1. Introduction
Austalides are a class of natural meroterpenoids with attractive scaffolds. Previous biosynthetic studies revealed that they are biosynthesized by cyclization and oxidative modification of 6-[(2
E, 6
E)farnesyl]-5,7-dihydroxy-4-methylphthalide [
1]. These meroterpenoids were mainly produced by the species of the fungal genera
Aspergillus and
Penicillium, especially those from marine environments. Since austalides A−E were first reported in 1981, a total of 36 analogues have been identified [
2,
3,
4,
5,
6,
7,
8,
9,
10,
11]. The structural variations of austalides are attributed to oxidation occurring at C-13, C-14, C-17, and the isopropyl (C-15, C-25, C-26) to generate alcohol, isopropenyl, lactone, ester, or ether functionalities. Some of the analogues exhibited significant bioactivities, such as cytotoxicity [
6,
10], antibacterial activity [
7], anti-influenza A virus (H1N1) activity [
8], endo-1,3-
D-glucanase inhibition [
9], and osteoclast differentiation inhibitory effects [
11].
In our efforts to search for new or bioactive molecules from deep-sea-derived
Penicillium or
Aspergillus strains [
12,
13,
14,
15], chemical examination of
Penicillium thomii YPGA3 afforded a new austalide meroterpenoid (
1) and seven known analogues (
2−
8) (
Figure 1). Compound
1 represented the third example of austalides bearing a phenol hydroxy at C-5 instead of the conserved methoxy in other known analogues. Additionally, two new labdane-type diterpenoids (
9 and
10) and a known analogue (
11) were also obtained (
Figure 1). The labdane-type diterpenoids are mainly produced by plants and are rarely found in fungus. To our knowledge, diterpenoids belonging to the labdane-type were discovered from species of
Penicillium for the first time. All compounds were evaluated for their inhibitions against α-glucosidase and NO (nitric oxide) production in lipopolysaccharide (LPS)-activated RAW 264.7 macrophages, and cytotoxicity toward two types of human breast carcinoma cells. Herein, the isolation, structural elucidation, and the bioactivity of compounds
1–
11 are described.
2. Results
Compound
1 had a molecular formula of C
25H
34O
7, as established by the high-resolution electrospray ionization mass spectroscopy (HRESIMS) (469.2202 [M + Na]
+, calcd. for 469.2197) (see
Figure S34 in the
Supplementary Materials), requiring nine degrees of unsaturation. The
1H nuclear magnetic resonance (NMR) spectrum exhibited signals for five methyl singlets (δ
H 2.02, 1.28, 1.21 × 2, 0.71), a methoxy (δ
H 3.67), an oxygenated methylene (δ
H 5.22), and a series of alkyl protons (see
Figure S1). The
13C NMR and the heteronuclear single-quantum coherence (HSQC) spectra exhibited 25 carbon resonances attributable to a benzene ring (δc 160.8, 153.8, 145.4, 112.3, 103.0, 111.2), five methyls (δc 33.2, 28.2, 27.7, 19.4, 10.6), a methoxy (δc 52.0), six methylenes (δc 70.6, 40.5, 22.6, 30.1, 34.9, 18.4) including an oxygenated one, two methines (δc 41.5, 52.1), two ester carbons (δc 174.2, 177.0), one sp
3 quaternary carbon (δc 42.8), and two oxygenated carbons (δc 78.2, 75.8) (see
Figures S2 and S3). As six degrees of unsaturation were accounted for by the benzene ring and two carbonyl carbons, the remaining three degrees of unsaturation required that compound
1 contained three additional rings. The aforementioned information was very similar to that of austalide P (
2) [
6], a co-isolated analogue first isolated from a sponge-associated fungus
Aspergillus sp., with the only difference owing to the absence of the aromatic methoxy group (δ
H 4.04, δ
C 62.2) in
2. This indicated that
1 was the 5-demethoxylated derivative of
2. The structure of
1 was further certified as correct by detailed interpretation of 2D NMR data (
Figure 2). The relative configuration of
1 was assigned by a nuclear Overhauser effect spectroscopy (NOESY) experiment (
Figure 3). The NOE correlations of H-21 (δ
H 1.68)/H
3-24 (δ
H 1.21), H-14 (δ
H 1.55), H
3-24/H-22β (δ
H 2.69), and H
3-27 (δ
H 0.71)/H-22α (δ
H 2.94) clarified the same orientation of H
3-24, H-21, and H-14, while H
3-27 was in the opposite orientation. Thus, the relative configuration of
1 was assigned as 11
S∗, 14
R∗, 20
S∗, and 21
R∗. In order to resolve its absolute configuration, the theoretical electronic circular dichroism (ECD) data of 11
S, 14
R, 20
S, 21
R-
1 were calculated by the time-dependent density-functional theory (TDDFT) method and showed an ECD curve with Cotton effects at 265 (−), 227 (+), and 214 (−) nm, which were in good agreement with the experimental cotton effects at 264 (−), 229 (+), and 212 (−) nm (
Figure 4), suggesting that compound
1 has the 11
S, 14
R, 20
S, 21
R configuration. Compound
1 was given the trivial name austalide Y and represented the third example of an austalide meroterpenoid without the 5-methoxy group.
Compound
9 was isolated as a colorless oil with the molecular formula of C
20H
30O
5 as determined by the HRESIMS at
m/
z 373.1991 [M + Na]
+ (calcd. 373.1985), requiring six degrees of unsaturation. The
1H NMR spectrum exhibited signals for an olefinic methyl (δ
H 2.13), two methyl singlets (δ
H 1.38 and 0.69), an olefinic methylene (δ
H 4.56, 4.90), an oxygenated proton (δ
H 3.18), and several aliphatic protons. The
13C NMR spectra exhibited 20 carbon resonances, which were classified by an HSQC experiment as three methyls (δ
C 24.8, 18.9, 13.5), six sp
3 methylenes (δ
C 40.7, 39.6, 38.7, 29.7, 27.2, 23.0), three methines (δ
C 79.0, 56.5 × 2) including one oxygenated methine, two sp
3 quaternary carbons (δ
C 50.3, 41.0), two carboxylic acid groups (δ
C 170.4, 180.5), and two double bonds (δ
C 161.7, 149.0, 107.3, 116.9). As four of the six degrees of unsaturation were covered by two carboxylic acid groups and two double bonds, the remaining two degrees of unsaturation required that
9 was bicyclic. The gross structure was further established by detailed analyses of the 2D NMR data (
Figure 2). The correlation spectroscopy (COSY) relationships from H
2-1 to H-3 and H-5 to H
2-7 coupled with the heteronuclear multiple-bond correlations (HMBCs) of H
3-20 (δ
H 0.69) to C-1 (δ
C 38.7), C-5 (δ
C 56.5), C-10 (δ
C 41.0), H
3-18 (δ
H 1.38) to C-3 (δ
C 79.0), C-4 (δ
C 50.3), C-5, C-19 (δ
C 180.5), the olefinic methylene protons H
2-17 (δ
H 4.56, 4.90) to C-7 (δ
C 39.6), C-8 (δ
C 149.0), C-9 (δ
C 56.5), and H-9 (δ
H 1.61) to C-11 (δ
C 23.0), C-20 (δ
C 13.5) established a bicyclic unit (unit A). Additional HMBCs from H
3-16 (δ
H 2.13) to C-12 (δ
C 40.7), C-13 (δ
C 161.7), C-14 (δ
C 116.9), and H-14 (δ
H 5.61) to C-15 (δ
H 170.4) established a senecioic acid moiety (unit B), which was linked to unit A at C-11 by the
1H-
1H COSY relationship between H
2-11 and H
2-12. Thus, the gross structure of
9 was established as depicted. The relative configuration of
9 was determined by an NOESY experiment (
Figure 3) and coupling constants. The coupling constants of
JH-3/H-2α (4.4 Hz) and
JH-3/H-2β (12.1 Hz) suggested that OH-3 was
β-orientated. The NOE correlations of H-5 with H-3, H-9 (δ
H 1.61), H
3-18 (δ
H 1.38), H
3-18 with H-6α (δ
H 2.03), and H
3-20 (δ
H 0.69) with H-6β (δ
H 1.93) indicated that H-3, H-5, and H-9 were α-orientated, while H
3-20 was
β-orientated. The NOE correlation between H-14 and H
2-12 (δ
H 2.30, 2.02) was indicative of an
E configuration of the double bond Δ
13. Comparison of the experimental ECD data with those of the calculated ECD data at the B3LYP/6-31+G(d,p) level for 3
S, 4
R, 5
R, 9
S, 10
R-
9 allowed the assignment of the 3
S, 4
R, 5
R, 9
S, 10
R configuration for
9 (
Figure 5). As the structure of
9 was 3-hydroxylated derivative of agathic acid (
10) [
16], it was named 3
β-hydroxy-agathic acid.
Compound
10 had a molecular formula of C
22H
32O
6, as determined by the HRESIMS at
m/
z 415.2096 [M + H]
+ (calcd. 415.2091), requiring seven degrees of unsaturation. The NMR data of
10 were similar to those of
9 with the obvious distinction due to the presence of an acetyl group (δ
H 2.04
, δ
C 172.8), suggesting that
10 was an acetylated derivative of
9. The downfield-shifted H-3 (δ
H 4.55) showed an HMBC with the acetyl carbonyl carbon (δ
C 172.8), locating the acetyl group at C-3 (
Figure 2). The relative configuration of
10 was determined to be the same as that of
9 based on their similar NOESY data. The structure of
10 was determined as depicted and is a C-3 epimer of a known analogue mumic acid A [
17]. The similar specific rotations and circular dichroism (CD) spectra of
9 and
10 confirmed the same absolute configuration of both
9 and
10 (
Figure 5), and compound
10 was named 3
β-acetoxy-agathic acid.
In addition, eight additional known compounds were identical to austalide P (
2) [
6], austalide H (
3) [
5], austalide P acid (
4) [
9], austalide H acid (
5) [
9], 17-O-demethylaustalide B (
6) [
9], austalide Q acid (
7) [
9], 13-deoxyaustalide Q acid (
8) [
9], and agathic acid (
10) [
16] based on comparisons of their NMR data and specific rotations with those reported in the literature. Furthermore, the
13C NMR data of austalide H (
3) and
13C NMR data of compounds
3,
4, and
7 in methanol-
d6 were reported for the first time.
All compounds were screened for their inhibitory activities against α-glucosidase at the initial concentration of 1 mM. Compounds 2 and 11 exhibited inhibition by more than 50% and were further evaluated to calculate the IC50 values. The results showed that compounds 2 and 11 inhibited α-glucosidase with IC50 values of 910 ± 4 and 525 ± 2 μM, being more active than the positive control acarbose (1.33 mM). Other compounds, on the other hand, showed inhibition less than 40% at the concentration of 1 mM. As for the labdane-type diterpenoids 9−11, the introduction of hydroxy and acetoxy groups at C-3 may lead to a sharp decrease in activity, since compounds 9 and 10 showed low inhibition when compared with that of 11.
The isolated compounds were also evaluated for their inhibitory effects against NO production in LPS-activated RAW 264.7 macrophages at the concentration of 50 μM following the same procedures in our previous study [
10]. The cell viability was further determined by the MTT assay to evaluate whether the inhibition on NO production was owing to the cytotoxicity. As results (see
Table S1 in the
Supplementary Materials), compounds
1,
2, and
10, possessing inhibition rates of more than 50% on NO production, showed obvious cytotoxic effects, which suggested that the inhibitory effects of NO production were due to the cytotoxicity. All compounds were further evaluated for their cytotoxicity toward two types of human breast carcinoma cells (MCF-7, MDA-MB-468) [
13], and the results showed that only compound
1 showed a weak inhibitory effect toward MDA-MB-468 cells with an IC
50 value of 38.9 ± 1.83 μM.
3. Experimental Section
3.1. General Experimental Procedure
Specific rotations were recorded by an SGW®-1 automatic polarimeter (Shanghai Jing Ke Industrial Co., Ltd., Shanghai, China). The NMR spectra were measured on a Bruker Avance III HD-400 spectrometer (Bruker, Fällanden, Switzerland). HRESIMS spectra were obtained on a Waters Xevo G2 Q-TOF spectrometer (Waters Corporation, Milford, MA, USA). Semi-preparative high-performance liquid chromatography (HPLC) was undertaken on a Shimadzu LC-6AD pump (Shimadzu Co., Kyoto, Japan) using a UV detector, and a YMC-Pack ODS-A HPLC column (semipreparative, 250 × 10 mm, S-5 μm, 12 nm, YMC Co., Ltd, Kyoto, Japan) was used for separation.
3.2. Fungal Strain and Identification
Fungus YPGA3 was isolated from deep sea water at a depth of 4500 m in the Yap Trench (West Pacific Ocean). The strain was identified as
Penicillium thomii based on microscopic examination and by internal transcribed spacer (ITS) sequencing. The ITS sequence was deposited in GenBank (
http://www.ncbi.nlm.nih.gov) with accession number MG835903. The strain YPGA3 (MCCC 3A01052) was deposited at the Marine Culture Collection of China.
3.3. Fermentation
The fermentation was carried out in 30 Fernbach flasks (500 mL), each containing 70 g of rice. Artificial seawater (90 mL) was added to each flask, and the contents were soaked for three hours before autoclaving at 15 psi for 30 min. After cooling to room temperature, each flask was inoculated with 3.0 mL of the spore inoculum and incubated at room temperature for 30 days.
3.4. Extraction and Isolation
The fermented materials were extracted with ethyl acetate (EtOAc) (3 × 5000 mL) in an ultrasonic bath at 30 °C for 20 min. After evaporation under vacuum, the EtOAc extract (3.1g) was subjected to ODS silica gel column chromatography (CC) eluting with MeOH/H2O (20:80→100:0) to afford ten fractions (F1–F10). F5 was further chromatographed over C-18 silica gel CC eluted with MeOH/H2O (65:35) to afford seven subfractions (F5a–F5g). F5d was further purified by HPLC on a semi-preparative YMC-pack ODS-A column using CH3CN/H2O (61:39, 3 mL/min) to afford 5 (31 mg, tR 40 min). F6 was further chromatographed over ODS silica gel CC eluted with MeOH/H2O (20:80→100:0) to afford fourteen subfractions (F6a–F6n). F6d was separated by HPLC CH3CN/H2O (54:46, 3 mL/min) to yield 3 (14 mg, tR 19.9 min). Purification of F6e by HPLC using CH3CN/H2O (54:46, 3 mL/min) gave 7 (10 mg, tR 21.9 min). F6h was purified by HPLC using CH3CN/H2O (52:48, 3 mL/min) to give 9 (3 mg, tR 23.9 min). F6i was purified by HPLC using CH3CN/H2O (53:47, 3 mL/min) to afford 4 (5 mg, tR 49.6 min) and 10 (2 mg, tR 53.6 min). F6j was purified by HPLC using CH3CN/H2O (50:50, 3 mL/min) to afford 6 (2 mg, tR 22.5 min). F6k was purified by HPLC using CH3CN/H2O (65:35, 3 mL/min) to afford 1 (1 mg, tR 36.2 min) and 2 (6 mg, tR 39.0 min). F6l was purified by HPLC using CH3CN/H2O (68:32, 3 mL/min) to afford 11 (20 mg, tR 24.4 min). F6m was purified by HPLC using CH3CN/H2O (70:30, 3 mL/min) to afford 8 (1.5 mg, tR 48.9 min).
Austalide Y (
1): colorless oil;
−33 (
c 0.05, MeOH); UV (MeOH)
λmax 222, 270 nm; ECD (
c 2.24 × 10
−4 M, MeOH)
λmax (Δε) 264 (−3.69), 229 (+5.21), 212 (−6.49);
1H and
13C NMR data, see
Table 1; HRESIMS
m/
z 469.2202 [M + Na]
+ (calcd. for C
25H
34O
7Na
+, 469.2197).
3
β-Hydroxy-agathic acid (
9): colorless oil;
+66 (
c 0.06, MeOH); ECD (
c 5.7 × 10
−4 M, MeOH)
λmax 217 (4.05);
1H and
13C NMR data, see
Table 2; HRESIMS
m/
z 373.1991 [M + Na]
+ (calcd. for C
20H
30O
5Na
+, 373.1985).
3
β-Acetoxy-agathic acid (
10): colorless oil;
+58 (
c 0.03, MeOH); ECD (
c 5.1 × 10
−4 M, MeOH)
λmax 225 (3.78);
1H and
13C NMR data, see
Table 2; HRESIMS
m/
z 415.2096 [M + Na]
+ (calcd. for C
22H
32O
6Na
+, 415.2091).
3.5. α-Glucosidase Assay
The α-glucosidase inhibitory effect was assessed as follows. First, 0.2 U of α-glucosidase from Saccharomyces cerevisiae was purchased from Sigma-Aldrich (St. Louis, MO, USA), and was diluted in a 0.067 M phosphate buffer consisting of Na2HPO4·12H2O and KH2PO4 (pH 6.8). The assay was conducted in a 60 μL reaction system containing 20 μL of diluted enzyme solution, and 20μL of dimethyl sulfoxide (DMSO) or sample (dissolved in DMSO). After 10 min of incubation in 96-well plates at 37 °C, a 20 μL portion of 4 mM 4-nitrophenyl-α-d-glucopyranoside (PNPG) (Aladdin, Shanghai, China) was added as a substrate to start the enzymatic reaction. The plate was incubated for an additional 20 min at 37 °C, and the reaction was quenched by adding 60 μL of 0.2 M Na2CO3. The final concentrations of tested compounds were between 0.2 and 2 mM. The optical density (OD) was measured at an absorbance wavelength of 405 nm using a Microplate Reader (Tecan, Mannedorf, Switzerland). All assays were performed in three replicates, and acarbose (Aladdin, Shanghai, China) was used as the positive control.