromatic methines (δ_H/C 6.60/115.8, 6.50/115.2, and 6.67/114.2), one methine connected to hydroxyl group (δ_H/C 5.27/67.0), and one oxygenated methyl (δ_H/C 3.57/51.5). The HMBC correlations (Figure 2) of H₃-9/C-1 revealed that the oxygenated methyl group 9-OCH₃ was located at C-1. The HMBC correlations of H-2/C-1, 3, 4, and 8 indicated that C-2 was next to C-1 and C-3. Moreover, it indicated the location of the phenolic hydroxyl group (C-4) and the aromatic methine (C-8). The HMBC correlations of H-8/C-4, 6, and H-5/C-3, 7, and the COSY correlations of H-5/H-6 revealed the location of C-5, C-6, and C-7. The optical rotation of 5 was almost zero, suggesting that it was a racemate. Finally, compound 5 was named methyl α,2,5-trihydroxybenzeneacetate.

Meanwhile, the other eight known compounds were identified as protocatechuic acid (6) [15], 5-[(3Z,5E)-3,5-nonadienyl]-1,3-benzenediol (7) [16], 2,4-Dihydroxy-6-(3E,5E)-3,5-nonadien-1-ylbenzoic acid (8) [17,18], 3,5-Dimethoxy-4-(1-methylethyl)[1,1′-biphenyl]-2,4′-diol (9) [19], 3-Methyl-6,8-dihydroxyisocoumarin (10) [20], 6,8-dihydroxy-5-methoxy-3-methyl-1H-isochromen-1-one (11) [21], butyrolactone I (12) [22], polybotrin (13) [23], respectively, by comparing their NMR data (Supplementary Information) to previous reports. According to the literature, 11 exhibited a-glucosidase inhibitory activity with an IC₅₀ value of 89.4 μM [24], and 12 had an acetylcholinesterase inhibiting effect [25].

As we can see, compounds 1/2 and 10/11 had similar parent nucleus structures, with the main difference being that 1/2 were five-membered lactone rings, while compounds 10/11 were six-membered lactone rings. The example of the development of furan and pyran rings in monoterpenoids illustrated the mechanism of biosynthesis. Aspergillus niger could be used to biotransformation linalool into both furanoid and pyranoid linalool oxide simultaneously [26]. The phenomena suggested that these two types of compounds had a certain biological relationship in the isolated strain.

2.2. Bioactivity Assay

The isolated compounds were evaluated for their antibacterial and cytotoxic activities (

Table 3. Antibacterial and cytotoxicity activities of compounds.

Compounds	Antibacterial (MIC, µg/mL)		Cytotoxicity (IC50, µM)
	S. aureus	S. suis	PC-3	LNCaP	DU145
1	- a	-	-	-	-
3	-	100	-	-	-
4	-	-	18.69	31.62	>100
7	100	100	-	-	-
8	-	100	-	-	-
9	100	100	-	-	-
12	-	-	-	-	-
Positive	25 b	50 c	0.12 d	0.0018 d	0.0033 d

^a “-” means no activity; ^b Streptomycin; ^c Penicillin; ^d Docetaxel.

). Compounds 7 and 9 exhibited activities against Staphylococcus aureus, with the MIC values of 100 µg/mL. Compounds 3, 7, 8, and 9 exhibited activities against Streptococcus suis, with MIC values of 100 µg/mL. In addition, compound 12 had a DPPH inhibitory activity with an EC₅₀ of 16.21 µg/mL, compared to the positive control (ascorbic acid, with an EC₅₀ of 11.22 µg/mL). Phosphodiesterase 4 (PDE4) is an important target for the treatment of inflammation [27]. In screening for PDE4 inhibitory activity, compounds 1–3 and 9–13 at 10 µM displayed inhibition against PDE4 with inhibitory ratios of 18.62%, 14.95%, 19.67%, 40.78%, 27.42%, 27.39%, 29.10%, and 26.22%, respectively. Among these, compound 9 exhibited moderate inhibitory activity against PDE4.

Compounds 1–4 and 7–13 were tested at 10 µM in the cytotoxicity assay on two human prostate cancer cell lines, PC-3 and 22Rv1 (Figure 6). Compound 4 exhibited significant inhibition on PC-3, with an IC₅₀ value of 18.69 µM. Furthermore, Compound 4 was evaluated against two additional prostate cancer cell lines, LNCaP and DU145. It effectively inhibited the LNCaP cell line with an IC₅₀ value of 31.62 μM, while its IC₅₀ value against the DU145 cell line was greater than 100 μM. The above results indicated that the cytotoxicity activity of 4 was selective, with the strongest activity against the PC-3 cell line. Among these three common prostate cancer cell lines, PC-3 was androgen-independent and had moderate metastatic potential [28]. Therefore, we speculated that compound 4 inhibited the mid-term stage of prostate cancer cell development. As a new natural product, 4 had a similar skeleton structure to auroglaucin-related analogs [29], both of which had a six-membered ring structure and a long chain. However, compounds 3 and 4 differed mostly in the characteristics of their six-membered rings. The cytotoxicity of 4 indicated that lengthy fatty chains had a minimal impact on the cytotoxicity of this kind of structure.

AKT1 (NP_005154.2) was involved in the growth, survival, and metabolism of cells [30]. Studies have shown that AKT1 protein could interact with UHRF1, but inhibited AKT1-induced phosphorylation of UHRF1 could induce degradation of UHRF1 protein, thereby reducing the interaction between UHRF1 and deubiquitinase USP7 and promoting its interaction with E3 ubiquitin-protein ligase BTRC, which is of great significance for the treatment of prostate cancer [31]. Therefore, AKT1 was selected as the docking target for further research. Compound 4 demonstrated a perfect interaction with the AKT1 protein (PDB ID: 3O96) with a docking score of −5.886. As shown in Figure 7, hydroxyl and ketone carbonyl groups of 4 formed two hydrogen bonds with the active site residues SER 205 and LYS 268.

3. Materials and Methods

3.1. General Experimental Procedures

Optical rotations were measured with an Anton Paar MPC500 (Anton, Graz, Austria) polarimeter. The UV spectra were recorded on a Shimadzu UV-2600 PC spectrometer (Shimadzu, Beijing, China), while the IR spectra were determined by an IR Affinity-1 spectrometer (Shimadzu). The ECD spectra were performed on a Chirascan circular dichroism spectrometer (Applied Photophysics, Leatherhead Surrey, UK). High-resolution electrospray ionization mass spectroscopy (HRESIMS) spectra were obtained on a Bruker maXis Q-TOF mass spectrometer (Bruker BioSpin International AG, Fällanden, Switzerland). The NMR spectra were collected on a Quantum-I Plus 500 MHz (Q-one Instrument Co., Ltd., Wuhan, China) operating at 500 MHz for ¹H NMR and 125 MHz for ¹³C NMR and were collected on an AVANCE III HD 700 MHz (Bruker Switzerland AG, Fällanden, Switzerland) operating at 700 MHz for ¹H NMR and 175 MHz for ¹³C NMR, which used tetramethylsilane as an internal standard. Semipreparative high-performance liquid chromatography (HPLC) was performed on the Hitachi Primaide with a DAD detector (Hitachi, Tokyo, Japan), using ODS columns (ChromCore 120 C18, 10 × 250 mm, 5 mm; YMC-pack ODS-A, 10 × 250 mm, 5 mm; COSMOSIL πNAP 10 × 250 mm; COSMOSIL 5C18-AR-II 10 × 250 mm). Column chromatography was detected by silica gel (200–300 mesh), and spots were detected on TLC (Qingdao Marine Chemical Factory, Qingdao, China) under 254 nm UV light, respectively. All solvents used were provided by Tianjin Fuyu Chemical and Industry Factory, Tianjin, China, and were of analytical grade.

3.2. Fungal Material

The strain Penicillium sp. SCSIO 41411 was isolated from the rhizosphere sediment sample of the mangrove Aegiceras corniculatum in Gaoqiao Mangrove, Zhanjiang. It was stored in the CAS Key Laboratory of Tropical Marine Bioresources and Ecology, South China Sea Institute of Oceanology, Chinese Academy of Sciences, Guangzhou, China. The strain was designated Penicillium sp. SCSIO 41411 is based on BLAST analysis of the ITS sequence (Supplementary Information) because it shared 99% of the similarities with Penicillium brefeldianum (NR_138263.1). Finally, the sequence was deposited in GenBank with the accession number OQ052995.

3.3. Fermentation and Extraction

The fungal strain Penicillium sp. SCSIO 41411 was statically cultivated on MA medium and then was cultured in 200 mL seed medium (1.5% malt extract, 2.0% sea salt) in 1 L Erlenmeyer flasks at 28 °C for 3 days on a rotary shaker (180 rpm). A large-scale fermentation was incubated at 26 °C for 28 days using a rice medium (200 g rice, 2% sea salt, 230 mL H₂O) in the 1 L flask (×80) under static conditions. The whole fermented culture was extracted with EtOAc three times to afford a brown extract (150 g).

3.4. Isolation and Purification

The whole ethyl acetate extract was subjected to a silica gel vacuum liquid chromatography using a step gradient elution of petroleum ether (PE)-dichloromethane (DCM) (ν:ν 1:0, 1:1, 0:1), DCM-methyl alcohol (CH₃OH) (ν:ν 100:1, 100:3, 50:3, 10:1, 5:1, 10:3, 5:2, 2:1, 10:7, 5:4, 10:9, 0:1), to yield 15 fractions (Frs. 1–15) in the light of TLC profiles. Fr. 2 was further purified by semipreparative HPLC (72% CH₃OH/H₂O, 3.0 mL/min) to afford 4 (2.4 mg, t_R 18.2 min). Fr. 4 to Fr. 6 were merged and then was divided into 18 subfractions (Frs. 4-1–4-18) by ODS silica gel eluting with CH₃OH/H₂O (5–100%). Based on this, Fr. 4–8 was directly separated by semipreparative HPLC (69% CH₃OH/H₂O, 3.0 mL/min) to offer 12 (37.4 mg, t_R 21.0 min). Compound 9 (6.3 mg, t_R 13.3 min) was further purified from Fr. 4–9 by semipreparative HPLC (60% CH₃CN/H₂O, 3.0 mL/min). Compound 3 (9.6 mg, t_R 13.3 min), compound 7 (25.0 mg, t_R 15.2 min), and Fr. 4-10-1 were further obtained from Fr. 4–10 by semipreparative HPLC (80% CH₃OH/H₂O, 2.0 mL/min; 60% CH₃CN/H₂O, 2.5 mL/min), respectively. Meanwhile, Fr. 4-10-1 was separated by semipreparative HPLC (52% CH₃CN/H₂O, 0.04% formic acid, 3.0 mL/min) to gain 8 (9.7 mg, t_R 22.0 min). Fr. 4–2 to Fr. 4–7 were merged once again and then were divided into 15 subfractions (Frs. 4-2-1–4-2-15) by ODS silica gel eluting with CH₃OH/H₂O (5–100%). Compound 10 (2.3 mg, t_R 9.8 min) and 11 (12.7 mg, t_R 10.4 min) were further obtained from Fr. 4-2-5 by semipreparative HPLC (63% CH₃OH/H₂O, 2.5 mL/min), respectively. Compound 1 (6.83 mg, t_R 14.0 min), 5 (2.2 mg, t_R 9.7 min), 6 (4.5 mg, t_R 7.8 min) and Fr. 7-5 were firstly obtained from Fr. 7 by semipreparative HPLC (30% CH₃OH/H₂O, 2.5 mL/min; 5% CH₃OH/H₂O, 0.04% formic acid, 3.0 mL/min; 13% CH₃CN/H₂O, 0.04% formic acid, 3.0 mL/min, respectively. Meanwhile, Fr. 7-5 was further separated by semipreparative HPLC (12% CH₃CN/H₂O, 0.04% formic acid, 2.5 mL/min) to gain 2 (56.09 mg, t_R 12.0 min). Compound 13 (24.9 mg, t_R 7.4 min) was further purified from Fr. 9 by semipreparative HPLC (15% CH₃OH/H₂O, 2.5 mL/min).

3.5. Spectroscopic Data of Compounds

Embeurekol D (1): brown oil; ${\left[\alpha \right]}_{\mathrm{D}}^{25}$ + 49.0 (c 0.1, CH₃OH); UV (CH₃OH) λ_max (logε) 214 (4.43), 257 (4.02), 302 (3.85) nm; ECD (0.83 mM, CH₃OH) λ_max 208 (+16.47), 226 (−2.28), 234 (−0.02), 256 (−3.33), 306 (+1.62); IR (film) ν_max 3370, 1732, 1616, 984 cm⁻¹; ¹H and ¹³C NMR data, see Table 1; HRESIMS m/z 241.0709 [M+H]⁺ (calculated for C₁₁H₁₃O₆⁺, 241.0707).

7-heptyl-4-methoxy-6-oxo-3H-pyran-2-carboxylic acid (3): red brown solid; UV (CH₃OH) λ_max (logε) 226 (4.29), 312 (3.61) nm; IR (film) ν_max 3256, 2930, 1742, 1636, 1260, 1207 cm⁻¹; ¹H and ¹³C NMR data, see Table 2; HRESIMS m/z 269.1389 [M+H]⁺ (calculated for C₁₄H₂₁O₅⁺, 269.1384).

Embelin A (4): yellow solid; UV (CH₃OH) λ_max (logε) 205 (4.05), 287 (4.27) nm; IR (film) ν_max 3339, 2924, 2853, 1663, 1634, 1593, 1204, 691 cm⁻¹; ¹H and ¹³C NMR data, see Table 2; HRESIMS m/z 253.1442 [M+H]⁺ (calculated for C₁₄H₂₁O₄⁺, 253.1434).

Methyl α,2,5-trihydroxybenzeneacetate (5): brown oil; [α${]}_{\mathrm{D}}^{25}$ 0.0 (c 0.1, CH₃OH); UV (CH₃OH) λ_max (logε) 205 (4.16), 301 (3.45) nm; IR (film) ν_max 3377, 2918, 2851, 1734, 1576, 1456, 1219, 762 cm⁻¹; ¹H and ¹³C NMR data, see Table 1; HRESIMS m/z 221.0429 [M+Na]⁺ (calculated for C₉H₁₀NaO₅⁺, 221.0420).

3.6. X-ray Crystallographic Analysis

Compound 4 was dissolved in CH₃OH and slowly evaporated to obtain the clear light crystal. Crystallographic data for the structure that have been submitted to the Cambridge Crystallographic Data Centre (CCDC) can be seen and copies of which can be freely accessed applying to CCDC, 12 Union Road, Cambridge, CB2 1EZ, UK [P: +44 (0) 1223 336408].

Crystal data for 4: C₁₄H₂₀O₄, Mr = 252.30, crystal size 0.3 × 0.05 × 0.05 mm³, triclinic, a = 5.1888 (8) Å, b = 9.6297 (8) Å, c = 14.1355 (17) Å, α = 74.175 (9)°, β = 89.844 (12)°, γ = 79.330 (10)°, V = 666.93 (15) Å3, Z = 2, T = 99.98 (10) K, space group P-1, µ (Cu Kα) = 0.746 mm⁻¹, D_calc = 1.256 g/cm³, 4793 reflections measured (6.508° ≤ 2Θ ≤ 151.438°), 2535 unique (R_int = 0.0620, R_sigma = 0.0720). The final R₁ was 0.0950 (I > 2σ(I)), and wR₂ was 0.2899 (all data). The goodness of fit on F² was 1.090 (CCDC 2363995).

3.7. ECD Computation Section

Compounds 1 and 2 were placed in Spartan’14, which used the MMFF molecular force field to perform a conformational search on the potential isomers separately. Then, the stable conformers of the first 5% in methanol solvent were optimized at the B3LYP/6-31+G (d) level via Gaussian 09 (D.01, Pittsburgh, PA, USA). A TDDFT polarizable continuum model at the level of B3LYP/6-311+G (d, p) was used to calculate the optimized low-energy conformations [32]. The calculated ECD spectra were generated from GaussView (6.0.16, Pittsburgh, PA, USA) and Origin 2021 with a half-bandwidth of 0.3–0.4 eV, wavelength corrected by the calculated UV curve with the measured curve and weighted by Boltzmann distribution to obtain the calculated ECD spectra.

3.8. Mosher’s Method

Compound 1 (2.01 mg, 0.0084 mmol), followed by the addition of (R)-MPA (8.50 mg, 0.0512 mmol), DCC (0.3455 mg, 0.0017 mmol), and DMAP (0.2046 mg, 0.0017 mmol), was dissolved in chloroform-d (0.6 mL) and then stirred at room temperature for 6 h. The reaction was monitored by TLC detection [12]. (R)-MPA ester (1a) was purified by semipreparative HPLC (60% CH₃OH/H₂O, 3.0 mL/min) to yield 1a (0.7 mg). Compound 1b (0.7 mg) was prepared using the same protocol as for 1a. The determination of the configuration of the hydroxyl group at C-1′ for 2 was performed the same way as for 1.

3.9. Antibacterial Activity Assay

Two bacteria (Staphylococcus aureus ATCC 25923, Streptococcus suis SC19) were assessed on 96-well plates. As positive controls against the two microorganisms, streptomycin and penicillin were employed, respectively. Then, compounds were assessed using a two-fold serial dilution approach on nutrient agar on a 96-well plate as previously described [33]. The medium’s OD value was determined at 490 nm (OD₄₉₀) using a microplate reader (Thermo Scientific, Bremen, Germany) following a 24-h incubation period.

3.10. Antioxidant Activity Assay

The obtained compound 12 was evaluated for its antioxidant activities against DPPH. The effect of the compound on DPPH radical was estimated according to the method of Hatano et al. (1988) [34] and Yen et al. (1995) [35]. In summary, a methanolic DPPH solution was supplemented with compound 12 to obtain a final concentration of 2.5–250 μg/mL. After shaking the mixture and letting it stand for 30 min at room temperature in the dark, the OD₅₁₇ values were measured using the PerkinElmer Enspire Multi-mode micro-orifice detector and enzyme labeling instrument (PerkinElmer, Waltham, MA, USA). Ascorbic acid was used as the positive control. Then, the free radical scavenging rate K (%) was computed based on the acquired OD₅₁₇ value using the formula, and Origin 2021 was utilized to determine EC₅₀.

3.11. PDE4 Inhibitory Screening Assays

The PDE4D2 expression, purification, and enzymatic assay procedures were comparable to those we previously described [36]. In brief, the inhibition of PDE4D2 by compounds 1–3 and 9–13 was measured using ³H-cAMP as substrates (20,000–30,000 cpm/assay), and reactions took place in a mixture that also contained 50 mM Tris-HCl (pH 7.5), 10 mM MgCl₂, 0.5 mM (DTT) at room temperature (25°) for 15 min. Subsequently, the reaction was stopped with 0.2 M ZnSO₄, the unreacted ³H-cAMP was precipitated out using 0.2 N Ba(OH)₂. The leftover supernatant was used to gauge the radioactivity in 2.5 mL of Ultima Gold liquid scintillation cocktails (PerkinElmer) by a liquid scintillation counter (PerkinElmer 2910) [36].

3.12. Cytotoxicity Bioassay

Human prostate cancer LNCaP, 22Rv1, PC-3, and DU145 cells were purchased from the Cell Bank/Stem Cell Bank of the Chinese Academy of Sciences (Shanghai, China). All cell lines were cultured in the medium according to the recommendations of the Chinese Academy of Sciences Cell Bank/Stem Cell Bank. The cells were incubated at 37 °C and 5% carbon dioxide. All culture media supplemented with 10% (v/v) fetal bovine serum (Biological Industries, Beit Haemek, Israel) and 1% penicillin/streptomycin. All cell lines were tested and found to be free of mycoplasma contamination. The MTT assay, which has been previously reported, was used to assess cytotoxicity [37]. To sum up, cells were treated with compounds after being seeded at a density of 5 × 10³ per well on a 96-well plate overnight after 72 h incubation. Following drug treatment, 10 μL of MTT solution (5 mg/mL) was added to each well at the indicated timings, and the wells were then incubated for 4 h at 37 °C in a humidified 5% CO₂ incubator. After the supernatants were eliminated, DMSO (100 μL) was added. Then, the OD₅₇₀ values after 10 min were measured using a TECAN Infinite 200 PRO Nano Quant multimode microplate reader [37,38]. Three separate runs of the experiment were conducted.

3.13. Molecular Docking

Using the structure of AKT1 (PDB code: 3O96) [30], which was obtained from the Protein Data Bank as a starting model, it was then handled following the Protein Preparation Wizard workflow in Maestro 11.9. The optimized ligand was docked into the receptor with the default parameters, which were obtained from Receptor Grid Generation [39,40,41,42].

4. Conclusions

In conclusion, three new polyketides (1, 3, and 5) and one new natural product (4), together with nine known ones (2 and 6–13), were isolated from the mangrove sediment-derived fungus Penicillium sp. SCSIO 41411. Their structures were determined by extensive spectroscopic analyses, ECD calculations, and X-ray single-crystal diffraction. Compound 9 inhibited PDE4 with an inhibitory ratio of 40.78% at 10 μM. Compound 12 showed moderate antioxidant inhibitory activity with an EC₅₀ of 16.21 µg/mL. Moreover, compound 4 demonstrated selective inhibitory activity against tumor cells, which provides a promising avenue for the development of anti-prostate cancer therapeutics.