Hyalodendriellins A–F, new 14-membered resorcylic acid lactones from the endophytic fungus Hyalodendriella sp. Ponipodef12

Abstract

Six new 14-membered resorcylic acid lactones (RALs), named hyalodendriellins A–F (1–6), were isolated from the culture of the endophytic fungus Hyalodendriella sp. Ponipodef12 associated with the hybrid ‘Neva’ of Populus deltoides Marsh × P. nigra L. The structures of the new compounds were deduced by analyses of the NMR spectroscopic and mass spectrometry data, in combination with chemical conversion, modified Mosher's method and TDDFT ECD calculations for determining the absolute configurations. Compounds 1, 2, 4, and 6 possess a 3S configuration, while 3 and 5 are 3R-configurated. The co-occurrence of RALs with different stereochemistry at C-3 in the same fungus is rare. The CD behaviors of 1–6 as well as their acetonide and hydrogenated derivatives were investigated. All the isolated compounds were evaluated for their antinematodal, larvicidal, cytotoxic, antibacterial and antifungal activities. Among them, hyalodendriellin A (1) displayed moderate antinematodal activity against Caenorhabditis elegans and Meloidogyne incognita. Hyalodendriellin C (3) exhibited larvicidal effect against the fourth-instar larvae of the mosquito Aedes aegypti.

---

Introduction

14-Membered resorcylic acid lactones (RALs), which feature a 14-membered macrocyclic ring fused to a β-resorcylic acid moiety, belong to a subclass of the benzenediol lactone family. Up to now, more than 100 RALs, such as zeaenols, zearalenones, caryospomycins, cochliomycins, and hamigeromycins, have been isolated from many fungal genera.¹ These metabolites displayed a diverse array of biological activities, such as inhibition of heat shock protein 90 and kinases,^2,3 cytotoxic,⁴ antiviral,⁵ anti-inflammatory,^6,7 estrogenic,⁸ and nematocidal activities.⁹ Due to their promising biological activities and unique structures, some RALs have been considered as the targets for realizing the total synthesis,¹⁰ or biosynthesis studies.^11,12

Endophytic fungi are known as a prolific source of bioactive natural products.¹³ In continuation of our interest in searching for bioactive substances from fungal endophytes,^14,15 an endophytic fungus, Hyalodendriella sp. Ponipodef12, isolated from the hybrid ‘Neva’ of Populus deltoides Marsh × P. nigra L., was selected for mycochemical investigation due to the interesting HPLC-DAD and ¹H NMR profiles of the crude extract. This fungus was previously reported to produce four benzo-α-pyrones in liquid culture on potato dextrose broth medium.^16,17 When cultured on rice media, the HPLC chromatogram and ¹H NMR spectrum of the extract showed additional peaks corresponding to 14-membered RALs. The subsequent fractionation has led to the isolation and identification of six new RALs, named hyalodendriellins A–F (1–6). Herein, we report the isolation, structural elucidation, as well as the biological activities of the new compounds.

Results and discussion

The EtOAc extract of the culture of Hyalodendriella sp. Ponipodef12 on rice media was subjected to column chromatography over silica gel and Sephadex LH-20, followed by purification using semi-preparative HPLC to afford compounds 1–6 (Fig. 1).

Fig. 1 Structures of hyalodendriellins A–F (1–6).

Hyalodendriellin A (1) was isolated as a pale yellow oil. It exhibited a prominent pseudomolecular peak at m/z 417.15153 [M + Na]⁺ in the HRESIMS spectrum, indicating a molecular formula of C₂₀H₂₆O₈. The ¹³C NMR spectrum (Table 1, in CDCl₃) displayed signals for twenty carbons, which could be classified into one ester carbonyl (δ_C 171.2), ten sp²-hybridized carbons that were assignable to two disubstituted double bonds, and one benzene ring, four oxygenated methines (δ_C 76.2, 72.7, 71.7, 71.5), two methoxy groups (δ_C 60.2, 55.7), two methylenes (δ_C 37.3, 36.2), and one methyl group (δ_C 19.6), by the aid of DEPT-135 and HMQC experiments. The ¹H NMR spectrum (DMSO-d₆) showed signals for one isolated aromatic proton (δ_H 6.43, s), four olefinic protons (δ_H 6.41, 5.84, 5.60, 5.45), four oxygenated methines (δ_H 4.83, 3.76, 3.42, 3.36), two methoxy groups (δ_H 3.55, 3.75), two methylene groups (δ_H 2.45, 2.07; 2.25), and one methyl group (δ_H 1.25, d). In addition, four D₂O-exchangeable protons (δ_H 9.85, 4.66, 4.58, 4.46) were ascribed to four hydroxy groups. The above functionalities account for seven of the eight degrees of unsaturation required by the molecular formula, thus suggesting a bicyclic nature of 1. The NMR data of 1 resembled those of zeaenol,¹⁸ which is a 14-membered resorcylic acid lactone, however, they differed in the benzene ring, in which an additional methoxy group (δ_H 3.55/δ_C 59.9) in 1 replaced the aromatic proton at 13 of the latter. This was corroborated by analysis of the HMBC correlations (Fig. 2) from H-15 (δ_H 6.43, s) to C-14, C-16, C-13, and C-16a, from 13-OMe (δ_H 3.55, s) to C-13, from 14-OMe (δ_H 3.75, s) to C-14, and from H-12 (δ_H 6.41, d) to C-13, C-12a, and C-16a. Thus, the planar structure of 1 was established as the 13-methoxylated derivative of zeaenol.

Table 1 ¹H and ¹³C NMR data of 1 and 2

Position	1 (CDCl3)		1 (DMSO-d6)		2 (DMSO-d6)
	δC, type	δH, mult. (J in Hz)	δC, type	δH, mult. (J in Hz)	δC, type	δH, mult. (J in Hz)
a Signals overlapped.
1	171.2, C		167.4, C		167.8, C
3	71.5, CH	5.29, m	73.2, CH	4.83, m	72.5, CH	5.19, m
4	37.3, CH2	2.47, m	37.3, CH2	2.25, m	35.9, CH2	2.29, ddd (15.8, 11.4, 9.0), 1.34, d (15.8)
5	128.7, CH	5.90, ov^a	128.9, CH	5.60, ddd (15.6, 7.8, 6.2)	72.1, CH	4.02, dd (11.4, 6.0)
6	131.7, CH	5.57, dd (15.6, 7.8)	133.4, CH	5.45, dd (15.7, 8.1)	37.6, CH2	1.71, ddd (13.1, 5.0, 2.3), 1.58, ddd (13.1, 10.7, 6.2)
7	71.7, CH	4.18, dd (8.6, 7.8)	74.0, CH	3.76, m	68.9, CH	3.45, m
8	76.2, CH	3.50, dd (8.6, 1.7)	78.5, CH	3.36, m	75.3, CH	2.86, td (8.2, 4.2)
9	72.7, CH	3.90, m	71.2, CH	3.42, m	76.5, CH	3.27, ddd (10.9, 8.7, 2.3)
10	36.2, CH2	2.56, m, 2.39, m	36.1, CH2	2.45, m, 2.07, m	33.3, CH2	2.52, ol^a, 2.05, ddd (12.7, 11.1, 6.5)
11	131.4, CH	5.86, ov^a	134.4, CH	5.84, ddd (15.9, 9.7, 4.2)	134.1, CH	5.85, ddd (16.1, 8.1, 6.6)
12	125.7, CH	6.61, d (15.9)	124.0, CH	6.41, d (15.9)	124.4, CH	6.40, d (16.1)
12a	133.6, C		130.4, C		131.0, C
13	140.0, C		138.4, C		138.5, C
14	158.4, C		153.8, C		154.1, C
15	99.2, CH	6.37, s	99.4, CH	6.43, s	99.6, CH	6.45, s
16	160.6, C		152.1, C		153.2, C
16a	104.0, C		112.4, C		111.8, C
3-Me	19.6, CH3	1.38, d (6.2)	20.4, CH3	1.25, d (6.1)	22.3, CH3	1.15, d (6.3)
13-OMe	60.2, CH3	3.55, s	59.9, CH3	3.55, s	60.0, CH3	3.57, s
14-OMe	55.7, CH3	3.83, s	55.6, CH3	3.75, s	55.6, CH3	3.76, s
16-OH		11.52, s		9.85, s		9.78, br.s
Other OH				7-OH: 4.66, br.d (2.6), 8-OH: 4.58, br.s, 9-OH: 4.46, d (6.0)		7-OH: 4.75, d (4.2), 8-OH: 4.86, d (5.1)

---

Fig. 2 Selected HMBC correlations of 1 and NOESY correlations of its 7,8-acetonide (1a).

Compound 1 has four chiral centers (C-3, C-7, C-8, and C-9) and incorporates three vicinal hydroxy groups at C-7–C-9. The absolute configuration of 1 was established by chemical conversion to its acetonide (1a). Treatment of 1 with 2,2-dimethoxypropane (DMP) and p-TsOH allowed the formation of its 7,8-acetonide (1a). The E-configuration for the double bonds was revealed by the large vicinal coupling constants (³J_H-12/H-11 = 15.9 Hz, ³J_H-5/H-6 = 15.4 Hz). The relative configuration of C-7–C-9 in 1a was established by analysis of the ¹H–¹H coupling constants and NOESY correlations. The anti relationship of H-7 and H-8 was suggested by the large J value (³J_H-7/H-8 = 8.3 Hz), while the gauche relationship of H-9 and H-8 was deduced from the small coupling constant between them (2.3 Hz), which were similar to those of the 7,8-acetonide of aigialomycin B (8.0, 1.6 Hz, respectively),¹⁹ and cochliomycin A (8.4, 2.4 Hz, respectively),²⁰ indicating that these compounds shared the same relative configuration for the triol unit. This was corroborated by analysis of the NOESY spectrum (Fig. 2), in which H-7 showed correlation to one methyl (δ_H 1.44, s) of the acetonide, while H-8 correlated to the other one (δ_H 1.38, s), confirming the anti relationship of H-7 and H-8, while the observed correlations of H-7/H-10b, H-8/H-9, and H-9/H-10a (δ_H 2.85, dddd), indicated H-9 was cis to H-8. In order to determine the absolute configuration of C-9, the modified Mosher's method was applied.^22,23 Compound 1a was reacted with (R)- or (S)-α-methoxy-α-phenylacetic acid (MPA) to give the corresponding (R)- or (S)-MPA ester (1aR/1aS), respectively. The resulting Δδ^RS values (δ_1aR − δ_1aS) indicating the S configuration for C-9 (Fig. 3). Taken into consideration of the aforementioned relative configuration, the 7S, 8S configuration of 1a was established.

Fig. 3 Δδ^RS (δ_1aR − δ_1aS) values for the MPA esters of 1a.

Nowadays, TDDFT ECD calculation has been a powerful tool to establish the absolute configuration of natural products by comparing the calculated spectra with the experimental data.²¹ Based on the established absolute configuration for C-7–C-9 of 1a, (3S, 7S, 8S, 9S)-1a was randomly selected for ECD calculation. (3S, 7S, 8S, 9S)-1a was first subjected to molecular Merck force field (MMFF) conformational search, followed by geometry optimization using the DFT method at the B3LYP/6-31G(d) level. The subsequent ECD calculation of the low-energy conformers (≥1%) were performed at the pbe0/TZVP level with the polarizable continuum model (PCM) in MeOH. The experimental ECD spectrum of 1a displayed negative cotton effect at 270 nm and positive cotton effect at 231 nm. The calculated ECD spectrum of (3S, 7S, 8S, 9S)-1a fit well with the experimental one (Fig. 4), suggesting the 3S, 7S, 8S, 9S configuration of 1a. Therefore, hyalodendriellin A (1) must possess the same 3S, 7S, 8S, 9S configuration. Interestingly, compound 1 shared the same planar structure as that of caryospomycin B, which was isolated from the aquatic fungus Caryospora callicarpa,⁹ however, they differed at C-9 for having an opposite configuration and the absolute configuration of the later compound was not determined.

Fig. 4 The experimental ECD spectrum of 1a and the calculated ECD spectrum of (3S, 7S, 8S, 9S)-1a.

Hyalodendriellin B (2) was isolated as a pale yellow oil, whose molecular formula was determined as C₂₀H₂₆O₈ by HRESIMS, which was the same as that of 1. The NMR data of 2 and 1 were similar (Table 1), however, the signals for a pair of olefinic protons were disappeared in 2, while signals for one oxygenated methine (δ_C 72.1; δ_H 4.02, dd) and a methylene (δ_C 37.6; δ_H 1.71, 1.58, each ddd) were observed. Analysis of the ¹H–¹H COSY spectrum revealed the C-5/C-6 double bond in 1 was replaced by the aforementioned oxygenated methine and methylene in 2, respectively (Fig. 5). Taken into consideration of the molecular formula, there must be an ether linkage between C-5 and C-9 in 2. This was corroborated by HMBC experiment, as key correlations were found from H-5 to C-9, and H-9 to C-5.

Fig. 5 Key ¹H–¹H COSY, HMBC and NOESY correlations of 2.

The relative configuration of 2 was established by analysis of the coupling constants and NOESY correlations (Fig. 5). The large coupling constants between H-6b (δ_H 1.58, ddd)/H-7 (10.7 Hz), H-7/H-8 (8.2 Hz), and H-8/H-9 (8.7 Hz), indicated these protons were axial with regard to the six-membered ring. These were consistent with the NOESY correlations. In the NOESY spectrum, cross-peaks were found between H-4a (δ_H 2.29, ddd), H-7, H-9, and H-10a (δ_H 2.52), suggesting that these protons were oriented to the same face (tentatively as β). While the correlations observed between H-10b (δ_H 2.05, ddd), H-8, H-6b, H-5, H-4b (δ_H 1.34, d), and H-3, indicated that these protons directed to the α-face. Thus, the relative configuration of 2 was established.

The absolute configuration of 2 was established by TDDFT ECD calculation. The calculated ECD spectrum of (3S, 5R, 7S, 8R, 9S)-2 matched the experimental spectrum (Fig. 6), which unambiguously supported the 3S, 5R, 7S, 8R, 9S configuration of 2.

Fig. 6 The experimental ECD spectrum of 2 and the calculated ECD spectrum of (3S, 5R, 7S, 8R, 9S)-2.

Hyalodendriellin C (3) was isolated as a white needle-like crystal. Its molecular formula was deduced as C₂₀H₂₈O₈ by HRESIMS, bearing two more protons than that of 1. Comparison of the NMR data revealed their great similarities, except that the C-5/C-6 olefinic carbons in 1 were replaced by two methylene groups in 3 (Table 2). This was confirmed by analysis of the HMBC spectrum, in which correlations were found from H-7 (δ_H 3.51, m) to C-6 (δ_C 31.1) and C-5 (δ_C 16.9).

Table 2 ¹H and ¹³C NMR data of 3

Position	δC, type^a	δH, mult.^a (J in Hz)	δC, type^b	δH, mult.^b (J in Hz)
a Recorded in DMSO-d6.b Recorded in CD3OD.c Assignments within a column may be interchanged.d Signals overlapped.
1	167.4, C		170.9, C
3	69.9, CH	5.02, m	72.8, CH	5.16, m
4	33.9, CH2	1.60, m	35.4, CH2	1.73, m
5	16.9, CH2	1.36, m	18.6, CH2	1.72, m, 1.53, m
6	31.1, CH2	1.37, m	32.3, CH2	1.77, m, 1.52, m
7	68.2, CH	3.51, m	70.6, CH	3.72, m^c
8	70.3, CH	3.19, t (4.9)	72.1, CH	3.47, t (4.9)
9	69.2, CH	3.51, m	71.4, CH	3.74, m^c
10	37.4, CH2	2.49, m, 2.23, m	38.6, CH2	2.66, dddd (14.2, 7.7, 6.1, 1.4), 2.39, ddd (14.2, 8.1, 2.6)
11	132.6, CH	6.24, ov^d	133.1, CH	6.31, ddd (16.2, 8.1, 6.0)
12	124.6, CH	6.24, ov^d	127.0, CH	6.41, d (16.2)
12a	128.3, C		131.4, C
13	138.8, C		141.1, C
14	153.5, C		156.7, C
15	99.4, CH	6.41, s	100.4, CH	6.43, s
16	150.6, C		154.9, C
16a	113.8, C		112.7, C
3-Me	19.4, CH3	1.21, d (6.3)	19.7, CH3	1.31, d (6.2)
13-OMe	59.7, CH3	3.54, s	60.5, CH3	3.61, s
14-OMe	55.6, CH3	3.75, s	56.3, CH3	3.83, s
16-OH		9.58, s

---

The absolute configuration of 3 was established by chemical conversions. Treatment of 3 with DMP and p-TsOH afforded two acetonides, including the 8,9-acetonide (3a) and the 7,8-acetonide (Scheme 1). The latter was identical to the co-isolated hyalodendriellin E (5) (vide infra) as revealed by HPLC and ¹H NMR analyses.

Scheme 1 Chemical conversion of 3 to 5 and 3a.

Hyalodendriellin E (5) was isolated as a pale yellow oil, with the molecular formula of C₂₃H₃₂O₈. The NMR data of 5 (Table 3) were closely related to those of 3, except for the presence of additional signals for two acetal methyl groups (δ_C 27.3, 27.4) and one acetal carbon (δ_C 107.5) in 5, which hinted the acetonide nature of 5. Detailed analysis of the ¹H–¹H COSY spectrum disclosed C-7/C-8 to be the position of the acetonide group, as H-9 (δ_H 3.58, m) showed correlation to 9-OH (δ_H 4.83, d) while no signals were found for 7-OH and 8-OH when recorded in DMSO-d₆. This was corroborated by analysis of the NOESY spectrum, in which correlations were found from H-7 (δ_H 3.90, m) to one acetal methyl group (α-methyl: δ_H 1.31, s), and H-8 (δ_H 3.62, t) to the other one (β-methyl: δ_H 1.28, s) (Fig. 7). The coupling constants (³J_H-7/H-8 = ³J_H-8/H-9 = 7.1 Hz) between H-7, H-8 and H-9 (δ_H 3.58, m) suggested these protons were in a 1,2-anti relationship. The NOESY correlations of H-7/H-9, and H-7/H-10b (δ_H 2.23, dd), as well as the weak correlation of H-9/α-methyl (δ_H 1.31, s), indicated these protons were in the same direction (α), whereas the correlation of H-8/H-5b (δ_H 1.40, m) implied they were oriented to the opposite face (β). In order to determine the absolute configuration of C-9, the modified Mosher's method was applied. The Δδ^RS (δ_5R − δ_5S) values of the MPA esters of 5 indicated the S configuration for C-9 (Fig. 8), which translated into a 7R, 8R configuration by the aforementioned relative configuration.

Table 3 ¹H and ¹³C NMR data of 3a, 5 and 6

Position	3a (CDCl3)		5 (DMSO-d6)		6 (DMSO-d6)
	δC, type	δH, mult. (J in Hz)	δC, type	δH, mult. (J in Hz)	δC, type	δH, mult. (J in Hz)
a Signals overlapped with the solvent residual peak.b Signals partially overlapped.c Assignments within a column may be interchanged.
1	170.9, C		167.5, C		167.1, C
3	73.4, CH	5.12, m	69.8, CH	4.98, m	70.0, CH	5.00, qdd (6.4, 6.4, 2.6)
4	35.0, CH2	1.70, m, 1.56, m	34.5, CH2	1.62, m	34.9, CH2	1.69, m, 1.58, m
5	18.5, CH2	1.71, m, 1.58, m	18.8, CH2	1.55, m, 1.40, m	15.0, CH2	1.31, m
6	32.4, CH2	1.71, m, 1.56, m	30.1, CH2	1.60, m	27.2, CH2	1.63, m, 1.55, m
7	72.0, CH	3.68, td (7.0, 2.2)	76.3, CH	3.90, m	71.4, CH	3.71, m
8	80.4, CH	3.87, m	79.7, CH	3.62, t (7.1)	60.9, CH	3.18, td (9.6, 7.7)
9	76.7, CH	3.96, dt (8.6, 3.6)	70.3, CH	3.58, m	71.6, CH	3.71, m
10	36.7, CH2	2.84, dt (15.8, 3.9), 2.45, ddd (15.8, 7.6, 3.8)	37.4, CH2	2.50, m^a, 2.23, dd (14.5, 7.8)	33.0, CH2	2.53, t^a, 2.34, dt (9.4, 4.5)
11	130.0, CH	6.42, ddd^b	131.6, CH	6.34, dt (16.2, 7.0)	132.3, CH	5.93, ddd (16.2, 10.2, 5.0)
12	127.2, CH	6.49, d (16.1)	125.3, CH	6.18, d (16.2)	126.0, CH	6.31, d (16.2)
12a	131.9, C		127.9, C		129.8, C
13	140.7, C		138.8, C		138.6, C
14	158.5, C		153.4, C		153.7, C
15	99.5, CH	6.42, s	99.5, CH	6.42, s	99.2, CH	6.40, s
16	160.0, C		150.4, C		150.5, C
16a	104.9, C		114.3, C		114.1, C
3-Me	18.8, CH3	1.36, d (6.2)	20.6, CH3	1.25, d (6.1)	18.9, CH3	1.22, d (6.3)
13-OMe	59.9, CH3	3.61, s	59.6, CH3	3.56, s	59.9, CH3	3.52, s
14-OMe	55.9, CH3	3.88, s	55.6, CH3	3.76, s	55.6, CH3	3.75, s
16-OH		11.17, s		9.57, s		9.54, s
Other OH				9-OH: 4.83, d (4.4)		8-OH: 4.50, d (7.7)
Acetonide group	108.3, C, 27.09, CH3^c, 27.06, CH3^c	β: 1.43, s, α: 1.40, s	107.5, C, 27.4, CH3^c, 27.3, CH3^c	β: 1.28, s, α: 1.31, s	96.8, C, 19.2, CH3, 29.2, CH3	1.38, s, 1.17, s

---

Fig. 7 Selected ¹H–¹H COSY and NOESY correlations of 5.

Fig. 8 Δδ^RS (δ_R − δ_S) values for the MPA esters of 5 and 3a.

Analysis of the stereochemistry of 3a provided further information to support the absolute configuration of C-7–C-9 as deduced from 5. The 1D NOESY experiment for 3a was performed. Irradiation of H-9 (δ_H 3.96, dt) caused significant enhance of H-7 (δ_H 3.68, td) and one acetal methyl group (α-methyl: δ_H 1.40, s), while H-9 was enhanced when irradiated at H-7. On the other hand, irradiation of H-8 (δ_H 3.87, m) caused obvious enhance of the other acetal methyl group (β-methyl: δ_H 1.43, s). These correlations defined the relative configuration for C-7–C-9. Similarly, we applied modified Mosher's method to determine the absolute configuration of C-7. As expected, the 7R was concluded by analysis of the Δδ^RS (δ_3aR − δ_3aS) values of the MPA esters of 3a (Fig. 8).

The absolute configuration for C-3 of 3 was resolved by analysis of the CD spectra, and confirmed by TDFT ECD calculations. Catalytic hydrogenation of 1 and 3 using Pd/C afforded the saturated congeners 1c and 3b, respectively (Scheme 2). Interestingly, the CD spectra of 1c and 3b are almost mirror images of each other (Fig. 9), which suggests a 3R configuration for 3b, opposite to that of 1c (3S), as the other chiral centers (C-7–C-9) are relatively far from the chromophore. Indeed, the calculated ECD spectrum of (3R, 7R, 8R, 9S)-3b match the experimental spectrum of 3b (Fig. 9). Therefore, the absolute configuration of 3 was established as shown in Fig. 1.

Scheme 2 Hydrogenation of 1 and 3.

Fig. 9 The experimental ECD spectra of 1c and 3b, and the calculated ECD spectrum of (3R, 7R, 8R, 9S)-3b.

Hyalodendriellin F (6) was also isolated as an acetonide of RAL, which showed the same molecular formula as that of 5. Comparison of the NMR data (Table 3), indicated that 6 was an analogue of 5, however the position of the acetonide group differed. Detailed analysis of the ¹H–¹H COSY spectrum revealed that the acetonide group was located at C-7/C-9, as H-8 (δ_H 3.18, td) correlated to 8-OH (δ_H 4.50, d). This also supported by the long-range HMBC correlation seen from the β-acetal methyl (δ_H 1.17, s) to C-7 (δ_C 71.4) and C-9 (δ_C 71.6) (Fig. 10).

Fig. 10 ¹H–¹H COSY, selected HMBC and NOESY correlations of 6.

The relative configuration for C-7–C-9 of 6 was established by analysis of the ¹³C NMR data, ³J value and NOESY correlations. According to the [¹³C]acetonide method,²⁴ 6 should be a syn-1,3-diol acetonide, which prefers chair conformation, as the acetal methyl groups resonate at δ_C 19.2 (axial) and 29.2 (equatorial), respectively, whereas the two acetal methyl groups of the anti-1,3-diol acetonides are nearly identical (ca. 25 ppm). The large coupling constants (³J_H-7/H-8 = ³J_H-8/H-9 = 9.6 Hz) between H-8, H-7 and H-9, revealed these protons to be axial. This was confirmed by the NOESY correlations (Fig. 10) observed between the axial acetal methyl (δ_H 1.38, s), H-7, and H-9, and between H-8, and H-5 (δ_H 1.31, m). The absolute configuration of C-8 was also determined by the modified Mosher's method. As shown in Fig. 11, the Δδ^RS values for the MPA esters of 6 indicated the 8R configuration. According to the established relative configuration, the 7R and 9S configuration for C-7 and C-9 was proposed.

Fig. 11 Δδ^RS (δ_6R − δ_6S) values for the MPA esters of 6.

Hyalodendriellin D (4) was isolated as a minor metabolite, with the molecular formula C₂₀H₂₈O₈ as deduced from HRESIMS. Inspection of the ¹H NMR spectrum revealed it to be an analogue of 3. However, the chemical shifts and coupling constants for C-7–C-9 were different, suggesting a different stereochemistry for the triol unit. The structure of 4 was deduced by chemical conversions.

Initially, we postulated that 4 could be the 4,5-dihydro derivative of 1. Catalytic hydrogenation of 1 yield the 4,5-dihydro (1b) and 4,5,11,12-tetrahydro (1c) derivatives (Scheme 2). However, 4 was not identical to that of 1b as revealed by ¹H NMR and HPLC analyses. It turned out to be the precursor of 6, as hydrolysis of 6 using dilute hydrochloric acid afforded 4. The absolute configuration for C-3 of 4 was determined to be S by TDDFT ECD calculations, as the calculated ECD spectrum of (3S, 7R, 8R, 9S)-4 matched that of the experimental one (Fig. 12). Thus, the 3S configuration for 6 was also deduced.

Fig. 12 The experimental ECD spectrum of 4, and the calculated ECD spectrum of (3S, 7R, 8R, 9S)-4.

In this study, we isolated six 14-membered RALs, hyalodendriellins A–F (1–6), each containing a 7,8,9-triol moiety and a trans 11,12-double bond. Hyalodendriellins A (1), B (2), D (4), and F (6) are 3S-configurated, while hyalodendriellins C (3) and E (5) possess a 3R configuration. According to a recent review, 119 of 14-membered RALs and related compounds have been reported up to the year of 2014, and more than 2/3 of them (86/119) contain a 3S chiral center, while the rest (33/119) possess a 3R configuration, such as monordens, monocillins, and pochonins.¹ However, the co-occurrence of RALs with different stereochemistry at C-3 in the same fungus is rare to the best of our knowledge. RALs are biosynthesized by iterative polyketide synthases (iPKSs), and iPKSs that produce the 14-membered RALs have been characterized in the producer fungi^25,26 and reconstituted both in vivo by heterologous expression in yeast and in vitro using isolated recombinant iPKS enzymes.^11,12,27 Interestingly, Zhou and co-workers found that Rdc1, one of the fungal iPKSs that involve in the biosynthesis of radicicol in Pochonia chlamydosporia, was tolerant of the opposite stereochemistry of the terminal hydroxyl nucleophile in macrolactonization, which resulted in the formation of epimers at C-3.¹¹ Thus, the metabolites with different absolute configuration at C-3 isolated in the present study should be the enzymatic products.

Structurally, hyalodendriellins B (2) and D (4) are the 7-hydroxy analogs of hamigeromycins B and A, respectively, which were previously isolated from the soil fungus Hamigera avellaneda BCC 17816.²⁸ Interestingly, hyalodendriellins E (5) and F (6) with an acetonide group were also obtained. Acetonides are generally recognized as artifacts generated during the extraction and/or isolation stage by reacting with acetone. So far only four RAL₁₄ acetonides have been reported, including caryospomycin A from Caryospora callicarpa,⁹ cochliomycins A and B from Cochliobolus lunatus,²⁰ and paecilomycin H from Paecilomyces sp. SC0924.²⁹ However, only cochliomycins A and B were claimed to be natural. In our case, though we dissolved 3 in acetone, mixed with silica gel, and left at 40 °C for a week, 5 was not detected. However, we cannot rule out the possibility that 5 and 6 are artifacts, as both compounds were co-isolated with their precursors.

Previously, the CD behaviors of E/Z-zearalenone and their 7α-/7β-hydroxy congeners were studied.³⁰ However, the influence of the 7,8,9-trihydroxy groups and the Δ⁵-double bond on the CD spectra has not been investigated. The CD spectra of the isolated compounds (1–6), the acetonide derivatives (1a, 3a), as well as the hydrogenation products of 1 (1b, 1c) and 3 (3b) were measured (in MeOH). As shown in Fig. S1 (see ESI†), compound 1 displayed a weak negative band at 340 nm (Δε = −0.28), a strong negative band at 270 nm (Δε = −3.60), and an intense positive band at 228 nm (Δε = +6.75), which resembled those of E-zearalenone and its 7α-hydroxy derivative (α-zearalenol),³⁰ implying the 5,6-double bond, and 8β,9α-hydroxy groups in 1 do not change the absolute conformation of the absorbing moiety as compared to that of α-zearalenol. Indeed, the 5,6-dihydro derivative of 1 (1b) showed a similar CD curve as that of 1. The CD spectrum of 1a is also of the same shape as that of 1, suggesting the 7,8-acetonide group does not introduce any particular steric hindrance. The saturated congener of 1 (1c) exhibits cotton effects at ca. 200 (Δε = +8.21), 226 (Δε = −1.58), 237 (Δε = +0.89), 261 (Δε = −5.90), and 316 (Δε = −1.25) nm, which are similar to that of zeranol,^30,31 and hypsochromically shifted as compared to 1. The ECD spectrum of 2 was similar to 1b, but the maxima for 2 are shifted bathochromically for the first two bands as compared to 1b, i.e. by 19 nm (from 336 to 355 nm), and 3 nm (from 307 to 310 nm), while shifted hypsochromically for the third (from 269 to 267 nm) and fourth band (from 230 to 225 nm). 1b and 4 are 7-epimers to each other, and their CD spectra are similar. However, compound 3, possessing a 3R configuration, shows a completely different CD (Fig. S2†), in which the strong negative band at around 270 nm, as found in 1b and 4, disappears, while new positive band at 247 nm (Δε = +4.01) and a weak negative one at 214 nm (Δε = −0.21) appear.

The CD spectrum of the 7,8-acetonide of 3 (i.e. 5) displays a similar shape as that of 3, though a weak negative peak at 275 nm (Δε = −0.71) appears, and the intensity for the negative peak at 210 nm obviously increases. On the contrary, the 8,9-acetonide of 3 (3a) gives a complete different CD spectrum, in which cotton effect has opposite sign at 237 nm (Δε = −2.36), indicating it adopts a quite different conformation for the macrocyclic ring. Similarly, the 7,9-acetonide of 4 (i.e. 6) shows a different CD spectrum to that of 4, while the former has similar CD to that of 3, to some extent. Taken together, it seems that the 8,9-, or 7,9-acetonide group dramatically changes the conformation of the macrocyclic ring (3a vs. 3; 6 vs. 4), while the 7,8-acetonide group does not (1a vs. 1; 5 vs. 3).

The isolated metabolites were evaluated for their antinematodal activities against the nematodes Caenorhabditis elegans and Meloidogyne incognita. Among them, hyalodendriellin A (1) exhibited moderate activity against C. elegans and M. incognita with LC₅₀ values of 29.9 and 59.8 μM, at 48 h, respectively. However, the other tested compounds were inactive (LC₅₀ > 200 μM) (Table 4). It seems that the 5,6-double bond (as only found in 1) is important for the antinematodal activity, while those without such function group do not exhibit any significant activity. Similarly, caryospomycins A–C, all containing a C5/C6-double bond, were found to have moderate activity against the nematode Bursaphelenchus xylophilus.⁹

Table 4 Antinematodal activity of the isolated compounds

Compound^a	Caenorhabditis elegans		Meloidogyne incognita
	Time (h)	LC50 (μM)	Time (h)	LC50 (μM)
a The other tested compounds were inactive (LC50 > 200 μM).b Positive control.
1	12	43.15	12	90.37
	24	33.16	24	67.77
	48	29.87	48	59.80
Avermectin^b	12	4.45	12	4.30
	24	3.84	24	3.72
	48	1.75	48	2.04

---

In addition, compounds 1–3, 5 and 6 were evaluated for their inhibition against the fourth-instar larvae of the mosquito Aedes aegypti. Hyalodendriellin C (3) displayed moderate larvicidal activity with LC₅₀ of 117.52 μg mL⁻¹, while the other tested compounds did not exhibit any significant activity at the highest tested concentration of 200 μg mL⁻¹ (Table 5).

Table 5 Larvicidal activity of the isolated compounds against A. aegypti

Compound^a	LC50 (μg mL^−1) (95% CL)	LC90 (μg mL^−1) (95% CL)	Slope ± SD	x^2
a The other tested compounds were inactive (LC50 > 200 μg mL^−1).b Positive control.
3	117.52 (104.14–135.10)	239.54 (194.82–334.89)	4.14 ± 0.57	5.99
Rotenone^b	3.49 (2.85–4.19)	10.36 (8.04–15.22)	2.71 ± 0.36	6.43

---

In vitro cytotoxic assay, revealed the isolated metabolites were non-cytotoxic against the five human carcinoma cell lines (HCT-116, HepG2, BGC-823, NCI-H1650 and A2780) with IC₅₀ > 10 μM. This was not unexpected, as the 14-membered RALs with potent cytotoxicity commonly contained an enone group, such as 4-O-demethylhypothemycin,³² monocillin II,³³ and 5Z-7-oxo-zeaenol.⁴

None of the isolated metabolites showed inhibitory effect against the tested bacteria of Agrobacterium tumefaciens, Bacillus subtilis, Ralstonia solanacearum, and Xanthomonas vesicatoria (MIC > 125 μM) and against the spore germination of Magnaporthe oryzae (IC₅₀ > 100 μg mL⁻¹).

Conclusions

In conclusion, six new 14-membered RALs, hyalodendriellins A–F (1–6), were isolated from the endophytic fungus Hyalodendriella sp. Ponipodef12 associated with the hybrid ‘Neva’ of Populus deltoides Marsh × P. nigra L. Their structures feature a 2-hydroxy-3,4-dimethoxybenzoic acid fused to a macrocyclic moiety, and they commonly contain an E double bond locating at C-11/C-12 and a 7,8,9-triol moiety. The absolute configuration of the new compounds were unambiguously determined by chemical conversion, modified Mosher's method, and TDDFT ECD calculations. Interestingly, hyalodendriellins A (1), B (2), D (4), and F (6) possess a 3S configuration, while hyalodendriellins C (3), and E (5) are 3R-configurated. To the best of our knowledge, RALs with different stereochemistry at C-3 have not been reported from the same fungus. Hyalodendriellin A (1) exhibited moderate antinematodal activity against C. elegans and M. incognita, while hyalodendriellin C (3) showed larvicidal effect against the fourth-instar larvae of the mosquito A. aegypti.

Experimental section

General experimental procedures

UV spectra were recorded on a TU-1810 UV-VIS spectrophotometer (Beijing Persee General Instrument Co., Ltd., Beijing, China). Optical rotations were recorded on a Rudolph Autopol IV automatic polarimeter (Rudolph Research Analytical, New Jersey). Circular dichroism (CD) spectra were recorded on a JASCO J-810 CD spectrometer (JASCO Corp., Tokyo, Japan). Infrared (IR) spectra were measured on a Thermo Nicolet Nexus 470 FT-IR spectrometer (Thermo Electron Scientific Instrument Crop., Wisconsin). High resolution electrospray ionization mass spectrometry (HRESIMS) spectra were recorded on a Bruker Apex IV FTMS instrument (Bruker Daltonics, Bremen, Germany) or an LC1260-Q-TOF/MS 6520 machine (Agilent Technologies, CA, USA). ¹H, ¹³C, and 2D NMR spectra were measured on an Avance 400 or 600 NMR spectrometer (Bruker BioSpin, Zürich, Switzerland). Chemical shifts were expressed in δ (ppm) referenced to the solvent residual peaks at δ_H 2.50/δ_C 39.5 for DMSO-d₆, δ_H 7.26/δ_C 77.0 for CDCl₃, or δ_H 3.31/δ_C 49.0 for CD₃OD and coupling constants (J) in hertz. Silica gel (200–300 mesh, Qingdao Marine Chemical Inc., Qingdao, China), and Sephadex LH-20 (Pharmacia Biotech, Uppsala, Sweden) were used for column chromatography. Medium pressure liquid chromatography (MPLC) separation was carried out on an Eyela-VSP-3050 instrument (Tokyo Rikakikai Co., Tokyo, Japan). HPLC-DAD analysis was performed using a Shimadzu LC-20A instrument with a SPD-M20A photodiode array detector (Shimadzu Corp., Tokyo, Japan) and an analytic C₁₈ column (250 mm × 4.6 mm i.d., 5 μm; Phenomenex Inc., Torrance, California). Semi-preparative HPLC separation was carried out on an Lab Alliance instrument (Scientific Systems Inc., State College, Pennsylvania) equipped with a Series III pump (flow rate: 3 mL min⁻¹) and an UV detector (Mode 201) using a Prevail C₁₈ column (250 mm × 10 mm, 5 μm, GRACE Corporate, Columbia, Maryland). The precoated silica gel GF-254 plates (Qingdao Marine Chemical Inc.) were used for analytical TLC. Spots were visualized under UV light (254 or 356 nm) or by spraying with 10% H₂SO₄ in 95% ethanol followed by heating.

Fungal material

The fungus was isolated from the healthy stems of the ‘Neva’ hybrid of P. deltoides Marsh × P. nigra L. previously,^16,34 and identified as Hyalodendriella sp. Ponipodef12 by morphological examination and analysis of the rDNA gene internal transcribed spacer sequence (GenBank accession number: HQ731647). A voucher specimen was deposited in Department of Plant Pathology, China Agricultural University.

Fermentation, extraction, and isolation

The fungus was grown on potato dextrose agar plates at 25 °C for 8–10 days. Then, 4–5 plugs of agar medium (0.5 cm × 0.5 cm) with fungal hyphae were transferred to a 250 mL Erlenmeyer flask containing 100 mL potato dextrose broth and incubated on a rotary shaker at 150 rpm and 25 °C for 7 days to produce the seed culture. The scale-up fermentation was carried out in 60 Erlenmeyer flasks (1 L) each containing 150 g of rice and 150 mL of distilled water. Each flask was inoculated using a seed culture. The fungus was grown statically at 25 °C for about 45 d before harvest.

The fungal hyphae along with rice were collected, dried and ground, followed by exhaustive maceration with MeOH (5 × 10 L) at room temperature. After filtration, the filtrate was concentrated under vacuum at 40 °C to give a brown residue, which was suspended in water and sequentially partitioned with petroleum ether, EtOAc, and n-BuOH to give their corresponding fractions. The EtOAc extract (82.8 g) was used for further investigation.

The EtOAc extract was subjected to column chromatography (CC) over silica gel (200–300 mesh) eluting with petroleum ether–acetone (8 : 1) to obtain six fractions (Frs. A–F). Fraction D (19.0 g) was separated by vacuum liquid chromatography (VLC) over silica gel eluting with a gradient of petroleum ether–acetone (100 : 0–0 : 100) to yield seven subfractions (D1–D7). Subfraction D3 was subjected to CC over Sephadex LH-20 eluting with CHCl₃–MeOH (1 : 1) to afford four subfractions (D3.1–D3.4). Subfraction D3.2 was purified by semi-preparative HPLC using MeOH–H₂O (65 : 35) as eluent to afford 5 (10.2 mg) and 6 (8.0 mg). Subfraction D6 was also chromatographed over Sephadex LH-20 using CHCl₃–MeOH (1 : 1) to obtain four further subfractions (D6.1–D6.4), among which subfractions D6.2 and D6.3 were purified by semi-preparative HPLC eluting with MeOH–H₂O (55 : 45) and ACN–H₂O (23 : 77) to afford 3 (33.0 mg), and 4 (0.8 mg), respectively. Fraction E (13.0 g) was processed in the same manner as that of fraction D to obtain five subfractions (E1–E5). Subfractions E4 and E5 were further purified by CC over Sephadex LH-20 eluting with CHCl₃–MeOH (1 : 1) and followed by semi-preparative HPLC using MeOH–H₂O (60 : 40) and MeOH–H₂O (50 : 50) as eluent to afford 2 (6.0 mg) and 1 (8.5 mg), respectively.

Hyalodendriellin A (1). Pale yellow oil; [α]²⁶_D +16.1 (c 0.3, MeOH); UV (MeOH) λ_max (log ε) 224 (4.08), 264 (3.53), 320 (3.43) nm; IR ν_max 3340, 2987, 2931, 2886, 1650, 1596, 1475, 1358, 1308, 1248, 1227, 1202, 1062, 1008, 965, 831 cm⁻¹; ECD (c = 6.34 × 10⁻⁴ M, MeOH) λ (Δε) 228 (+6.75), 270 (−3.60), 299 (+0.17), 340 (−0.28) nm; ¹H NMR (CDCl₃ or DMSO-d₆, 400 MHz) and ¹³C NMR (CDCl₃ or DMSO-d₆, 100 MHz) data see Table 1; HRESIMS m/z 417.15153 [M + Na]⁺ (calcd for C₂₀H₂₆O₈Na, 417.15199).

Hyalodendriellin B (2). Pale yellow oil; [α]²⁶_D +12.2 (c 0.47, MeOH); UV (MeOH) λ_max (log ε) 230 (3.66) nm, 310 (3.02) nm; IR ν_max 3395, 2924, 2854, 1674, 1618, 1598, 1570, 1451, 1368, 1223, 1211 cm⁻¹; ECD (c = 6.34 × 10⁻⁴ M, MeOH) λ (Δε) 225 (+12.26), 267 (−4.49), 310 (+2.49), 355 (−1.10) nm; ¹H NMR (DMSO-d₆, 400 MHz) and ¹³C NMR (DMSO-d₆, 100 MHz) data see Table 1; HRESIMS m/z 395.17021 [M + H]⁺ (calcd for C₂₀H₂₇O₈, 395.17004), 417.15206 [M + Na]⁺ (calcd for C₂₀H₂₆O₈Na, 417.15199).

Hyalodendriellin C (3). White needle-like crystal; mp 166–167 °C; [α]²⁶_D +96.1 (c 0.875, MeOH); UV (MeOH) λ_max (log ε) 230 (3.77), 310 (3.00) nm; IR ν_max 3396, 2931, 2854, 1708, 1648, 1593, 1452, 1360, 1313, 1245, 1204, 1115, 1050, 1018 cm⁻¹; ECD (c = 6.31 × 10⁻⁴ M, MeOH) λ (Δε) 214 (−0.21), 232 (+4.13), 247 (+4.01), 278 (+0.12), 294 (+0.15), 328 (+0.47) nm; ¹H NMR (DMSO-d₆ or CD₃OD, 400 MHz) and ¹³C NMR (DMSO-d₆ or CD₃OD, 100 MHz) data see Table 2; HRESIMS m/z 397.18584 [M + H]⁺ (calcd for C₂₀H₂₉O₈, 397.18569).

Hyalodendriellin D (4). White amorphous powder; [α]²⁴_D −21.6 (c 0.05, MeOH); UV (MeOH) λ_max (log ε) 229 (3.88), 267 (3.42), 318 (3.23) nm; IR ν_max 3448, 2933, 1637, 1597, 1471, 1359, 1255, 1229, 1021 cm⁻¹; ECD (c = 1.26 × 10⁻³ M, MeOH) λ (Δε) 229 (+7.54), 268 (−5.91), 308 (+0.45), 345 (−0.13) nm; ¹H NMR (CD₃OD, 400 MHz) δ 6.68 (1H, dd, J = 15.9, 2.1 Hz, H-12), 6.47 (1H, s, H-15), 5.92 (1H, ddd, J = 15.9, 9.8, 4.0 Hz, H-11), 5.12–5.04 (1H, m, H-3), 3.92 (1H, ddd, J = 8.1, 5.7, 2.2 Hz, H-9), 3.86 (3H, s, 14-OMe), 3.85–3.80 (1H, m, H-7), 3.68 (1H, dd, J = 4.9, 2.1 Hz, H-8), 3.58 (3H, s, 13-OMe), 2.72 (1H, ddd, J = 14.9, 9.8, 8.1 Hz, H-10a), 2.55 (1H, dddd, J = 14.9, 5.8, 3.8, 2.3 Hz, H-10b), 1.89–1.57 (5H, m), 1.50–1.41 (1H, m) (H₂-4, H₂-5, H₂-6), 1.36 (3H, d, J = 6.2 Hz, 3-Me); HRESIMS m/z 395.1725 [M − H]⁻ (calcd for C₂₀H₂₇O₈, 395.1711).

Hyalodendriellin E (5). Pale yellow oil; [α]²⁶_D +44.2 (c 1.57, MeOH); UV (MeOH) λ_max (log ε) 230 (4.09), 306 (3.43) nm; IR ν_max 3398, 2928, 1718, 1651, 1592, 1452, 1368, 1244, 1206, 1053, 1018 cm⁻¹; ECD (c = 5.73 × 10⁻⁴ M, MeOH) λ (Δε) 210 (−1.69), 232 (+4.98), 275 (−0.71), 309 (−0.07) nm; ¹H NMR (DMSO-d₆, 600 MHz) and ¹³C NMR (DMSO-d₆, 150 MHz) data see Table 3; HRESIMS m/z 459.19935 [M + Na]⁺ (calcd for C₂₃H₃₂O₈Na, 459.19894).

Hyalodendriellin F (6). Pale yellow oil; [α]²⁶_D +60.3 (c 0.53, MeOH); UV (MeOH) λ_max (log ε) 228 (3.81), 302 (3.17) nm; IR ν_max 3395, 2928, 1674, 1620, 1569, 1454, 1367, 1205, 1163, 1043 cm⁻¹; ECD (c = 5.73 × 10⁻⁴ M, MeOH) λ (Δε) 212 (−3.83), 232 (+5.71), 253 (+4.49), 279 (−0.27), 288 (−0.08), 316 (−0.25) nm; ¹H NMR (DMSO-d₆, 600 MHz) and ¹³C NMR (DMSO-d₆, 150 MHz) data see Table 3; HRESIMS m/z 459.19938 [M + Na]⁺ (calcd for C₂₃H₃₂O₈Na, 459.19894).

Preparation of the acetonide derivatives

A mixture of 1 (2.0 mg), 2,2-dimethoxypropane (0.4 mL), and p-TsOH (0.4 mg) was stirred at room temperature for 5 h. Saturated aqueous NaHCO₃ (5 mL) was added, and the reaction mixture was extracted with EtOAc (5 mL × 3). The organic layer were combined and concentrated in vacuum to obtain a crude product, which was purified by semi-preparative HPLC (65% MeOH/H₂O) to obtain 1a (1.1 mg). Similarly, the acetonides, including 5 (0.8 mg), and 3a (2.7 mg) were prepared from 3 (5.0 mg).

1a. White amorphous powder; ECD (c = 5.75 × 10⁻⁴ M, MeOH) λ (Δε) 231 (+22.17), 270 (−8.94), 312 (+0.58), 350 (−0.22) nm; ¹H NMR (CDCl₃, 400 MHz) δ 11.29 (1H, s, OH-16), 6.73 (1H, dd, J = 15.9, 2.4 Hz, H-12), 6.42 (1H, s, H-15), 6.00 (1H, ddd, J = 15.4, 7.9, 4.9 Hz, H-5), 5.92 (1H, ddd, J = 15.9, 9.8, 3.4 Hz, H-11), 5.50 (1H, br.dd, J = 15.5, 8.9 Hz, H-6), 5.43 (1H, m, H-3), 4.56 (1H, t, J = 8.5 Hz, H-7), 4.20 (1H, ddd, J = 12.3, 4.9, 2.4 Hz, H-9), 3.874 (3H, s, 14-OCH₃), 3.873 (1H, dd, J = 8.3, 2.3 Hz, H-8), 3.59 (3H, s, 13-OCH₃), 2.85 (1H, dddd, J = 15.5, 4.9, 3.4, 2.4 Hz, H-10a), 2.54 (1H, ddd, J = 16.0, 7.9, 2.7 Hz, H-4a), 2.42 (1H, dddd, J = 16.1, 5.8, 4.9, 1.9 Hz, H-4b), 2.31 (1H, ddd, J = 15.5, 12.3, 9.7 Hz, H-10b), 1.44 (3H, s, β-acetal methyl), 1.42 (3H, d, J = 6.4 Hz, 3-Me), 1.38 (3H, s, α-acetal methyl); HRESIMS m/z 435.20163 [M + H]⁺ (calcd for C₂₃H₃₁O₈, 435.20134), 457.18395 [M + Na]⁺ (calcd for C₂₃H₃₀O₈Na, 457.18329).

3a. White amorphous powder; ECD (c = 5.73 × 10⁻⁴ M, MeOH) λ (Δε) 216 (+0.35), 223 (+0.71), 237 (−2.36), 251 (+1.07), 269 (−0.12), 283 (+0.77), 294 (+0.29), 324 (+1.10), 368 (−0.14) nm; ¹H NMR (CDCl₃, 400 MHz) and ¹³C NMR (CDCl₃, 100 MHz) data, see Table 3; HRESIMS m/z 437.21688 [M + H]⁺ (calcd for C₂₃H₃₃O₈, 437.21699), 459.19884 [M + Na]⁺ (calcd for C₂₃H₃₂O₈Na, 459.19894).

Preparation of the (R)- and (S)-MPA ester derivatives of 1a, 3a, 5, and 6³⁵

To a CH₂Cl₂ solution (600 μL) of 1a (1.0 mg, 0.0023 mmol), (R)-MPA or (S)-MPA (2.0 mg, 0.012 mmol), 4-dimethylaminopyridine (catalytic amount), and N,N′-dicyclohexylcarbodiimide (2.5 mg, 0.012 mmol) were added. The reaction was kept at room temperature for 6 h until all the starting material was consumed. After evaporation of the solvent in vacuum, the crude product was dissolved in MeOH followed by centrifugation. The supernatant was purified by semi-preparative HPLC (80% MeOH/H₂O as the mobile phase) to afford (R)-MPA ester (1aR, 1.0 mg), or (S)-MPA ester (1aS, 1.1 mg). Following the same protocol, (R)-MPA esters of 3a (3aR, 1.0 mg), 5 (5R, 1.0 mg), and 6 (6R, 1.1 mg), and (S)-MPA esters of 3a (3aS, 1.1 mg), 5 (5S, 1.0 mg), and 6 (6S, 0.9 mg) were obtained.

Hydrogenation of 1 and 3

To a solution of 1 (3.0 mg) in MeOH (3 mL) was added Pd/C (0.4 mg), and the mixture was stirred under H₂ at room temperature for 5 h. The suspension was filtered, and the filtrate was concentrated in vacuum to give a crude product, which was subjected to semi-preparative HPLC (eluting with 55% MeOH/H₂O) to obtain the 5,6-dihydro derivative 1b (0.5 mg) and the 5,6,11,12-tetrahydro derivative 1c (1.5 mg). Similarly, 3b (1.3 mg) was prepared from 3 (3.0 mg).

1b. White amorphous powder; ECD (c = 6.31 × 10⁻⁴ M, MeOH) λ (Δε) 230 (+17.59), 269 (−12.41), 307 (+0.12), 336 (−0.44) nm; ¹H NMR (CD₃OD, 400 MHz) δ 6.66 (1H, dd, J = 15.9, 2.3 Hz, H-12), 6.48 (1H, s, H-15), 5.81 (1H, ddd, J = 15.1, 10.6, 3.5 Hz, H-11), 5.13–5.03 (1H, m, H-3), 4.05–3.96 (2H, m), 3.87 (3H, s, 14-OMe), 3.55 (3H, s, 13-OMe), 3.49–3.45 (1H, m), 2.69–2.61 (1H, m, H-10a), 2.50 (1H, dt, J = 14.8, 10.7 Hz, H-10b), 1.93–1.77 (2H, m), 1.68–1.55 (1H, m), 1.51–1.30 (3H, m), 1.37 (3H, d, J = 6.0 Hz, 3-Me); HRESIMS m/z 397.18646 [M + H]⁺ (calcd for C₂₀H₂₉O₈, 397.18569), 419.16821 [M + Na]⁺ (calcd for C₂₀H₂₈O₈Na, 419.16764).

1c. White amorphous powder; ECD (c = 6.27 × 10⁻⁴ M, MeOH) λ (Δε) ca. 200 (+8.21), 226 (−1.58), 237 (+0.89), 261 (−5.90), 284 (−0.24), 316 (−1.25) nm; ¹H NMR (CDCl₃, 400 MHz) δ 12.21 (1H, s, OH-16), 6.40 (1H, s, H-15), 5.15 (1H, dq, J = 11.8, 6.0 Hz, H-3), 4.14 (1H, br.d, J = 11.4 Hz), 3.98–3.90 (1H, m), 3.87 (3H, s, 14-OMe), 3.73 (3H, s, 13-OMe), 3.64 (1H, br.s), 2.98 (1H, td, J = 11.7, 5.2 Hz, H-12a), 2.90 (1H, td, J = 11.7, 5.4 Hz, H-12b), 2.10–1.77 (3H, m), 1.68–1.53 (3H, m), 1.52–1.35 (4H, m), 1.38 (3H, d, J = 6.1 Hz, 3-Me); HRESIMS m/z 399.20215 [M + H]⁺ (calcd for C₂₀H₃₁O₈, 399.20134), 421.18355 [M + Na]⁺ (calcd for C₂₀H₃₀O₈Na, 421.18329).

3b. White amorphous powder; ECD (c = 6.27 × 10⁻⁴ M, MeOH) λ (Δε) 209 (−6.88), 223 (+1.92), 233 (+0.09), 252 (+6.04), 273 (+0.53), 293 (+1.35) nm; ¹H NMR (CD₃OD, 400 MHz) δ 6.39 (1H, s, H-15), 5.26 (1H, m, H-3), 3.82 (3H, s, 14-OMe), 3.73 (3H, s, 13-OMe), 3.74–3.69 (1H, m), 3.64 (1H, dt, J = 7.2, 3.5 Hz), 3.42 (1H, t, J = 5.1 Hz, H-8), 2.74 (2H, dd, J = 9.6, 7.2 Hz, H₂-12), 1.85 (1H, dq, J = 13.5, 6.7 Hz), 1.80–1.57 (6H, m), 1.50 (1H, dtd, J = 13.7, 7.0, 3.1 Hz), 1.34 (3H, d, J = 6.4 Hz, 3-Me), 1.32–1.21 (2H, m); HRESIMS m/z 399.20238 [M + H]⁺ (calcd for C₂₀H₃₁O₈, 399.20134), 421.18396 [M + Na]⁺ (calcd for C₂₀H₃₀O₈Na, 421.18329).

Computation details

Molecular Merck Force Field (MMFF) and DFT/TDDFT calculations were performed with the SYBYL-X 2.0 software package and the Gaussian 09 program package, respectively, using default grids and convergence criteria. MMFF conformational search-generated low-energy conformers within a 10 kcal mol⁻¹ energy window were subjected to geometry optimization using the DFT method at the B3LYP/6-31G(d) level in vacuo. TDDFT ECD calculations of the low-energy conformers (≥1%) were performed at the Pbe0/TZVP level, with the PCM solvent model for MeOH. The number of excited states per conformer was 30. The ECD spectrum of each conformer was simulated by the program SpecDis³⁶ using a Gaussian band shape with 0.3 eV exponential half-width using dipole-velocity computed rotational strengths. The equilibrium population of each conformer at 298.15 K was calculated from the ZPVE-corrected energies using Boltzmann statistics. The Boltzmann-averaged ECD spectra for (3S, 7S, 8S, 9S)-1a, (3S, 5R, 7S, 8R, 9S)-2, (3R, 7R, 8R, 9S)-3b, and (3S, 7R, 8R, 9S)-4 were generated according to the Boltzmann distributions of the lowest energy conformers for each structure. Theoretical ECD spectra were then compared with the experimental ones to determine the absolute configuration. The calculated ECD spectra (y-axes) of 1a, 2, and 4 were multiplied by 1/2, 1/2, and 1/4, respectively, to match the experimental data.

Antinematodal assay

The isolated compounds (except 4) were tested for their antinematodal activity against C. elegans and M. incognita, as described previously.¹⁷ Avermectin was used as the positive control and three replicates were carried out for each treatment. The nematode C. elegans was kindly supplied by Dr Chonglin Yang of the Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, and nematode M. incognita was from Department of Plant Pathology, China Agricultural University.

Larvicidal assay

Compounds 1–3, 5 and 6 were tested for their larvicidal activity against the fourth-instar larvae of A. aegypti as described previously.³⁷ Rotenone was used as the positive control. The eggs of mosquito A. aegypti were obtained from the Department of Vector Biology and Control, Institute for Infectious Disease Control and Prevention, Chinese Centre for Disease Control and Prevention.

Cytotoxic assay

Cytotoxicity was tested against five human cancer cell lines including colon cancer cells (HCT-116), liver hepatocellular carcinoma cells (HepG2), gastric cancer cells (BGC-823), non-small-cell lung carcinoma cells (NCI-H1650), and ovarian cancer cells (A2780) using the microculture tetrazolium method as described previously.¹⁴ All the cells were obtained from the Cell Culture Center, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences.

Antibacterial assay

The antibacterial activity of 1–3, 5 and 6 was evaluated against four plant pathogenic bacteria, including A. tumefaciens ATCC 11158, B. subtilis ATCC 11562, R. solanacearum ATCC 11696, and X. vesicatoria ATCC 11633, by the modified broth dilution colorimetric assay.³⁸ All the bacteria were from the American Type Culture Collection.

Antifungal assay

The antifungal activity of 1–3, 5 and 6 was tested against M. oryzae using the spore germination assay as described previously.¹⁵ The fungal pathogen M. oryzae 131 was obtained from the Department of Plant Pathology, China Agricultural University.

Conflict of interest

The authors declare no competing financial interests.

Acknowledgements

We acknowledge the National Basic Research Program of China (2013CB127805), the Special Fund for Agro-Scientific Research in the Public Interest of China (201203037), the Hi-Tech R&D Program of China (2011AA10A202), and the Chinese Universities Scientific Fund (2016QC047) for financial support.

Notes and references

1. W. Shen, H. Mao, Q. Huang and J. Dong, Eur. J. Med. Chem., 2015, 97, 747–777.
2. W. Nicolas, F. Jean-Gonzague and B. Sofia, Curr. Trends Med. Chem., 2009, 9, 1419–1435.
3. J. Patocka, O. Soukup and K. Kuca, Mini-Rev. Med. Chem., 2013, 13, 1873–1878.
4. S. Ayers, T. N. Graf, A. F. Adcock, D. J. Kroll, S. Matthew, E. J. Carcache de Blanco, Q. Shen, S. M. Swanson, M. C. Wani, C. J. Pearce and N. H. Oberlies, J. Nat. Prod., 2011, 74, 1126–1131.
5. V. Hellwig, A. Mayer-Bartschmid, H. Müller, G. Greif, G. Kleymann, W. Zitzmann, H.-V. Tichy and M. Stadler, J. Nat. Prod., 2003, 66, 829–837.
6. P. Rawlins, T. Mander, R. Sadeghi, S. Hill, G. Gammon, B. Foxwell, S. Wrigley and M. Moore, Int. J. Immunopharmacol., 1999, 21, 799–814.
7. J. Ninomiya-Tsuji, T. Kajino, K. Ono, T. Ohtomo, M. Matsumoto, M. Shiina, M. Mihara, M. Tsuchiya and K. Matsumoto, J. Biol. Chem., 2003, 278, 18485–18490.
8. J. C. Turcotte, P. J. B. Hunt and J. D. Blaustein, Horm. Behav., 2005, 47, 178–184.
9. J. Dong, Y. Zhu, H. Song, R. Li, H. He, H. Liu, R. Huang, Y. Zhou, L. Wang, Y. Cao and K. Zhang, J. Chem. Ecol., 2007, 33, 1115–1126.
10. P.-Y. Dakas, S. Barluenga, F. Totzke, U. Zirrgiebel and N. Winssinger, Angew. Chem., Int. Ed., 2007, 46, 6899–6902.
11. H. Zhou, K. Qiao, Z. Gao, J. C. Vederas and Y. Tang, J. Biol. Chem., 2010, 285, 41412–41421.
12. H. Zhou, K. Qiao, Z. Gao, M. J. Meehan, J. W. H. Li, X. Zhao, P. C. Dorrestein, J. C. Vederas and Y. Tang, J. Am. Chem. Soc., 2010, 132, 4530–4531.
13. J. Zhao, T. Shan, Y. Mou and L. Zhou, Mini-Rev. Med. Chem., 2011, 11, 159–168.
14. D. Lai, A. Wang, Y. Cao, K. Zhou, Z. Mao, X. Dong, J. Tian, D. Xu, J. Dai, Y. Peng, L. Zhou and Y. Liu, J. Nat. Prod., 2016, 79, 2022–2031.
15. T. Shan, J. Tian, X. Wang, Y. Mou, Z. Mao, D. Lai, J. Dai, Y. Peng, L. Zhou and M. Wang, J. Nat. Prod., 2014, 77, 2151–2160.
16. L. Zhong, L. Xu, X. Meng, Y. Peng, Y. Chen, P. Sui, M. Wang and L. Zhou, Afr. J. Biotechnol., 2011, 10, 18174–18178.
17. X. Meng, Z. Mao, J. Lou, L. Xu, L. Zhong, Y. Peng, L. Zhou and M. Wang, Molecules, 2012, 17, 11303–11314.
18. F. Sugawara, K.-W. Kim, K. Kobayashi, J. Uzawa, S. Yoshida, N. Murofushi, N. Takahashi and G. A. Strobel, Phytochemistry, 1992, 31, 1987–1990.
19. M. Isaka, C. Suyarnsestakorn, M. Tanticharoen, P. Kongsaeree and Y. Thebtaranonth, J. Org. Chem., 2002, 67, 1561–1566.
20. C.-L. Shao, H.-X. Wu, C.-Y. Wang, Q.-A. Liu, Y. Xu, M.-Y. Wei, P.-Y. Qian, Y.-C. Gu, C.-J. Zheng, Z.-G. She and Y.-C. Lin, J. Nat. Prod., 2011, 74, 629–633.
21. G. Bringmann, T. Bruhn, K. Maksimenka and Y. Hemberger, Eur. J. Org. Chem., 2009, 2717–2727.
22. I. Ohtani, T. Kusumi, Y. Kashman and H. Kakisawa, J. Am. Chem. Soc., 1991, 113, 4092–4096.
23. J. M. Seco, E. Quiñoá and R. Riguera, Chem. Rev., 2004, 104, 17–118.
24. S. D. Rychnovsky, B. N. Rogers and T. I. Richardson, Acc. Chem. Res., 1998, 31, 9–17.
25. I. Gaffoor and F. Trail, Appl. Environ. Microbiol., 2006, 72, 1793–1799.
26. S. Wang, Y. Xu, E. A. Maine, E. M. K. Wijeratne, P. Espinosa-Artiles, A. A. L. Gunatilaka and I. Molnár, Chem. Biol., 2008, 15, 1328–1338.
27. C. D. Reeves, Z. Hu, R. Reid and J. T. Kealey, Appl. Environ. Microbiol., 2008, 74, 5121–5129.
28. M. Isaka, P. Chinthanom, S. Veeranondha, S. Supothina and J. J. Luangsa-ard, Tetrahedron, 2008, 64, 11028–11033.
29. L. Xu, J. Xue, Y. Zou, S. He and X. Wei, Chin. J. Chem., 2012, 30, 1273–1277.
30. M. Gelo, Z. Raza, V. Šunjić, J. Guo and G. Snatzke, Tetrahedron: Asymmetry, 1991, 2, 1005–1010.
31. G. Snatzke, C. Angeli, E. Decorte, F. Moimas, B. Kojié-Prodié, Z. i. Ruz̆ić-Toroš and V. S̆unjić, Helv. Chim. Acta, 1986, 69, 734–748.
32. J. L. Wee, K. Sundermann, P. Licari and J. Galazzo, J. Nat. Prod., 2006, 69, 1456–1459.
33. H. Wei, L. Xu, M. Yu, L. Zhang, H. Wang, X. Wei and Y. Ruan, ChemBioChem, 2012, 13, 465–475.
34. L. Zhong, Y. Zhou, S. Gao, L. Xu, J. Zhao, T. Shan, W. He and L. Zhou, Afr. J. Microbiol. Res., 2011, 5, 3924–3929.
35. D. Lai, Y. Li, M. Xu, Z. Deng, L. van Ofwegen, P. Qian, P. Proksch and W. Lin, Tetrahedron, 2011, 67, 6018–6029.
36. T. Bruhn, A. Schaumloeffel, Y. Hemberger and G. Bringmann, Chirality, 2013, 25, 243–249.
37. J. Tian, X. C. Liu, Z. L. Liu, D. Lai and L. Zhou, Pest Manage. Sci., 2016, 72, 961–965.
38. S. Lu, W. Sun, J. Meng, A. Wang, X. Wang, J. Tian, X. Fu, J. Dai, Y. Liu, D. Lai and L. Zhou, J. Agric. Food Chem., 2015, 63, 3501–3508.

---

Footnotes

† Electronic supplementary information (ESI) available: The CD spectra of 1–6, 1a–1b, and 3a; ECD calculation data for 1a, 2, 3b, and 4; MS, IR, 1D and 2D NMR spectra of 1–6. See DOI: 10.1039/c6ra24009g

‡ These authors contributed equally.

---