Brominated Azaphilones from the Sponge ... - ACS Publications

Article Cite This: J. Nat. Prod. XXXX, XXX, XXX−XXX

pubs.acs.org/jnp

Brominated Azaphilones from the Sponge-Associated Fungus Penicillium canescens Strain 4.14.6a

Marian Frank,† Rudolf Hartmann,‡ Malte Plenker,‡ Attila Mań di,§ Tibor Kurtań ,§ Ferhat Can Ö zkaya,⊥ Werner E. G. Müller,∥ Matthias U. Kassack,# Alexandra Hamacher,# Wenhan Lin,¶ Zhen Liu,*,† and Peter Proksch*,†

Downloaded via NOTTINGHAM TRENT UNIV on July 30, 2019 at 13:43:02 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

Institute of Pharmaceutical Biology and Biotechnology, Heinrich-Heine-Universität Düsseldorf, Universitätsstrasse 1, 40225 Düsseldorf, Germany ‡ Institute of Complex Systems: Strukturbiochemie, Forschungszentrum Juelich, Wilhelm-Johnen-Strasse, 52428 Juelich, Germany § Department of Organic Chemistry, University of Debrecen, Egyetem tér 1, Debrecen 4032, Hungary ⊥ ̇ ̇ Faculty of Fisheries, Izmir Katip Ç elebi University, Ç iğli, 35620 Izmir, Turkey ∥ Institute of Physiological Chemistry, Universitätsmedizin der Johannes Gutenberg-Universität Mainz, 55128 Mainz, Germany # Institute for Pharmaceutical and Medicinal Chemistry, Heinrich-Heine-Universität Düsseldorf, Universitätsstrasse 1, 40225 Düsseldorf, Germany ¶ State Key Laboratory of Natural and Biomimetic Drugs, Peking University, Beijing 100191, China S Supporting Information *

ABSTRACT: The fungus Penicillium canescens was isolated from the inner tissue of the Mediterranian sponge Agelas oroides. Fermentation of the fungus on solid rice medium yielded one new chlorinated diphenyl ether (1) and 13 known compounds (2−14). Addition of 5% NaBr to the rice medium increased the amounts of 4−6, while lowering the amounts of 8, 12, and 14. Furthermore, it induced the accumulation of 17 and two new brominated azaphilones, bromophilones A and B (15 and 16). Compounds 15 and 16 are the first example of azaphilones with the connection of a benzene moiety and the pyranoquinone core through a methylene group. The structures of the new compounds were elucidated based on the 1D and 2D NMR spectra as well as on HRESIMS data. The absolute configuration of the condensed bicyclic moiety of 15 and 16 was determined by sTDA ECD calculations. Compound 16 exhibited moderate cytotoxicity against the mouse lymphoma cell line L5178Y (IC50 8.9 μM), as well as against the human ovarian cancer cell line A2780 (IC50 2.7 μM), whereas the stereoisomer 15 was considerably less active.

M

interesting biological activities including cytotoxic,9 antibacterial,10 antiviral,11 antifungal,11 antimalarial,12 and anti-inflammatory activities.13 While chlorination of naturally occurring azaphilones is rather common,14 brominated analogues are only seldomly observed in OSMAC experiments.15,16 The OSMAC approach has also been implemented in this study on the sponge-derived fungus Penicillium canescens strain 4.14.6a that was isolated from the Mediterranean Sponge Agelas oroides. This fungus is a known producer of several bioactive secondary metabolites, including the antifungal compounds canescin, griseofulvin, curvulinic acid, and several tetrapeptides.17−19 The EtOAc extract of a solid rice medium fermentation of P. canescens showed promising cytotoxic activity against the mouse lymphoma cell line L5178Y (100% inhibition at a dose of 10 μg/mL) and was thus

arine-derived fungi are prominent producers of novel bioactive compounds and drug lead structures.1,2 However, under laboratory conditions fungi often transcribe only a fraction of their biogenetic gene clusters, whereas other genes are silent. Thus, only a part of the potential chemical diversity of natural products can be obtained from these cultures.3,4 Numerous strategies aim at the activation of silent biogenetic gene clusters in order to enlarge the diversity of metabolites. One of these strategies is the OSMAC approach, which aims at diversifying media and culture conditions in order to generate clues that induce silent biogenetic gene clusters and thus the accumulation of cryptic metabolites.5 Azaphilones are an example of such fungal secondary metabolites that are controlled by silent gene clusters and can be induced under certain conditions such as OSMAC approaches,6 cocultivation,7 and epigenetic modification.8 Azaphilones are polyketides with a bicyclic core and a conjugated chromophore and exhibit a wide array of © XXXX American Chemical Society and American Society of Pharmacognosy

Received: February 16, 2019

A

DOI: 10.1021/acs.jnatprod.9b00151 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

(δC 55.5, δH 3.76) and the deshielded C-4′ (+4.1 ppm) in 1. The protons of this additional methoxy group exhibited an HMBC correlation to C-4′ and ROESY correlations to H-3′ (δH 6.29) and H-5′ (δH 6.35), indicating its location at C-4′. Detailed analysis of the 2D NMR spectra of 1 revealed that the remaining structure of 1 was identical to that of 2. Thus, compound 1 was elucidated as the 4′-O-methyl derivative of 2. Further known compounds that were isolated from the fermentation of P. cancescens on solid rice medium included two xanthones, griseoxanthone C (3)21 and norlichexanthone (4),22 three benzophenones, griseophenones C, B, and G (5− 7),23 two spirocyclic polyketides, griseofulvin (8)23 and dechlorogriseofulvin (9),24 two curvulinic acid derivatives, methylcurvulinate (10)25 and 3′-O-methylmethylcurvulinate (11),26 the tetraketide mycotoxin penicillic acid (12),27 the meroterpenoid citreohybridonol (13),28 and the diketopiperazine alkaloid piscarinine B (14).29 In an OSMAC approach, 5% NaBr was added to the solid rice medium. This concentration of NaBr was chosen based on previous experiments with other marine-derived fungi.30 The metabolite profile of P. cancescens following fermentation in the presence of NaBr differed markedly from that obtained following cultivation only on rice (Table 2). The amounts of

further investigated, which resulted in the isolation of 14 natural products (1−14) including a new diphenyl ether (1). In an attempt to expand the metabolic profile, the fungus was cultivated on solid rice medium following addition of 5% NaBr. Investigation of the fungal extract following fermentation on NaBr-spiked rice medium yielded two new brominated azaphilones (15 and 16) and a known xanthone derivative (17) in addition to the previously isolated compounds (1− 14). Compounds 15−17 were not detected when the fungus was cultivated in the absence of NaBr. Herein, we describe the structure elucidation of the new metabolites (1, 15, and 16) as well as the biological activities of the isolated compounds against tumor cells.

Table 2. Relative Intensities at 235 nm of Selected Compounds in Cultures of P. canescens Grown on Solid Rice Medium (control) vs Cultures Grown in the Presence of NaBr (n = 5) compound 4 5 6 8 12 14 15 16 17

■

RESULTS AND DISCUSSION The molecular formula of 1 was established as C18H19ClO7 by HRESIMS data with nine degrees of unsaturation. The 1H and 13 C NMR data of 1 (Table 1) resembled those of 2, a coisolated known chlorinated diphenyl ether.20 The obvious difference was the appearance of an additional methoxy group

control (mAU·min) n.d.a 3.7 n.d. 146 409 59 n.d. n.d. n.d.

± 0.1 ± 13 ± 15 ±5

5% NaBr (mAU·min) 42 54 8.8 11 101 23 0.46 0.36 19

± ± ± ± ± ± ± ± ±

2 2 0.5 2 11 1 0.05 0.10 6

a

1

Table 1. H and

13

C NMR Data for Compound 1

position

δC

1 2 3 4 5 6 7 1′ 2′ 3′ 4′ 5′ 6′ 7′ 4-OMe 6-OH 7-OMe 2′-OMe 4′-OMe

101.2, C 156.4, C 106.7, C 160.1, C 95.6, CH 162.2, C 170.8, C 138.8, C 149.7, C 98.6, CH 154.9, C 107.2, CH 129.6, C 17.4, CH3 56.4, CH3 52.3, CH3 56.5, CH3 55.5, CH3

n.d. = not detected in the extract. Compounds 4 and 6 were not detected in control cultures due to their low amounts and partial overlap with 14 and 8, respectively.

a

δH, m (J in Hz)

compounds 4−6 increased strongly in the presence of NaBr, whereas the amounts of 8, 12, and 14 decreased. In addition, two new brominated azaphilone derivatives (15 and 16) as well the known 1,3,5,6-tetrahydroxy-8-methylxanthone (17)31 were obtained from solid rice cultures containing NaBr but were not detected when the fungus was grown only on rice. Compound 15 was isolated as a reddish-brown, amorphous powder. The molecular formula was determined as C33H35BrO13 with 16 degrees of unsaturation based on the HRESIMS data. The 1H NMR data of 15 (Table 3) showed eight olefinic protons at δH 7.35 (H-14), 6.98 (H-11), 6.82 (H12), 6.80 (H-13), 6.40 (H-10), 6.24 (H-4), 6.19 (H-19), and 6.07 (H-15) and a methoxy group at δH 3.79 (OMe-20), in addition to three methyl groups at δH 2.57 (Me-25), 1.57 (Me9), and 1.17 (Me-6′). The 13C NMR data of 15 (Table 3) displayed five carbonyls at δC 203.2 (C-24), 201.6 (C-8), 186.7 (C-6), 171.0 (C-1′), and 167.6 (C-16), as well as 16 olefinic carbons at δC 162−107, accounting for 13 degrees of unsaturation. Thus, compound 15 was suggested to be tricyclic. The HMBC correlations from H-1ab (δH 4.49 and

6.34, s

6.29, d (2.7) 6.35, d (2.7) 2.29, s 3.90, s 11.57, s 3.74, s 3.48, s 3.76, s

a

Measured in CDCl3 (1H at 600 MHz and 13C at 150 MHz). B


Journal of Natural Products Table 3. 1H and

13

Article

C NMR Data of Compounds 15 and 16 15a

position

δC, type

δH (J in Hz)

1

70.3, CH2

3 4 4a 5 6 7 8 8a 9 10 11

161.2, C 107.3, CH 150.3, C 110.9, C 186.7, C 85.0, C 201.6, C 54.7, C 24.0, CH3 129.0, CH 136.2, CH

12

139.4, CH

13

136.1, CH

14

144.4, CH

15 16 17

124.5, CH 167.6, C 40.9, CH2 3.59, d (14.4) 3.52, d (14.4) 125.2, C 109.2, CH 6.19, s 148.5, C 133.2, C 145.5, C 124.0, C 203.2, C 33.2, CH3 2.57, s 171.0, C 42.6, CH2 2.64, dd (15.1, 7.1) 2.61, dd (15.1, 5.6) 68.4, CH 4.27, m 45.5, CH2 1.73, dt (13.9, 3.7) 1.61, dt (13.9, 9.0) 67.2, CH 4.02, m 24.2, CH3 1.17, d (6.1) 56.3, CH3 3.79, s

18 19 20 21 22 23 24 25 1′ 2′

3′ 4′ 5′ 6′ 20-OMe 22-OH

4.49, d (11.3) 4.12, d (11.3) 6.24, s

1.57, s 6.40, d (15.1) 6.98, dd (15.1, 10.2) 6.82, dd (15.0, 10.2) 6.80, dd (15.0, 10.2) 7.35, dd (15.3, 10.2) 6.07, d (15.3)

16b δC, type 71.4, CH2 161.1, C 106.5, CH 150.2, C 115.9, C 185.8, C 82.1, C 203.6, C 53.2, C 22.2, CH3 128.6, CH 135.9, CH 139.2, CH 136.0, CH 144.3, CH

δH (J in Hz) 4.64, d (11.4) 3.98, d (11.4) 6.43, s

Figure 1. COSY and key HMBC correlations of compound 15.

H-4′ab/H-5′ (δH 4.02)/Me-6′ as well as the HMBC correlations from H-2′ab to C-1′ and the long-range 4JHMBC correlation from Me-9 to C-1′ indicated the presence of a 3,5-dihydroxyhexanoyl side chain and its attachment at C7 through an ester bond. In addition, the HMBC correlations from H-19 to C-21 (δC 133.2) and C-23 (δC 124.0), from Me25 to C-23 and C-24, and from the methoxy group to C-20 (δC 148.5), the weak 2J-HMBC correlations from H-19 to C-18 (δC 125.2) and C-20, the long-range 4J-HMBC correlation from H-19 to C-22 (δC 145.5), and the ROESY correlation between H-19 and the methoxy group established a benzene moiety with a methoxy group at C-20, two hydroxy groups at C-21 and C-22, and an acetyl group at C-23. Moreover, the HMBC correlations from H-17ab (δH 3.59 and 3.52) to C-1 (δC 70.3), C-4a, C-8, C-8a, C-18, C-19, and C-23 indicated the connection of the benzene moiety and the pyranoquinone core through a methylene group (C-17). The bromine substituent is connected to C-5 based on the molecular formula and the chemical shift of C-5 (δC 110.9).15,16 Thus, the planar structure of 15 was elucidated as shown, representing a new brominated azaphilone derivative, for which the name bromophilone A is proposed. In the ROESY spectrum of 15, the correlation between the axial Me-9 and the axial H-1b (δH 4.12) indicated that Me-9 and H-1b were on the same side of the ring, while the C-17 methylene group was on the other side, implying a (7R*,8aR*) relative configuration for 15. The relative configuration of the C-3′ and C-5′ chirality centers in the 3,5-dihydroxyhexanoyl side chain was assigned as (3′R*,5′R*) by NOE correlations and coupling constants using Newman projections and density functional theory (DFT) conformers of (7R,8aR,3′R,5′R)-15. The large coupling constants (9.0 Hz) of H-4′b with H-3′ and H-5′ and the small coupling constants (3.7 Hz) of H-4′a with H-3′ and H-5′ together with the NOE relationships between H-4′a/H-3′ and H-4′a/H-5′ indicated that H-4′a was in gauche position from H-3′ and H-5′, whereas H-4′b was in anti position from H-3′ and H-5′ (Figure 2a). The (3′R*,5′R*) relative configuration of the side chain could not be correlated with the (7R*,8aR*) relative configuration of the bicycle core in 15, which enabled four stereoisomers, namely, (7R,8aR,3′R,5′R), (7S,8aS, 3′S,5′S), (7R,8aR,3′S,5′S), and (7S,8aS,3′R,5′R). For the configurational assignment of 15, conformational analysis and the sTDA (simplified Tamm−Dancoff approximation) approach32−35 were applied on the arbitrarily chosen (7R,8aR,3′R,5′R) and (7R,8aR,3′S,5′S) stereoisomers. The sTDA method was developed for large molecules and/or for compounds with a large number of electronic circular dichroism (ECD) transitions, the TDDFT-ECD calculation of which would not be feasible at an advanced level. The developers demonstrated its applicability on helicenes and fullerenes.34,35 For natural products, only UV calculation of

1.34, s 6.52, d (15.2) 7.05, dd (15.2, 10.7) 6.86, dd (14.9, 10.7) 6.81, dd (14.9, 10.7) 7.36, dd (15.2, 10.7) 6.07, d (15.2)

124.3, CH 167.2, C 39.0, CH2 3.56, d (14.6) 3.29, d (14.6) 124.0, C 107.5, CH 6.24, s 148.5, C 133.4, C 145.2, C 124.6, C 202.8, C 32.9, CH3 2.53, s 171.4, C 41.9, CH2 2.48, dd (15.4, 7.0) 2.43, dd (15.4, 5.9) 68.1, CH 4.10, m 45.3, CH2 1.56, dt (14.0, 3.7) 1.50, dt (14.0, 8.9) 67.1, CH 3.95, m 24.0, CH3 1.13, d (6.2) 56.1, CH3 3.81, s 8.19, s

a

Recorded at 700 MHz (1H) and 175 MHz (13C) in acetone-d6. Recorded at 750 MHz (1H) and 187.5 MHz (13C) in acetone-d6.

b

4.12) to C-3 (δC 161.2), C-4a (δC 150.3), C-8, and C-8a (δC 54.7), from H-4 to C-3, C-5 (δC 110.9), and C-8a, and from Me-9 to C-6, C-7 (δC 81.0), and C-8, as well as the long-range 4 J-HMBC correlation from H-4 to C-6, established the pyranoquinone bicyclic core with a methyl group at C-7 and two carbonyls at C-6 and C-8, respectively (Figure 1). The COSY correlations between H-10/H-11/H-12/H-13/H-14/ H-15 together with the HMBC correlations from H-15 to C-16 and from H-10 to C-3 and C-4 indicated a hepta-2,4,6-trienoic acid side chain at C-3. The three double bonds are E configured as indicated by the typical coupling constants of the olefinic protons, which are in the range 15.0−15.3 Hz (Table 3). The COSY correlations between H-2′ab/H-3′ (δH 4.27)/ C



Article

Figure 2. Lowest-energy CAM-B3LYP/TZVP PCM/MeOH conformer and characteristic NOE correlations of (7R,8aR,3′R,5′R)-15 (a) and (7S,8aR,3′R,5′R)-16 (b).

indigo34 and ECD calculations of agathisflavone36 have been described, which, however, do not really have large molecular weights requiring the simplified approach. Merck molecular force field (MMFF) conformational searches of (7R,8aR,3′R,5′R)- and (7R,8aR,3′S,5′S)-15 resulted in 1317 and 1443 conformers in a 21 kJ/mol energy window, respectively. These initial conformers were then reoptimized at the AM1 semiempirical level to reduce the number of the conformers37 and then at the CAM-B3LYP/TZVP38 PCM/ MeOH level. In (7R,8aR,3′R,5′R)-15, the orientations of the C-7 and C-8a substituents are governed by the intramolecular hydrogen-bonding of the 3′-OH with an acetyl or hydroxy group of the aryl moiety. In contrast, the 3′-OH of the dihydroxyhexanoyl side chain formed an intramolecular hydrogen bond with the C-6 carbonyl oxygen in the lowestenergy conformer (72.6%) of (7R,8aR,3′S,5′S)-15 resulting in a different orientation of the C-7 and C-8a substituents. sTDAECD calculations were performed for conformers above 1% population at the CAM-B3LYP, LC-BLYP,39 and ωB97X40 levels, for which the required single-point calculations were performed with the same functionals and the TZVP basis set. The Boltzmann-averaged ECD spectra of both diastereomers gave moderate agreement with the experimental spectrum, allowing elucidation of the absolute configuration at C-7 and C-8 as (7R,8aR), whereas no conclusion could be reached about the C-3′ and C-5′ absolute configurations (Figure 3). The computed ECD spectra also showed that the different orientations of the C-7 and C-8a substituents did not result in different ECD spectra of the conformers, suggesting that the ECD transitions are governed by the unsaturated bicycle and the conjugating C-3 side chain. Because the sTDA-ECD method is a relatively new and less validated approach, TDDFT-ECD calculations41,42 (Figure 3) were also performed on the low-energy CAM-B3LYP/TZVP PCM/MeOH conformers of (7R,8aR,3′R,5′R)-15 at various levels (B3LYP, BH&HLYP, CAM-B3LYP, and PBE0 functionals with the TZVP basis set and PCM for MeOH) to confirm the results of the sTDA-ECD calculations. The well-established TDDFTECD method performed better in the 250−300 nm spectral range, but the overall agreement with the experimental ECD spectrum and the conclusion were the same, verifying the applicability of the sTDA-ECD method for azaphilone derivatives. The UV spectrum and 1H NMR data of bromophilone B (16) were comparable to those of 15 (Table 3). The molecular formula of 16 was identical to that of 15 as determined by HRESIMS. Detailed analysis of the 2D NMR spectra of 16 revealed that it shared the same planar structure as 15. The major changes in the 13C NMR data of 16 were the shielded signals of C-7 (−2.9 ppm), C-17 (−1.9 ppm), and C-9 (−1.8 ppm) and the deshielded signals of C-5 (+5.0 ppm) and C-8

Figure 3. Comparison of the experimental ECD spectrum of 15 in MeOH (black line) with the TDDFT-B3LYP/TZVP PCM/MeOH spectrum of (7R,8aR,3′R,5′R)-15 computed for the CAM-B3LYP/ TZVP PCM/MeOH conformers (orange line); the sTDA-CAMB3LYP spectrum of (7R,8aR,3′R,5′R)-15 computed from the CAMB3LYP/TZVP single-point data, level of optimization: CAM-B3LYP PCM/MeOH (red dashed line); and the sTDA-ωB97X spectrum of (7R,8aR,3′S,5′S)-15 computed from the ωB97X/TZVP single-point data, level of optimization: CAM-B3LYP PCM/MeOH (blue dashed line). Computed spectra are Boltzmann statistics of the low-energy (≥1%) conformers.

(+2.0 ppm) when compared to 15. In the ROESY spectrum of 16, the axial Me-9 (δH 1.34) showed correlations to H-17b (δH 3.29) of the axial methylene group and H-19 (δH 6.24) of the phenyl group rather than to H-1b as observed for 15, indicating that Me-9 and the C-17 methylene group were on the same side of the ring in 16 (Figure 2b). These NOE correlations were also checked and found feasible on the computed conformers of (7S,8aR,3′R,5′R)-16, which allowed determining the (7S*,8aR*) relative configuration for 16. The similar carbon chemical shifts and coupling constants in the 3,5-dihydroxyhexanoyl side chain from C-1′ to C-6′ of compounds 15 and 16 suggested that both compounds shared the same (3′R*,5′R*) relative configuration at C-3′ and C-5′. Since the two blocks of chirality centers could not be correlated, four stereoisomers are possible for 16, which are (7S,8aR,3′R,5′R), (7R,8aS,3′S,5′S), (7S,8aR,3′S,5′S), and (7R,8aS,3′R,5′R). The above computational protocol was also performed on (7S,8aR,3′R,5′R)-16 and (7S,8aR,3′S,5′S)16, and the initial MMFF conformational search showed similar conformational flexibility with 1134 and 1092 conformers, respectively. The sTDA calculations for the CAMB3LYP/TZVP PCM/MeCN conformers gave moderate to good agreement with the experimental ECD spectrum for both D



Article

and isomerization results in the para quinone methide intermediate D.51 Through a nucleophilic attack of the carbanion C to the carbocation E, 15 and 16 are formed. All isolated compounds were evaluated for their cytotoxic activities against the mouse lymphoma cell line L5178Y and the human ovarian cancer cell line A2780. Penicillic acid (12) and bromophilone B (16) showed cytotoxicity against the mouse lymphoma cell line L5178Y, with both having an IC50 value of 8.9 μM. Bromophilone A (15) had an IC50 value of 13.9 μM, higher than its epimer 16. Bromophilone B (16) exhibited moderate cytotoxicity against the human ovarian cancer cell line A2780 with an IC50 value of 2.7 μM. Bromophilone A (15) was inactive, with an IC50 value of 37 μM. All other compounds were inactive (IC50 > 10 μM). In summary, 14 compounds (1−14) including a new diphenyl ether, penicanether (1), were isolated from the axenic solid rice medium fermentation of the sponge-derived fungus P. canescens. Addition of 5% NaBr to the rice medium significantly changed the pattern of metabolites, leading to the induction of two new brominated azaphilones (15 and 16) and the isolation of one additional known xanthone (17). Bromophilone B (16) exhibited moderate cytotoxicity against the mouse lymphoma cell line L5178Y (IC50 8.9 μM) and the human ovarian cancer cell line A2780 (IC50 2.7 μM), whereas its epimer 15 was considerably less active.

diastereomers, allowing elucidation of the absolute configuration of the condensed bicyclic moiety as (7S,8aR) (Figure 4). Similarly to 15, the C-3′ and C-5′ absolute configurations could not be determined by the ECD calculations.

Figure 4. Comparison of the experimental ECD spectrum of 16 in MeOH (black line) with the sTDA-CAM-B3LYP spectrum of (7S,8aR,3′R,5′R)-16 computed from the CAM-B3LYP/TZVP single-point data, level of optimization: CAM-B3LYP PCM/MeOH (red dashed line); and the sTDA-CAM-B3LYP spectrum of (7S,8aR,3′S,5′S)-16 computed from the CAM-B3LYP/TZVP singlepoint data, level of optimization: CAM-B3LYP PCM/MeOH (blue dashed line). Computed spectra are Boltzmann statistics of the lowenergy (≥1%) conformers.

■

EXPERIMENTAL SECTION

General Experimental Procedures. Optical rotations were measured with a JASCO P-2000 polarimeter. UV spectra were obtained from the HPLC-DAD chromatograms of this system in concentrations that allowed for absorption in the Lambert−Beer region. ECD spectra were recorded on a JASCO J-810 spectrometer. FT-IR measurements were carried out on a Bruker TENSOR 37 IR spectrometer at ambient temperature in a KBr pellet at a range of 4000−400 cm−1 with a 4 cm−1 resolution (16 scans per measurement). NMR spectra were recorded on Bruker Avance III 600, 700, or 750 spectrometers. HRESIMS data were recorded with Bruker Daltonics UHR-QTOF Maxis 4G or Thermo Fisher LTQ FT Ultra mass spectrometers. HPLC-DAD analysis was performed using a Dionex UltiMate-3400SD system (Thermo Fisher) with an LPG3400SD pump, a DAD3000RS photodiode array detector, and a Knauer Eurospher C18 analytical column (125 × 4 mm, 5 μm). Semipreparative HPLC was performed using a Lachrom-Merck Hitachi system (an L-7100 pump, an L-7400 UV detector, and a 300 × 8 mm Knauer Eurospher C18 column) with MeOH and H2O as eluents. Column chromatography was performed with Sephadex LH20 (Merck) or silica gel 60 M (Macherey-Nagel). Routine thin layer chromatography (TLC) was performed on precoated silica plates (Merck silica gel F254), and spots were detected at 254 or 365 nm or by dipping the plates into anisaldehyde reagent followed by heating at 105 °C. Fungal Material and Identification. The fungus was isolated from the inner tissues of the marine sponge Agelas oroides. The specimen was collected at a depth of 10 m at the coast of Sığaçık̇ Izmir, Turkey, in 2013 by local fishermen. The identification of the sponge was performed by Prof. Semih Engin in December 2013. The fungus was identified as Penicillium canescens, using a molecular biology protocol for the amplification and sequencing of the 5.8S rDNA and the internal transcribed spacer region (18S rDNA) and subsequent BLAST search as described before.52 The resulting sequence data have been submitted to GenBank under the accession number MH820167. A deep frozen sample of the fungal strain P. canescens (4.14.6a) has been stored at one of the authors’ laboratory (P.P.). Fermentation, Extraction, and Isolation. P. canescens was cultivated on solid rice medium in 10 Erlenmeyer flasks (1 L each, 100 g rice, 110 mL demineralized H2O, and 3.8 g Sigma-Aldrich

In an attempt to assign the absolute configuration of the 3,5dihydroxyhexanoyl side chain in 15 and 16, biosynthetic aspects were taken into consideration. It has been reported that the side chain can be produced via reverse β-oxidation in enantiopure (3R,5R)-form by fungal metabolism.43,44 The side chain may be formed by reverse peroxisomal multifunctional βoxidation to yield the 3-hydroxyacyl-CoA-ester intermediate by reduction of the 3-ketoacyl-CoA-ester.45−47 These reactions are known to be stereospecific in the microbodies of fungi to yield the R-form of the resulting 3-hydroxyacyl-CoA, whereas fungi lack the epimerase necessary to convert the product into the S-form.47 The (3R,5R)-dihydroxyhexanoic acid enantiomer has been isolated in polymer and lactone form from fungi.48 These findings combined with the relative syn-configuration suggested a (3′R,5′R) absolute configuration of the dihydroxyhexanoyl side chain in 15 and 16. The structural novelty of 15 and 16 lies in the aromatic side chain at position 8a of the azaphilone core, which is atypical14 and hitherto unreported. A plausible biosynthetic pathway for 15 and 16 is proposed in Figure 5. The azaphilone core shares a high similarity to chaetoviridins and chaetomugulins, for which the biosynthesis has been extensively studied.49 The 3,5dihydroxyhexanoic acid side chain as a product of reverse βoxidation44 is then attached to the epimeric hydroxy unit to form the ester intermediate A. Reduction of the Δ1 double bond and enolization of the adjacent C-8 keto group give intermediate B. Deprotonation of the C-8 enol followed by isomerization leads to the formation of the carbanion intermediate C. The aromatic side chain is likely derived from vulvulic acid, a fungal secondary metabolite found in Penicillium sp.,50 which is structurally similar to compounds 10 and 11. Decarboxylation of vulvulic acid followed by oxidation E



Article

Figure 5. Proposed biosynthesis for 15 and 16. artificial sea salt per flask). The medium was sterilized by autoclaving it at 121 °C for 20 min prior to fungal inoculation. After 17 days at 20 °C under static conditions, the fermentation was terminated by the addition of 350 mL of EtOAc per flask. After an overnight soak, the rice was cut using a scalpel and subsequently shaken at 150 rpm for 8 h. The extract was collected through a paper filter, and the residue was washed with 50 mL of EtOAc, which was poured through the same filter. The extract was evaporated to dryness at 40 °C using the rotary evaporator, to yield around 10 g of EtOAc extract. Liquid−liquid separation between n-hexane and 90% MeOH−H2O was performed. A 2 g amount of insoluble griseofulvin (8) and 2.4 g of oily phase were removed to give 5.6 g of a MeOH-soluble fraction. The latter was separated by vacuum liquid chromatography on a silica gel column using a solvent gradient (from 100% n-hexane to 100% EtOAc and subsequently from 100% CH2Cl2 to 100% MeOH), resulting in 14 fractions (V1−V14). Fraction V2 was subjected to a silica gel column (10 × 200 mm) with CH2Cl2−MeOH (99:2), yielding 1 (2.1 mg) and 3 (4.3 mg). Fraction V3 was subjected to a Sephadex LH20 column using CH2Cl2−MeOH (1:1) as mobile phase to give 4 (109.0 mg) and three additional subfractions (V3S1−V3S4).

Subfraction V3S2 was further separated by a silica gel column (CH2Cl2−MeOH, 95:5) to give 12 (11.9 mg). Subfraction V3S3 was purified by semipreparative HPLC to yield 2 (13.4 mg), while subfraction V3S4 was purified by semipreparative HPLC to yield 5 (4.1 mg), 6 (15.5 mg), 7 (1.5 mg), 10 (2.5 mg), and 11 (6.4 mg). Fraction V4 was separated by a Sephadex LH20 column (CH2Cl2− MeOH, 1:1) followed by purification using semipreparative HPLC to yield 13 (20.4 mg). The OSMAC experiment used the same cultivation and extraction parameters as the initial fermentation. It was performed on five flasks (1 L each, 100 g rice, 110 mL demineralized H2O, 3.8 g artificial sea salt, and 5.0 g NaBr per flask). The resulting extract (6.3 g) was partitioned between n-hexane and 90% MeOH−H2O to give 4.9 g of MeOH-soluble extract. The latter was separated by VLC with a gradient solvent system (from 80% n-hexane to 100% EtOAc and subsequently from 100% CH2Cl2 to 100% MeOH) to give 17 fractions (V1−V17). All fractions were analyzed by HPLC-DAD, and the peaks with unidentified UV spectra were further investigated. Fraction V4 was purified by semipreparative HPLC to yield 17 (27.3 mg). Fraction V5 was analyzed by cochromatography with authentic F



■

dechlorogriseofulvin standard and identified as such, yielding 9 (200 mg). Fraction V9 was purified by semipreparative HPLC to yield 14 (31.1 mg). Under the exclusion of light by using brown glass and aluminum foil, fraction V12 was separated by a Sephadex LH20 column (CH2Cl2−MeOH, 1:1) followed by purification with semipreparative HPLC to yield 15 (1.2 mg) and 16 (1.1 mg). Methyl 3-chloro-2-(2,4-dimethoxy-6-methylphenoxy)-6-hydroxy-4-methoxybenzoate (1): white powder; UV (MeOH) λmax (log ε) 283 (3.29) nm; IR νmax (KBr) 3443, 2950, 2921, 1651, 1602, 1561, 1431, 1322, 1248, 1203, 1090, 809 cm−1; 1H and 13C NMR data, Table 1; HRESIMS m/z 383.0892 [M + H]+ (calcd for C18H20ClO7, 383.0892). Bromophilone A (15): red powder; [α]26D −19 (c 0.2, MeOH); UV (MeOH) λmax (log ε) 428 (3.96) and 311 (3.71) nm; ECD (0.291 mM, MeOH), λmax (Δε) 464 (+3.10), 382sh (−3.09), 356sh (−3.88), 318sh (−8.01), 295 (−10.72), 260sh (−3.00), 229 (−7.38), 204sh (+13.27); IR νmax (KBr) 3443, 2920, 2851, 1700, 1677, 1651, 1616, 1543, 1505, 1241, 1137, 1008, 868 cm−1; 1H and 13C NMR data, Table 2; HRESIMS m/z 719.13398 [M + H]+ (calcd for C33H36BrO13, 719.13338). Bromophilone B (16): red powder; [α]26D −99 (c 0.2, MeOH); UV (MeOH) λmax (log ε) 424 (3.36) and 306 (3.36) nm; ECD (0.278 mM, MeOH), λmax (Δε) 423 (−8.24), 346sh (+2.66), 319 (+4.20), 266 (+3.25), 207sh (+6.57); IR νmax (KBr) 3433, 2920, 2850, 1720, 1623, 1545, 1385, 1272, 1135, 1009 cm−1; 1H and 13C NMR data, Table 2; HRESIMS m/z 719.13312 [M + H]+ (calcd for C33H36BrO13, 719.13338). Computational Section. Mixed torsional/low-frequency mode conformational searches were carried out by means of the MacroModel 10.8.011 software using the MMFF with an implicit solvent model for CHCl3.53 Geometry reoptimizations were carried out first at the semiempirical AM1 level and then at the CAMB3LYP/TZVP level with the PCM solvent model for MeOH. Singlepoint calculations for the sTDA ECD calculations and TDDFT-ECD calculations were run with various functionals (CAM-B3LYP, LCBLYP, and ωB97X for the single-point and B3LYP, BH&HLYP, CAM-B3LYP, and PBE0 for the TDDFT) and the TZVP basis set as implemented in the Gaussian09 package.54 The sTDA calculations were performed with the sTDA 1.6 package.55 ECD spectra were generated as sums of Gaussians with 2400 to 3300 cm−1 widths at half-height, using dipole-velocity-computed rotational strength values.56 Boltzmann distributions were estimated from the CAMB3LYP/TZVP energies. The MOLEKEL software package was used for visualization of the results.57 Cytotoxicity Assay. Cytotoxicity against the mouse lymphoma cell line L5178Y was evaluated by the MTT assay as previously described.58 Kahalalide F (IC50 4.3 μM) and 0.1% ethylene glycol monomethyl ether in DMSO were used as positive and negative controls, respectively. The cytotoxicity assay against the human ovarian cancer cell line A2780 was carried out as previously described.59 In brief, A2780 cells were seeded at a density of 9500 cells/well in 96-well plates (Corning). After 24 h, cells were exposed to increased 3.6-fold serially diluted concentrations of test compounds. Incubation was ended after 72 h, and cell survival was determined by addition of MTT solution (5 mg/mL in phosphatebuffered saline). The formazan precipitate was dissolved in DMSO (VWR). Absorbance was measured at 544 and 690 nm in a FLUOstar microplate reader (BMG LabTech). Cisplatin (IC50 1.7 μM) and 0.9% sodium chloride were used as positive and negative controls, respectively.

■

Article

AUTHOR INFORMATION

Corresponding Authors

*Tel: +49 211 81 15979. Fax: +49 211 81 11923. E-mail: [email protected] (Z.L.). *Tel: +49 211 81 14163. Fax: +49 211 81 11923. E-mail: [email protected] (P.P.). ORCID

Attila Mándi: 0000-0002-7867-7084 Wenhan Lin: 0000-0002-4978-4083 Zhen Liu: 0000-0003-3314-7853 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS P.P. wants to thank the DFG (270650915/GRK 2158) and the Manchot Foundation for support. T.K. thanks the National Research, Development and Innovation Office (NKFI K120181). M.U.K. acknowledges financial support from the BMBF (16GW0108). A.M. wants to thank the János Bolyai Research Scholarship of the Hungarian Academy of Sciences for financial support. The Governmental Information-Technology Development Agency (KIFÜ ) is acknowledged for CPU time. Furthermore, we wish to thank Prof. H. Weber (HHU Duesseldorf) for helpful discussions.

■

REFERENCES

(1) Newman, D. J.; Cragg, G. M. J. Nat. Prod. 2016, 79, 629−661. (2) Blunt, J. W.; Carroll, A. R.; Copp, B. R.; Davis, R. A.; Keyzers, R. A.; Prinsep, M. R. Nat. Prod. Rep. 2018, 35, 8−53. (3) Brakhage, A. A.; Schroeckh, V. Fungal Genet. Biol. 2011, 48, 15− 22. (4) Gerke, J.; Bayram, O.; Feussner, K.; Landesfeind, M.; Shelest, E.; Feussner, I.; Braus, G. H. Appl. Environ. Microbiol. 2012, 78, 8234− 8244. (5) Daletos, G.; Ebrahim, W.; Ancheeva, E.; El-Neketi, M.; Lin, W. H.; Proksch, P. In Chemical Biology of Natural Products; Newman, D. J., Cragg, G. M., Grothaus, P. G., Eds.; CRC Press: Boca Raton, FL, USA, 2017; pp 233−284. (6) Hewage, R. T.; Aree, T.; Mahidol, C.; Ruchirawat, S.; Kittakoop, P. Phytochemistry 2014, 108, 87−94. (7) Zhang, Z. X.; Yang, X. Q.; Zhou, Q. Y.; Wang, B. Y.; Hu, M.; Yang, Y. B.; Zhou, H.; Ding, Z. T. Molecules 2018, 23, 1816. (8) Wang, M. H.; Jiang, T.; Ding, G.; Niu, S. B.; Wang, X. W.; Yu, M.; Gu, Y. C.; Zhang, Q. B.; Chen, J. H.; Jia, H. M.; Zou, Z. M. J. Antibiot. 2017, 70, 801−804. (9) Wang, W.; Liao, Y.; Chen, R.; Hou, Y.; Ke, W.; Zhang, B.; Gao, M.; Shao, Z.; Chen, J.; Li, F. Mar. Drugs 2018, 16, 61. (10) Wang, W. X.; Kusari, S.; Laatsch, H.; Golz, C.; Kusari, P.; Strohmann, C.; Kayser, O.; Spiteller, M. J. Nat. Prod. 2016, 79, 704− 710. (11) Arunpanichlert, J.; Rukachaisirikul, V.; Sukpondma, Y.; Phongpaichit, S.; Tewtrakul, S.; Rungjindamai, N.; Sakayaroj, J. Chem. Pharm. Bull. 2010, 58, 1033−1036. (12) Panthama, N.; Kanokmedhakul, S.; Kanokmedhakul, K.; Soytong, K. J. Nat. Prod. 2011, 74, 2395−2399. (13) Tang, J. L.; Zhou, Z. Y.; Yang, T.; Yao, C.; Wu, L. W.; Li, G. Y. J. Agric. Food Chem. 2019, 67, 2175−2182. (14) Gao, J. M.; Yang, S. X.; Qin, J. C. Chem. Rev. 2013, 113, 4755− 4811. (15) Matsuzaki, K.; Tahara, H.; Inokoshi, J.; Tanaka, H.; Masuma, R.; Omura, S. J. Antibiot. 1998, 51, 1004−1011. (16) Chen, M.; Zheng, Y. Y.; Chen, Z. Q.; Shen, N. X.; Shen, L.; Zhang, F. M.; Zhou, X. J.; Wang, C. Y. J. Nat. Prod. 2019, 82, 368− 374.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jnatprod.9b00151. UV, IR, HRESIMS, and NMR spectra of compounds 1, 15, and 16 as well as ECD calculations for compounds 15 and 16 (PDF) G



Article

(52) Kjer, J.; Debbab, A.; Aly, A. H.; Proksch, P. Nat. Protoc. 2010, 5, 479−490. (53) MacroModel 9.0; Schrödinger, LLC.: New York, NY, USA, 2015. https://www.schrodinger.com/macromodel. (54) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; et al. Gaussian 09 Revision E.01; Gaussian, Inc.: Wallingford, CT, 2013. (55) Grimme, S.; Bannwarth, C.; Shushkov, P.; Wergifosse, M. d. sTDA - A Program Package to calculate Excited States via a simplified Tamm-Dancoff Approximation to TD-DFT Version 1.6; Mulliken Center for Theoretical Chemistry, Ed.; Institut für Physikalische und Theoretische Chemie der Universität Bonn: Bonn, Germany, 2018. (56) Stephens, P. J.; Harada, N. Chirality 2009, 22, 229−233. (57) Varetto, U. MOLEKEL 5.4; Swiss National Supercomputing Centre: Manno, Switzerland, 2009. (58) Ashour, M.; Edrada, R.; Ebel, R.; Wray, V.; Wätjen, W.; Padmakumar, K.; Müller, W. E. G.; Lin, W. H.; Proksch, P. J. Nat. Prod. 2006, 69, 1547−1553. (59) Engelke, L. H.; Hamacher, A.; Proksch, P.; Kassack, M. U. J. Cancer 2016, 7, 353−363.

(17) Nicoletti, R.; Lopez-Gresa, M. P.; Manzo, E.; Carella, A.; Ciavatta, M. L. Mycopathologia 2007, 163, 295−301. (18) Bertinetti, B. V.; Peña, N. I.; Cabrera, G. M. Chem. Biodiversity 2009, 6, 1178−1184. (19) Brian, P. W.; Hemming, H. G.; Moffatt, J. S.; Unwin, C. H. Trans. Br. Mycol. Soc. 1953, 36, 243−247. (20) Xia, X.; Li, Q.; Li, J.; Shao, C.; Zhang, J.; Zhang, Y.; Liu, X.; Lin, Y.; Liu, C.; She, Z. Planta Med. 2011, 77, 1735−1738. (21) Mutanyatta, J.; Matapa, B. G.; Shushu, D. D.; Abegaz, B. M. Phytochemistry 2003, 62, 797−804. (22) Abdel-Lateff, A.; Klemke, C.; König, G. M.; Wright, A. D. J. Nat. Prod. 2003, 66, 706−708. (23) Cacho, R. A.; Chooi, Y.-H.; Zhou, H.; Tang, Y. ACS Chem. Biol. 2013, 8, 2322−2330. (24) MacMillan, J. J. J. Chem. Soc. 1953, 1697−1702. (25) Varma, G. B.; Fatope, M. O.; Marwah, R. G.; Deadman, M. E.; Al-Rawahi, F. K. Phytochemistry 2006, 67, 1925−1930. (26) Beekman, A. M.; Castillo Martinez, E.; Barrow, R. A. Org. Biomol. Chem. 2013, 11, 1109−1115. (27) Sekiguchi, J.; Katayama, S.; Yamada, Y. Appl. Environ. Microbiol. 1987, 53, 1531−1535. (28) Kosemura, S. Tetrahedron 2003, 59, 5055−5072. (29) Kozlovsky, A.; Vinokurova, N. G.; Adanin, V. M.; Gräfe, U. Nat. Prod. Lett. 2000, 14, 333−340. (30) Frank, M.; Ö zkaya, F.; Müller, W.; Hamacher, A.; Kassack, M.; Lin, W.; Liu, Z.; Proksch, P. Mar. Drugs 2019, 17, 99. (31) Belofsky, G. N.; Gloer, K. B.; Gloer, J. B.; Wicklow, D. T.; Dowd, P. F. J. Nat. Prod. 1998, 61, 1115−1119. (32) Grimme, S. J. Chem. Phys. 2013, 138, 244104. (33) Risthaus, T.; Hansen, A.; Grimme, S. Phys. Chem. Chem. Phys. 2014, 16, 14408−14419. (34) Bannwarth, C.; Grimme, S. Comput. Theor. Chem. 2014, 1040− 1041, 45−53. (35) Bannwarth, C.; Grimme, S. J. Phys. Chem. A 2015, 119, 3653− 3662. (36) Covington, C. L.; Junior, F. M. S.; Silva, J. H. S.; Kuster, R. M.; de Amorim, M. B.; Polavarapu, P. L. J. Nat. Prod. 2016, 79, 2530− 2537. (37) Kicsák, M.; Mándi, A.; Varga, S.; Herczeg, M.; Batta, G.; Bényei, A.; Borbás, A.; Herczegh, P. Org. Biomol. Chem. 2018, 16, 393−401. (38) Yanai, T.; Tew, D. P.; Handy, N. C. Chem. Phys. Lett. 2004, 393, 51−57. (39) Iikura, H.; Tsuneda, T.; Yanai, T.; Hirao, K. J. Chem. Phys. 2001, 115, 3540−3544. (40) Chai, J. D.; Head-Gordon, M. J. Chem. Phys. 2008, 128, 84106. (41) Superchi, S.; Scafato, P.; Górecki, M.; Pescitelli, G. Curr. Med. Chem. 2018, 25, 287−320. (42) Mándi, A.; Mudianta, I. W.; Kurtán, T.; Garson, M. J. J. Nat. Prod. 2015, 78, 2051−2056. (43) Giver, L. J.; Newman, L. M.; Mundorff, E.; Huisman, G. W.; Jenne, S. J.; Zhu, J.; Behrouzian, B.; Grate, J. H.; Lalonde, J. U.S. Patent 8,273,547 B2, 2012. (44) Gonzalez, R.; Clomburg, J. M. U.S. Patent 9,994,881 B2, 2014. (45) Hiltunen, J. K.; Wenzel, B.; Beyer, A.; Erdmann, R.; Fosså, A.; Kunau, W. H. J. Biol. Chem. 1992, 267, 6646−6653. (46) Poirier, Y.; Antonenkov, V. D.; Glumoff, T.; Hiltunen, J. K. Biochim. Biophys. Acta, Mol. Cell Res. 2006, 1763, 1413−1426. (47) Winkler, U.; Säftel, W.; Stabenau, H. Plant Physiol. 2003, 131, 753−762. (48) Maloney, K. N. Biologically Active Natural Products from Plants and their Endophytes. Ph.D. Thesis, Cornell University, Ithaca, NY, USA, 2007. (49) Winter, J. M.; Sato, M.; Sugimoto, S.; Chiou, G.; Garg, N. K.; Tang, Y.; Watanabe, K. J. Am. Chem. Soc. 2012, 134, 17900−17903. (50) Kimura, Y.; Nishibe, M.; Nakajima, H.; Hamasaki, T. Agric. Biol. Chem. 1991, 55, 1137−1138. (51) Sugumaran, M. Int. J. Mol. Sci. 2016, 17, 1576. H


Brominated Azaphilones from the Sponge ... - ACS Publications

Recommend Documents