Talaromarins A–F: Six New Isocoumarins from Mangrove-Derived Fungus Talaromyces flavus TGGP35

Six new isocoumarin derivative talaromarins A-F (1–6), along with 17 known analogues (7–23), were isolated from the mangrove-derived fungus Talaromyces flavus (Eurotiales: Trichocomaceae) TGGP35. Their structures were identified by detailed IR, UV, 1D/2D NMR and HR-ESI-MS spectra. The absolute configurations of new compounds were determined by the modified Mosher’s method and a comparison of their CD spectra with dihydroisocoumarins described in the literature. The antioxidant, antibacterial, anti-phytopathogenic and inhibitory activity against α-glucosidase of all the isolated compounds were tested. Compounds 6–11, 17–19 and 21–22 showed similar or better antioxidant activity than the IC50 values ranging from 0.009 to 0.27 mM, compared with the positive control trolox (IC50 = 0.29 mM). Compounds 10, 18, 21 and 23 exhibited strong inhibitory activities against α-glucosidase with IC50 values ranging from 0.10 to 0.62 mM, while the positive control acarbose had an IC50 value of 0.5 mM. All compounds showed no antibacterial or anti-phytopathogenic activity at the concentrations of 50 μg/mL and 1 mg/mL, respectively. These results indicated that isocoumarins will be useful to developing antioxidants and as diabetes control agents.


Results and Discussion
Compound 1 was isolated as a colorless oil. The molecular formula of 1 was established as C 17 6 and 111.5), one methoxyl group at δ C (56.6), two oxygenated methine groups at δ C (77.4 and 70.9), four methylene groups at δ C (35.7, 34.7, 28.0 and 21.0), two methyl groups at δ C (21.6 and 20.1). The above 1D NMR spectroscopic data indicated that 1 had an isocoumarin skeleton structure, and 1 is similar to penicimarin G (12) [16]. The major difference was the presence of an acetoxy group at [(δ C 171.1 (C), 21.6 (CH 3 ) and δ H 2.04, (s)] in 1, indicating that the hydroxyl group in 12 was replaced by an acetoxy group in 1. The 1 H-1 H COSY correlations of CH 2 (2 )-CH 2 (3 )−CH(4 )−CH 3 (5 ), combined with the HMBC correlations from H-5 to C-4 /C-3 ( Figure 2), confirmed the acetoxy group connected at C-4 in 1, and the planar structure of 1 was determined ( Figure 1).  The absolute configurations of C-3 and C-4 in 1 were determined by chemical hydrolysis, modified Mosher's method and a comparison of CD spectra with dihydroisocoumarins described in the literature [28,34]. The major hydrolysis product (1a) of 1 was obtained with K 2 CO 3 and anhydrous ethanol at 28 • C for 1.5 h (Figure 3), and 1a showed the same planar structure with 12 [16]. The modified Mosher's method was used to determine the configuration of C-4 for 1a. The differences in 1 H NMR chemical shifts of 1a between (S)-and (R)-MTPA esters (∆δ = δ S − δ R ) were calculated to assign the absolute configuration of C-4 to be R (Figure 4), the same as 12 [16]. The negative cotton effect at 266 nm suggested the R configuration at C-3 ( Figure 5), by comparison with data for dihydroisocoumarins described in the literature [34]. Thus, the structure of 1 was determined and named talaromarin A.  Compound 2 was isolated as a white powder. The molecular formula was deduced to be C 15 H 20 O 5 on the basis of HR-ESI-MS spectrum, implying six degrees of unsaturation. According to the IR spectrum, the hydroxyl group (3414 cm −1 ) and aromatic rings (1668, 1619, 1586, 1502 and 1442 cm −1 ) were observed. The 1 H and 13 C NMR spectroscopic data (Tables 1 and 2) revealed that 2 also belonged to the isocoumarin class and had a similar structural relationship to penicimarin M [17], except for the presence of one oxygenated methine at [δ H 3.80 (m), δ C 67.9 (CH)] for C-4 , and the absence of a carbonyl group at δ C 208.4 (C) in 2. The above results suggested a carbonyl group in penicimarin M was replaced by an oxygenated methine group in 2. Furthermore, the 1 H-1 H COSY correlations of H-5 to H-4 and H-3 to H-2 /H-4 , and the HMBC correlations from the methyl H-5 to C-3 /C-4 established the oxygenated methine at C-4 ( Figure 2). The absolute configuration of C-4 was determined as R by Mosher's method [28] (Figure 4). The negative cotton effect at 265 nm suggested the R configuration at C-3 ( Figure 5), by comparison with data for dihydroisocoumarins described in the literature [34]. Thus, the absolute configuration of 2 was established as 3R,4 R and named talaromarin B. Compound 3 was isolated as a yellow oil, with the molecular formula C 16 H 22 O 6 (six degrees of unsaturation) by the HR-ESI-MS spectrum. The IR spectrum indicated that 3 had hydroxyl group (3415 cm −1 ) and aromatic ring (1638, 1618 and 1384 cm −1 ). The 1 H, 13 C NMR data (Tables 1 and 2) and HR-ESI-MS data revealed that 3 closely resembled those of 2, the main differences were the presence of a methoxyl group at [δ H 3.89 (s), δ C 61.8 (CH 3 )] in 3, and an aromatic proton signal at δ H 6.62 (d, J = 8.0 Hz) was absented in 3. Moreover, the chemical shift of C-5 at δ C (117.1) in 2 was downfield-shifted to δ C (147.7) in 3. The HMBC correlations from 8-OCH 3 to C-8, 7-OCH 3 to C-7 and H-6 to C-8/C-7/C-4a ( Figure 2), indicated the additional methoxyl group was attached to C-8 and the hydroxyl group was connected to C-5 ( Figure 2). The absolute configurations of C-3 and C-4 were determined to be the same R by comparison with CD data described in the literature [34] and Mosher's method [28] (Figures 4 and 5). Thus, the structure of 3 was determined and named talaromarin C.
Compound 4 was isolated as a yellow oil, and the molecular formula was established as C 16 H 22 O 5 (six degrees of unsaturation) on the basis of its HR-ESI-MS spectrum. The IR spectrum of 4 displayed absorption bands for hydroxyl (3475 cm −1 ), carbonyl (1706 cm −1 ) and aromatic (1637 and 1617 cm −1 ) groups. The 1 H and 13 C NMR spectroscopic data (Tables 1 and 2) suggested that 4 was very similar to those of 2, the only difference was the presence of a methoxyl group at [δ H 3.96 (s) and δ C 61.7 (CH 3 )] in 4. The location of the methoxyl groups at C-7 and C-8 were established by HMBC correlations from 7-OCH 3 to C-7, 8-OCH 3 to C-8, H-6 to C-8/C-4a and H-5 to C-7/C-8a/C-4 ( Figure 2). The 1 H-1 H COSY, HMQC, and HMBC spectra established the complete assignment for 4 ( Figure 2). Mosher's method was tried to determine the absolute configuration of C-4 in 4 [28]; unfortunately, the reaction failed. The negative cotton effect at 259 nm suggested the R configuration at C-3 ( Figure 5) [34]. Thus, the structure of 4 was determined and named talaromarin D.
Compound 5 was isolated as a colorless oil and had the molecular formula of C 16 H 20 O 5 as determined by HR-ESI-MS and NMR data, requiring seven degrees of unsaturation. The presence of an aromatic ring (1638, 1617 cm −1 ) was observed in the IR spectrum. The 1 H and 13 C NMR data (Tables 1 and 2) of 5 were structurally similar to those of 4, except for the presence of a ketone carbonyl carbon at δ C 208.7 (C) and the absence of one oxygenated methine carbon at [δ C 68.1 (CH), δ H 3.83 (m)] at C-4 in 5, indicating that the oxygenated methine group in 4 was replaced by a carbonyl group for C-4 in 5. The HMBC correlations from H-3 to C-1 , H-5 to C-4 /C-3 further confirmed 5 with a carbonyl unit at C-4 ( Figure 2). The whole structure was further determined by the 2D NMR spectra (Figure 2). The absolute configuration of C-3 was determined as R by CD spectra ( Figure 5) [34], and 5 was named talaromarin E.
Compound 6 was isolated as a yellow oil. The molecular formula of 6 was established as C 14 H 16 O 5 (seven degrees of unsaturation) on the basis of its HR-ESI-MS data. The IR spectrum of 6 showed the hydroxyl group at 3461 and 3407 cm −1 and the aromatic rings at 1736 and 1671 cm −1 . The 1 H and 13 C NMR spectroscopic data (Tables 1 and 2) revealed that 6 was an isocoumarin derivative, with a similar structural relationship to peniciisocoumarin D (18), the obvious difference was that 6 lacked a methoxy group at C-8. The methoxy group (8-OMe) in 18 was replaced by a hydroxy group (8-OH) in 6, which was supported by the appearance of a hydrogen-bonded hydroxyl group at δ H 11.00 (s). The HMBC correlations from the hydroxyl group 8-OH to C-8a/C-8/C-7 further confirmed the 8-OH was connected at C-8 (δ C 149.1) (Figure 2). The 1 H-1 H COSY, HMQC, and HMBC spectra determined the complete assignment for 6 ( Figure 2). The absolute configuration of C-3 was determined as R by CD spectroscopy ( Figure 5) [34] and 6 was named talaromarin F.
The plausible biosynthetic pathways of compounds 1-23 were also proposed (Scheme 1). Isocoumarins were originated from the acetate-malonate or the polyketide synthase (PKS) pathway [10]. Peniciisocoumarin C (8), penicimarin C (13) and peniciisocoumarin A (14) would be biosynthesized from malonyl-CoA and acetyl-CoA and can be considered as intermediates which would be transformed to other isolated compounds from the fungus TGGP35. Compound 8 would be transformed to 7, 9 and 20-23 by condensation, aromatization, esterification, dihydroxylation, methoxylation reaction and so on. Compound 13 would be transformed to 10 by dehydroxylation, demethoxylation and dehydroxylation reaction.  (Table 3). The preliminary structure-activity relationship of the isolated isocoumarins was discussed. The substitution site, orientation of hydroxyl and methoxy groups on the skeleton of isocoumarins, and the substitution of different groups by side chain C-4 can affect their antioxidant activity. Compound 2 which possesses a hydroxyl group on C-8, showed better antioxidant activity than that of 3 and 4, indicating that the chelated hydroxyl group at C-8 is important in enhancing antioxidant activity. Compound 6 possesses two hydroxyl groups at C-7 and C-8, which showed higher antioxidant activity than 16 (only one hydroxyl group on C-8), indicating that the hydroxyl group at C-7 is an important antioxidant activity site. Compounds 17 and 18 possess a ketone group at C-4 , which showed higher antioxidant activity than 12 and 13, which have an oxygenated methine at C-4 , suggesting that the substitution of different groups by side chain at C-4 can affect antioxidant activity. Compound 22 possesses a hydroxyl group at C-8, showed higher antioxidant activity than 23, suggesting that the chelated hydroxyl group at C-8 is important in enhancing antioxidant activity. Furthermore, compounds 7, 8, 19 and 21 show antioxidant activities, which may be due to the existence of a chelated hydroxyl group.
All compounds were tested for their antibacterial activities against Staphylococcus. aureus, Escherichia coli, S. epidermidis and Pseudomonas aeruginosa; however, all compounds showed no antibacterial activity at the concentration of 50 µg/mL. All compounds showed no biological activity against five phytopathogens (Colletotrichum asianum, C. acutatum, Fusarium oxysporum, Pyricularia oryzae and Curvularia australiensis) at the concentration of 1 mg/mL.

General Experimental Procedures
Optical rotations were measured on a JASCO P-1020 digital polarimeter (JASCO, Tokyo, Japan). IR spectra were recorded on a Thermo Nicolet 6700 (using KBr disks) spectrophotometer. UV spectra were measured on a PERSEE TU-1990 spectrophotometer. CD spectra were recorded on a Mos-450 spectrometer and 1D and 2D NMR spectra were recorded on a Bruker AV spectrometer (400 MH Z for 1 H and 100 MH Z for 13 C) and a JNM-ECZS spectrometer (600 HM Z for 1 H and 125 MH Z for 13 C). HR-ESI-MS spectra were obtained on a Q-TOF Ultima Global GAA076 LC mass spectrometer. ESI-MS spectra were recorded on a MAT-95-MS mass spectrometer. HPLC were used for the Agilent 1100 prep-HPLC system with an Agilent C18 analytical (9.4 × 250 mm, 5 µm) HPLC column. Silical gel (200−300 mesh, Qingdao Marine Chemical Factory, Qingdao, China) were used for column chromatography (CC). Sephadex LH-20 gel column (GE Healthcare, Bio-Sciences Corp, Piscataway, NJ, USA) were used for CC. Biological activities were tested on an ultraclean workbench (Suzhou Sujing Company, Suzhou, China) and the biological activities' results were tested with a full wavelength multifunctional microplate reader (Bio-Tek Instruments, Winooski, VT, USA). Methanol, ethyl acetate, petroleum ether, chloroform, dimethyl sulfoxide and other conventional chemical reagents were used in the experimental operation (Guangzhou Xilong Chemical Reagent Factory, Guangzhou, China).

Fungal Materials
The fungus TGGP35 was isolated from the stem of the mangrove plant Acanthus ilicifolius, which were collected in the Dongzhai Port, Haikou, Hainan Province in August, 2015. The fungus was identified according to its morphological characteristics and a molecular biological protocol by 18S rRNA amplification and sequencing of the ITS region. The sequence data have been submitted to GeneBank, with accession number MT071116, and the fungal strain was identified as Talaromyces flavus (Eurotiales: Trichocomaceae). The strains have been stored in the Key Laboratory of Tropical Medicinal Resources Chemistry of the Ministry of Education, School of Chemistry and Chemical Engineering, Hainan Normal University (PDA medium, stored at 4 • C).

Fermentation, Extraction and Isolation
The seed culture was prepared in potato liquid medium (6 g sea salt and 10 g peptone in 2 L of potato infusion, in 1 L × 4 erlenmeyer flasks each containing 500 mL seed medium), and incubated on a rotary shaker (170 rpm) for 4 days at 28 • C. In total, 20 mL seed culture was then transferred into 1 L erlenmeyer flasks with solid rice medium with a total of 100 bottles of fermentation (each flask contained 60 mL rice, 0.6 g sea salt and 1.0 g peptone) at 28 • C for 4 weeks. The whole rice fermented material was extracted three times with EtOAc, and then concentrated in vacuo to yield crude extracts (120.8 g). The total crude extracts were subjected to silica gel column chromatography (CC) eluted with petroleum ether/EtOAc (v/v, gradient 100:0-0:100) and EtOAc/MeOH (v/v, gradient 100:0-70:30) to generate fifteen fractions (Fr. A-Fr. L). Fr. F (20.5 g) was fractionated by silica gel CC (200-300 mesh) using a gradient elution of petroleum ether/EtOAc system (7:1-0:1) to obtain ten fractions (Fr. F1-Fr.

Preparations of the (S)-and (R)-MTPA Esters of Compounds 1, 2 and 3
Compound 1 was hydrolyzed in anhydrous ethanol solution for 90 min with potassium carbonate in an equivalent ratio of 1:2; the mixed product after hydrolysis was purified by semi-preparative HPLC (MeOH-H 2 O, 70:30, v/v) to obtain 1a. The preparation of (S)-and (R)-MTPA ester derivatives of 1a, 2 and 3 was performed as described previously [28].
Hydrolysate of 1 (1a): 1  The antioxidant activity assay was based on the reported methods [15]. The assay was performed on a 96-well microplate, the reaction was initiated by adding 10 µL of sample solution to 200 µL of ABTS working solution. All test group gradients (including positive control) were of 2.0, 1.0, 0.5, 0.25 mg/mL, respectively. PBS buffer was used as the blank control, DMSO as the negative control, and trolox as the positive control (IC 50 = 0.29 mM). The antioxidant effect was evaluated by a full wavelength multifunctional microplate reader measurement at 734 nm. The inhibition rate of each sample was calculated according to the following formula: inhibition rate = [(A blank − A compound )/A blank ] × 100%. Finally, the SPSS software was used to calculate the IC 50 value.

Antibacterial Activity
All compounds were determined against four pathogenic bacteria: Staphylococcus aureus (ATCC 25923), Escherichia coli (ATCC 25922), S. epidermidis (ATCC 17749) and Pseudomonas aeruginosa (ATCC 17749). The concentration value of the test group and positive control was 1mg/mL by the microplate assay method [28]. The antibacterial effect was evaluated by a full wavelength multifunctional microplate reader measurement at 630 nm; the broth medium containing pathogenic bacteria was used as the blank group and DMSO as the negative control, ciprofloxacin was used as the positive control. The positive control ciprofloxacin showed antibacterial activities against four pathogenic bacteria S. aureus, E. coil, S. epidermidis and P. aeruginosa with the MIC values of 0.097, 0.78, 0.195 and 0.78 µg/mL, respectively.

Anti-Phytopathogenic Activity
All compounds were tested against five plant pathogens (Colletotrichum asianum, C. acutatum, Fusarium oxysporum, Pyricularia oryzae and Curvularia australiensis) by disk method [40]. DMSO was used as a negative control, carbendazim as a positive control.
The concentration values of all test groups, negative control and positive control were 1 mg/mL; the anti-phytopathogenic results were recorded on a vernier caliper.

Inhibitory Activity against α-Glucosidase
The α-glucosidase inhibitory activity of the tested compounds was determined using the method in [17], with modifications for carrying it out in 96-well plates. The initial concentration of all test samples (including positive control and negative control) was 1 mg/mL, the optimized method was a mixture of 0.1 mM potassium phosphate buffer (pH = 6.8, 0.5 mL) and 10 mg/L α-glucosidase (100 µL), the testing sample (0.5 mL) was incubated at 37 • C for 5 min, and the 2.5 mM (4-nitrophenyl-β-D-glucopyranoside) PNPG (0.5 mL) was added, followed by mixing. The reaction was carried out at 37 • C for 15 min and then stopped by adding 0.2 M solution of Na 2 CO 3 (0.75 mL). The inhibitory activity against α-Glucosidase was evaluated by a full wavelength multifunctional microplate reader measurement at 405 nm. Finally, inhibition rate = [(A control − A compound )/A control ] × 100%. The SPSS software was used to calculate the IC 50 value. DMSO was used as the negative control and acarbose was used as the positive control (IC 50 = 0.5 mM).