Secondary metabolites from the marine-derived fungus Penicillium chrysogenum Y20-2, and their pro-angiogenic activity.

A systematic chemical study of the secondary metabolites of the marine fungus, Penicillium chrysogenum (No. Y20-2), led to the isolation of 21 compounds, one of which is new (compound 3). The structures of the 21 compounds were determined by conducting extensive analysis of the spectroscopic data. The pro-angiogenic activity of each compound was evaluated using a zebrafish model. The results showed that compounds 7, 9, 16, and 17 had strong and dose-dependent pro-angiogenic effects, with compound 16 demonstrating the strongest pro-angiogenic activity, compounds 6, 12, 14, and 18 showing moderate activity, and compounds 8, 13, and 19 exhibiting relatively weak activity.

Keywords: bioactive metabolite; marine-derived fungus; Penicillium chrysogenum; pro-angiogenesis

Marine natural products are an important source of drug compounds and can serve as the material basis for the development of new drugs [[1]]. Marine microorganisms contain abundant natural active substances, among which marine fungi are a crucial component. These substances have attracted increasing attention as a source of novel bioactive secondary metabolites [[2]]. In a study of the secondary metabolites of marine fungi, Penicillium sp. has received much attention as an important source of novel natural products with multiple chemical structures and biological activities. As of 2020, approximately 600 compounds have been isolated from marine sources of Penicillium [[4]]. Various activity studies have shown that these compounds have prominent antibacterial, anticancer, and antioxidant activities and have potential applications in the development of novel drugs.

Penicillium sp. fungi, as an important group of marine microorganisms, have great potential for exploitation. Despite numerous studies on the activity of marine compounds (e.g., antibacterial, anticancer, cytotoxic, antitumor, and antioxidant), relatively little research has been performed on the pro-angiogenic activity of the compounds. During our ongoing search for new pro-angiogenic agents from marine fungi [[5]], we screened and obtained a pro-angiogenic active marine fungus, Penicillium chrysogenum (No. Y20-2), the isolation, structure, and biology of which are described in the current study.

Fermentation of the fungus P. chrysogenum (No. Y20-2) was conducted on a large scale using rice medium, and the fermentation products were extracted and concentrated to obtain 21.71 g of crude extracts. The extracts were separated using silica gel, octadecyl silica (ODS), Sephadex gel column chromatography, and high-performance liquid chromatography (HPLC) to yield the following 21 compounds: cerebroside A (1) [[7]], cerebroside B (2) [[7]], (2S,2′R,3R,3′E,4E,8E)-N-2′-hydroxyhexadecanoyl-2-amino-9-methyl-4,8-octadecadiene-1,3-diol (3), (2S,2′R,3R,4E,8E)-N-2′-hydroxyhexadecanoyl-2-amino-9-methyl-4,8-octadecadiene-1,3-diol (4) [[8]], methyl linolenate (5) [[9]], palmitic acid (6) [[10]], n-octadecanoic acid (7) [[11]], 13-docosenamide (8) [[12]], (E)-5-acetoxy-3-methylpent-2-enoic acid (9) [[13]], methyl ester (5Z)-5-hexacosenoate (10) [[14]], (E)-1-aminotridec-5-en-2-ol (11) [[15]], linolenic acid (12) [[16]], R-mevalonolactone (13) [[17]], walterolactone A (14) [[18]], 3,6-dihydrixy-4-decen-9-olide (15) [[19]], N-(2-hydroxyphenyl)-acetamide (16) [[20]], ethyl formyltyrosinate (17) [[22]], (–)-dechloroxylariamide A (18) [[23]], meleagrin (19) [[24]], [[25]], [[26]], [[27]], (22E,24R)-5α,8α-peroxide ergostere-6,22-diene-3-ol (20) [[28]], [[29]], [[30]], and 6β-methoxyergosta7,9(11),22E-triene-3β,5α-diol (21) [[31]]. Among these 21 isolated compounds, compound 3 was newly identified. The chemical structures of the compounds were analyzed using spectroscopic data and physicochemical properties. The chemical structures of known compounds were identified by comparing their nuclear magnetic resonance (NMR) data with those reported previously.

Compound 1 was isolated as a white powder (soluble in chloroform). The 1H and 13C-NMR spectra of compound 1 showed signals of an amide group at δH 7.44 (1H, NH) and δC 174.8, and a methine linked to amide nitrogen at δH 3.83 (1H) and δC 53.7 [[32]]. The 1H NMR spectrum also indicated the presence of two terminal methyl groups [δH 0.88 (6H)], the aliphatic methylenes [δH 1.25–1.33 (34H)], a methyl group attached to the double bond [δH 1.56 (3H)], four methylenes attached to the double bonds [δH 1.91–2.00 (8H)], a methylene attached to an O-atom [δH 3.35 (1H) and 3.51 (1H)], two methines attached to an O-atom [δH 4.37 (1H) and 4.55 (1H)], and five olefinic methines [δH 5.07 (1H), 5.48 (2H), 5.73 (1H), and 5.81 (1H)]. These data suggest that compound 1 was a ceramide. Additionally, the H-atom signals at δH 3.46, 3.51, 3.52, 3.84, 4.06, and 5.73 in the 1H NMR spectrum and the C-atom signals at δC 61.2, 72.5, 73.1, 76.2, 81.5, and 103.2 in the 13C NMR spectrum (Table 1) indicated the presence of a glucose structure. Finally, compound 1 was identified as cerebroside A by comparing its NMR data with those reported earlier [[7]] (Figure 1).

Table 1: 1D NMR data of compounds 1 and 3.

Graph: Figure 1: Structures of the 21 compounds isolated from Penicillium chrysogenum.

Compound 3 was obtained as brown oil. Its molecular formula was C35H65NO4, as determined using high resolution electrospray ionization mass spectroscopy (HRESIMS), implying four degrees of unsaturation. Both the 1H and 13C NMR data of compound 3 closely resembled those of 1 (Table 1), except for a lack of the glucose group in 3. Hence, we inferred that compound 3 was the aglycone structure of compound 1, and this was confirmed by further analysis of the NMR spectra. Further analysis of the 1H and 13C NMR together with the heteronuclear single-quantum correlation spectroscopy (HSQC) data (Table 1) indicated the presence of an amide fragment (δH 7.00 (1H) and δC 173.1), three methyls [δH 0.88 (6H) and 1.58 (3H) and δC 14.3 × 2 and 16.2], 21 aliphatic methylenes [δH 1.26–1.38 (34H), 1.93 (2H), and 2.02–2.10 (6H) and δC 22.8 × 2, 27.7, 28.2, 29.7 × 12, 32.1 × 2, 32.4, 32.7, and 40.0], a methine linked to an N-atom [δH 3.87 (1H) and δC 54.7], two methines linked to hydroxyls [δH 4.32 (1H) and 4.54 (1H) and δC 73.3 and 74.6], a methylene linked to hydroxyl [δH 3.71 (1H) and 3.94 (1H) and δC 62.3], five sp2 methines [δH 5.08 (1H), 5.53 (1H), 5.56 (1H), 5.78 (1H), and 5.88 (1H) and δC 123.2, 127.4, 128.9, 134.2, and 136.7], and an olefinic quaternary carbon (δC 136.5). The 1H–1H correlated spectroscopy (COSY) correlations (Figure 2) helped to identify two coupling systems, CH2-1 (δH 3.71 and 3.94) and NH (δH 7.00)-CH-2 (δH 3.87)-CH-3 (δH 4.32)-CH-4 (δH 5.53)-CH-5 (δH 5.78)-CH2-6 (δH 2.02–2.10)-CH2-7 (δH 2.02–2.10)-CH-8 (δH 5.08) and CH-2′ (δH 4.54)-CH-3′ (δH 5.56)-CH-4′ (δH 5.88)-CH2-5′ (δH 2.02–2.10)-CH2-6′ (δH 1.26–1.38). The 1H–1H COSY correlation of NH (δH 7.00) with CH-2 placed the amide group at C-2, which was confirmed using the heteronuclear multiple bond correlations (HMBC) from NH to C-3 (δC 74.6). The methyl, CH3-19, was located at C-9 according to the HMBC correlations of CH3-19 (δH 1.58) with C-8 (δC 123.2) and C-10 (δC 40.0). The side chains from C-2′ to C-16′ were linked at the acyl group according to the HMBC correlation of CH-2′ with C-1′ (δC 173.1). The rest of the two aliphatic side chains were identified based on a comparison of the corresponding NMR data between compounds 1 and 3 (Table 1). Other HMBC correlations are shown in Figure 2. The configuration of the double bond at C-4 and C-3′ was deduced to be E based on the large 1H–1H coupling constant (J = 15.6 Hz) between H-4 and H-5 and between H-3′ and H-4′ [[7], [33]]. The E geometry of the double bond at C-8 was determined by the chemical shift value of the olefinic methyl group (δC 16.2 for CH3-19) [[7], [34]]. The values of δC-1–C-18 and δC-1′–C-16′ and the specific rotation ([α]25D 3.2°) of 3 were close to those of cerebroside A (1) ([α]25D 5.6°), (2S,2′R,3R,4E,8E)-N-2′-hydroxyhexadecanoyl-2-amino-9-methyl-4,8-octade]cadiene-1,3-diol (4) ([α]25D 3.2°), and (2S,2′R,3R,3′E,4E,8E)-N-(2′-hydroxy-3′-hexadecenoyl)-9-methyl-4,8-icosadien-1,3-diol [[35]], suggesting that compound 3 shared the same 2S,2′R,3R- and 3′E,4E,8E-configurations. Taken together, the structure of X compound 3 was determined to be (2S,2′R,3R,3′E,4E,8E)-N-2′-hydroxyhexadecanoyl-2-amino-9-methyl-4,8-octadecadiene-1,3-diol.

Graph: Figure 2: Key COSY and HMBC correlations of compound 3.

Zebrafish are characterized by small size for high-throughput screening and transparent body of juvenile fish for easy observation. Zebrafish are 87 % genetically homologous to humans and have physiological and pharmacological responses similar to those of humans and other higher mammals, making it well suited for the identification of drugs and bioactive natural products with therapeutic potential [[36]]. The pro-angiogenic activity of the isolated compounds was evaluated using a zebrafish model. The formation of zebrafish intersegmental vessels (ISVs) can be blocked by selective inhibitors of VEGFR tyrosine kinases, such as PTK787 [[37]], which can be used to generate a zebrafish model of vascular injury. Danhong injection (DHI) has been shown to have significant efficacy in the treatment of ischemic cardio-cerebrovascular diseases [[39]]. Moreover, studies have shown that DHI has a significant regenerative function on zebrafish ISVs and can repair and protect against vascular damage, with reliable methods and stable results [[40]]. Compared to the blank group, the zebrafish in the model control group had significantly fewer ISVs, indicating successful experimental modeling [[41]]. The positive control group had a significant increase in the intersegmental angiogenesis of zebrafishes compared to those in the model group. Compounds 12 and 14 were lethally toxic to zebrafish at high concentrations but exhibited obvious pro-angiogenic activity at low to medium concentrations. Compounds 11, 15, and 21 were purified, but the resulting mass was insufficient to perform activity experiments. The remaining compounds were subjected to two batches of activity experiments depending on the time of isolation. As shown in Figures 3 and 4, four compounds, namely, n-octadecanoic acid (7), (E)-5-acetoxy-3-methylpent-2-enoic acid (9), N-(2-hydroxyphenyl)-acetamide (16), and ethyl formyltyrosinate (17), showed strong pro-angiogenic activities at all tested concentrations. Among them, compound 16 had the strongest activity, and this was also obtained at the highest amount (48.5 mg), suggesting that it may be the primary angiogenic active substance of this strain. Compounds 6 and 18 were inactive at low concentrations but active at medium to high concentrations, and compounds 8, 13, and 19 were relatively inactive. Additionally, compound 10 showed lethal toxicity at all tested concentrations and still caused mortality to zebrafish when its dosing concentration was reduced to 10 μg/mL.

Graph: Figure 3: Angiogenic activities of compounds 1–4, 6, 7, 12, 19, and 20 (n = 10, mean ± SEM). (A) In vivo observation of the intersegmental blood vessels (ISVs) of zebrafish (Scale bar = 200 μm). (B) Statistical analysis results of the zebrafish ISVs for all of the treated groups. Comparisons with the control group; ####p < 0.0001. Comparisons with the model group; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Graph: Figure 4: Angiogenic activities of compounds 5, 8, 9, 13, 14, and 16–18 (n = 10, mean ± SEM). (A) In vivo observations of the intersegmental blood vessels (ISVs) of zebrafish (Scale bar = 200 μm). (B) Statistical analysis results of the zebrafish ISVs for all of the treated groups. Comparisons with the control group; ####p < 0.0001. Comparisons with the model group; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

The optical rotations were measured on a Perkin-Elmer 241 MC at 20 °C. The HRESIMS data were obtained on an Agilent 6210 ESI/TOF mass spectrometer. The NMR spectra were acquired on a Bruker Avance spectrometer operating at 400 (1H) and 100 (13C) MHz, with tetramethylsilane (TMS) as an internal standard. HPLC was performed on an Agilent 1260 G1311A pump equipped with a G1322A degasser, a G1315D DAD detector, and a C18 column (preparation, Cosmosil 5C18-MS-II, 10 × 250 mm, 5 μm; analysis, Diamonsil C18(2), 4.6 × 250 mm, 5 μm). Column chromatography (CC) was conducted using silica gel (200–300 mesh), Sephadex LH-20 (25–100 mm), and reversed-phase C18 silica gel (150–200 mesh). Thin layer chromatography (TLC) was performed on plates pre-coated with silica gel GF254, where the spots were visualized under ultra violet (UV) (254 and 365 nm) light and by heating after spraying with a solution of H2SO4–EtOH (1:9, v/v).

The fungus Y20-2 was isolated from a seawater sample collected from the Indian Ocean on November 5, 2012, and was identified as P. chrysogenum using the ITS rDNA sequences (GenBank: MH443755.1). The strain was deposited in the Drug Screening Research Laboratory, the Institute of Biology Institute of the Shandong Academy of Sciences.

To obtain the seed culture, the fungus was cultivated in two 500 mL Erlenmeyer flasks, each of which contained 100 mL of potato dextrose broth (PDB), on a rotary shaker (150 rpm) at 28 °C for 7 days. Large-scale fermentation was conducted on a rice medium in twenty 500 mL Erlenmeyer flasks, the rich medium containing 65 g of rice and 100 mL of seawater, which was sterilized and used, at room temperature for 50 days. The fermented medium containing the mycelium was extracted using ethyl acetate (3 × 6 L) and methanol (3 × 6 L) in turn, and the extracts were collected, filtered, concentrated under reduced pressure, and dried to obtain 21.71 g of crude extract.

The crude extract was preliminarily separated into 10 fractions (Fr.1–Fr.10) by silica gel CC using a gradient elution with dichloromethane-methanol (100:0–0:100, v/v). Fr.2 was chromatographed on a Sephadex LH-20 CC using isocratic conditions for elution using the mobile phase CH2Cl2–MeOH (1:1) to give five subfractions (Fr.2-1–Fr.2-5). Among the subfractions, Fr.2-3 and Fr.2-5 were individually subjected to the ODS reverse CC with the MeOH–H2O elution and further purified by semi-preparative HPLC using different concentrations of methanol in H2O to give the following compounds: 11 (45 % MeOH–H2O, 1.3 mg), 5 (95 % MeOH–H2O, 16.7 mg), 12 (95 % MeOH–H2O, 24.7 mg), 20 (95 % MeOH–H2O, 25.4 mg), 21 (95 % MeOH–H2O, 1.0 mg), 13 (12 % MeOH–H2O, 18.9 mg), 14 (12 % MeOH–H2O, 3.5 mg), 15 (12 % MeOH–H2O, 1.2 mg), 9 (50 % MeOH–H2O, 25.7 mg), 16 (50 % MeOH–H2O, 48.5 mg), 6 (50 % MeOH–H2O, 21.9 mg), 7 (50 % MeOH–H2O, 28.1 mg), and 10 (50 % MeOH–H2O, 6.2 mg). Fr.3 was subjected to ODS reverse CC with MeOH/H2O elution to give compound 17 (13 mg). Fr.4 was fractionated on a Sephadex LH-20 column using isocratic conditions for elution with CH2Cl2–MeOH (1:1) to obtain six subfractions (Fr.4-1–Fr.4-6). Fr.4-4 and Fr.4-5 were separated following a similar procedure to that used for fractions Fr.2-3 and Fr.2-5 to afford compounds 3 (100 % MeOH–H2O, 5.4 mg), 4 (100 % MeOH–H2O, 9.8 mg), and 19 (70 % MeOH–H2O, 15.2 mg). Fr.4-6 was subjected to silica gel CC, followed by ODS reverse CC with MeOH/H2O elution and was further purified using semi-preparative HPLC to yield compound 18 (28 % MeOH–H2O, 15.6 mg). Fr.6 was chromatographed on a Sephadex LH-20 column using isocratic conditions for elution with CH2Cl2–MeOH (1:1) to obtain four subfractions (Fr.6-1 to Fr.6-4). Fr.6-2 was purified by semi-preparative HPLC to give compounds 1 (MeOH, 29.2 mg) and 2 (MeOH, 29.7 mg). Fr.6-3 was subjected to silica gel CC, followed by ODS reverse CC with MeOH/H2O elution, and further purified by thin layer preparation to yield compound 8 (1.6 mg).

The zebrafish strain used in this experiment was the Tg (FLI1-EGFP) transgenic line [[42]], which was maintained in an automatic circulating tank system (ESEN, Beijing, China) with a 10 h dark/14 h light cycle at (28 ± 0.5) °C, pH 7.0 ± 0.5, and fed live brine shrimp twice a day. The embryos were obtained from natural spawning. Then, the fertilized eggs were collected, disinfected with methylene blue solution, washed three times with fish water, and placed in a light-operated incubator at 28 °C.

The pro-angiogenesis assay was conducted according to a previously described method [[5]]. At 24 hpf (hours postfertilization), the zebrafish larvae were removed from the egg membrane with pronase E and randomly placed in 24-well cell culture plates with 10 embryos per well. The exposure groups were divided into six groups, including a blank control group (dimethyl sulfoxide [DMSO] solution, 0.4 %, v/v), a model control group (PTK787, 0.25 μg/mL), a positive control group (0.25 μg/mL of PTK787 + 9 μL/mL of Danhong injection [DHI]), and three sample groups (0.25 μg/mL of PTK787 + 25, 50, and 100 μg/mL of the compound). After drug treatment, the zebrafish in each group were placed in a light incubator at 28 °C for 24 h. Following incubation, the zebrafish were observed under a fluorescent microscope and photographed. The total lengths of the zebrafish ISVs were measured using Image-Pro Plus 5.1 software. Three replicates were conducted for each test group.

All data are presented as the mean ± standard error (SEM). Statistical analysis was performed using GraphPad Prism 6.0 for Windows (GraphPad Prism Software, San Diego, CA). One-way ANOVA was used to evaluate differences among multiple groups. *p < 0.05 was treated as statistically significant.

In summary, 21 compounds were isolated from the marine fungus Y20-2, one of which was newly identified (compound 3). The structures of the compounds were determined using spectroscopic data and by comparison with previously published data. Preliminary screening of these compounds for pro-angiogenic activity was performed using a zebrafish model, and compound 16 was identified as the primary pro-angiogenic bioactive substance of this strain. Additionally, compounds 7, 9, and 17 showed strong pro-angiogenic activity. The results suggest that these compounds are promising candidates for further pharmacologic and biosynthetic research.