Biologically Active Tetranorditerpenoids from the Fungus Sclerotinia homoeocarpa, Causal Agent of Dollar Spot in Turfgrass

Abstract

Nine new tetranorditerpenoid dilactones (2–10), together with two previously reported norditerpenoids dilactones (1, 11), and two known putative biosynthetic intermediates oidiolactone-E (12) and 13 were isolated from an ethyl acetate extract of a culture medium of Sclerotinia homoeocarpa. Structures and absolute configuration of these compounds were determined by spectroscopic methods and confirmed by X-ray crystallographic analysis of representative compounds. Compounds were evaluated for herbicidal, antiplasmodial and cytotoxic activities. Compounds 1, 2, 6, 7, 11 were more active as growth inhibitors in a duckweed bioassay (I₅₀ values of 0.39 - 0.95 µM) than more than half of 26 commercial herbicides previously evaluated using the same bioassay. Some of these compounds exhibited strong antiplasmodial activities as well, but they also had cytotoxic activity thus precluding them as potential antimalarial agents.

Malaria continues to be a major cause of morbidity and mortality in many parts of the tropics and subtropics. Every year about 500 million people become ill with malaria and over a million people, most of them young children, die of this disease.¹ Wide-spread resistance to first-line antimalarial drugs has hampered the effective control of this disease. Many efforts are underway to develop new classes of antimalarials to counter this trend.² One of these involves searching for compounds to inhibit unique metabolic pathways in the apicoplast.^2,3

Plasmodium parasites, which cause malaria in humans, contain an organelle called the apicoplast.⁴ The apicoplast is very similar to plastids (chloroplasts are one of several plastid forms) of plants and is believed to have been acquired by the engulfment of an ancestral alga and retention of the algal plastid. The apicoplast is essential for the survival of parasites, and it contains many plant-like metabolic pathways such as essential amino acid, heme, and type II fatty acid biosynthesis not present in the vertebrate hosts of malaria parasites.⁴ Several studies have been initiated to evaluate antimalarial activity of herbicides and natural phytotoxins which are known to inhibit metabolic pathways of the plastid in plants.^5–7

A number of plant pathogenic fungi are known to release phytotoxins that disrupt biological processes in plants.^8–10 Some of these have also been shown to inhibit metabolic pathways in the plastid;¹¹ thus, plant pathogens could be a potential source of antimalarial compounds.

As part of a program to search for antimalarial compounds from natural sources, we initiated a project to screen phytopathogenic fungi for antimalarial activity. Sclerotinia homoeocarpa is the causal agent of “dollar spot,” the most prevalent disease of turfgrass in North America.¹² Previous studies on S. homoeocarpa led to the isolation of two tetranorditerpene dilactones with potent phytotoxic activity.¹³

In preliminary studies, S. homoeocarpa was isolated from an infected grass leaf and was grown in potato/dextrose (PD) broth. The ethyl acetate extract of PD broth showed herbicidal activity, as well as in vitro antimalarial activity, against chloroquine sensitive (D-6) and chloroquine sensitive resistant (W-2) strains of Plasmodium falciparum. Chemical investigation of this extract led to the isolation of eleven tetranorditerpene dilactones and two putative biogenetic precursors with partially and fully opened lactones.

A number of compounds of this class have been previously isolated from fungi.^13–20 These compounds were very similar to podolactones from the plant family Podocarpaceae.²¹ Norditerpene dilactones have shown herbicidal,^{13,15,19,22,23} antitumor,²⁰ antifungal,^15,17,24,25 insecticidal,²⁷ anti-feedant,^28,29 and cytokine production inhibitory activities.¹⁴

Nine compounds (2–10) isolated in the present study, all norditerpenoid dilactones, have not been previously reported. Compound 1 had been identified from an isolate of S. homoeocarpa from Japan, and its herbicidal activity was described.¹³ Spectroscopic data of compound 9 resembled those of a previously reported compound to which a different structure was assigned.^19,30 Compound 11 was first reported by synthesis³⁰ and was later isolated from an unidentified fungus.²⁰ Compound 12 and 13 were reported from Oidiodendron truncata ¹⁴ and from an Acrostalagmus species,¹⁸ respectively. Most of the compounds described herein showed both phytotoxic and antiplasmodial activity; however, they also had a relatively high level of general cytotoxcity.

Results and Discussion

Spectroscopic data indicated that a majority of the compounds isolated from S. homoeocarpa belong to the teteranorditerpene dilactone class.^15,25,30 The molecular weight of compound 1 was determined to be C₁₆H₁₆O₆ by HRESIMS. The NMR data of 1 agreed well with those reported for N-1400A, which was previously isolated from an isolate of S. homoeocarpa.¹³ Since the assignments of ¹H and ¹³C NMR data for N-1400A were not previously reported, they are presented in Tables 1 and 2. The relative configuration of this compound was also determined for the first time. The ROESY spectrum of compound 1 showed correlations of CH₃- 15 to H-3, H-5, and H-6, indicating that they are on the same face. However, the absence of any ROESY correlation of CH₃-16 prevented assignment of the relative configuration of all the asymmetric centers of this compound by spectroscopic methods. X-ray crystallography (Figure 1) was used to establish the relative configuration of all the asymmetric centers.

Table 1.

¹H NMR Spectroscopic Data (400 MHz, CDCl₃/ CD₃OD) for Compounds 1–10.

	1	2	3	4	5	6	7	8	9	10
Position				δH (J in Hz)
1	6.37, d (10.0)	6.23, d (6.0)	1.68, m, 1.93, m	1.51, dd (13.5, 7.5) 2.32, dd (13.5, 9.0)	3.90, t (6.0)	6.61, d (9.6)	6.33, dd (9.6, 3.2)	1.66, m 1.83, m	1.34, dd (13.2, 7.2) 2.20, dd (13.6, 9.2)	3.78, t (6.4)
2	5.83, dd (10.0, 5.6)	5.90, ddd (6.8, 6.0, 2.0)	1.75, m, 1.89, m	3.94, m	1.63, m, 1.82, m	6.07, dd (10.0, 5.6)	5.90, ddd (9.6, 6.8, 2.4)	1.56, m 1.83, m	3.81, m	1.55, m, 1.76, m
3	4.16, d (5.6)	2.23, dd (18.8, 6.8) 2.79, d (18.8, 2.0)	4.25, dd (8.8, 3.2)	1.86, dd (13.5, 5.0) 2.05, dd (13.5, 12.5)	1.52, ddd (14.4, 8.4, 6.2), 2.24, ddd (14.0, 8.0, 5.6 )	4.42, d (5.6)	2.83, ddd (18.0, 3.6, 2.4), 2.26, dd (18.0, 7.0)	4.31, dd (11.2, 4.0)	1.68, dd (13.6, 4.8) 1.96, t (13.2)	1.38, ddd, (14.4, 6.4, 6.4), 2.13, ddd (14.0, 8.0, 5.6)
5	2.02, d (4.8)	1.98, d (4.8)	2.05, d (4.8)	1.91, d ( 5.0)	1.81, d (4.4)	2.29, d (5.2)	2.08, d (4.6)	2.01, d (4.8)	1.81, d (5.2)	1.75, d (4.8)
6	4.74, d (4.4)	5.01, d (4.0)	5.08, dd (4.8, 1.2)	5.01, dd (5.0, 1.5)	4.96, dd (4.8, 1.2)	5.01, t, (5.0)	5.07, t (4.4)	4.94, brt (4.4)	4.89, td, (4,4, 1.2)	4.89, dt (4.8, 1.2)
7	3.81, d (1.2)	3.9, s	4.00, br.s	3.97, d (1.5)	3.92, br.s	6.24, m	6.25, m	6.10, m	6.08, m	6.05, m
11	6.08, s	6.23, s	6.00, s	6.04, s	6.56, s	6.00, d (1.6)	5.97, d (1.6)	5.71, br.s	5.62, d ( 1.6)	6.09, d (2.0)
13	3.97, d (12.8) 4.57, d (12.8)	4.13, d (12.4) 4.72, d (12.4)	4.12, d (12.8) 4.77, d (12.8)	4.12, d (12.4) 4.74, d (12.4)	4.07, d (12.4) 4.69, d (12.4)	4.91, d (13.6), 5.01, dt (14.0, 2.0)	4.89, d (13.2) 1.8), 4.99, dt (13.2, 2.0)	4.78, d (13.5) 4.85, brd (13.5)	4.73, d (14.0) 4.81, dt (13.6, 2.0)	4.70, d (13.5) 4.77, dt (13.5, 2.0)
15	1.23, s	1.34, s	1.30, s	1.36, s	1.23, s	1.49, s	1.38, s	1.22, s	1.21, s	1.11, s
16	1.26, s	1.12, s	1.17, s	1.25, s	1.14, s	1.47, s	1.14, s	1.05, s	1.06, s	0.98, s

Table 2.

¹³C NMR Spectroscopic Data (δ) (100 MHz, CDCl₃/CD₃OD) for Compounds 1–10.

	1	2	3	4	5	6	7	8	9	10
position					δC
1	133.6	128.7	27.7	39.1	68.9	135.1	130.1	28.6	36.5	69.1
2	129.4	127.7	26.7	63.5	29.2	129.5	127.4	26.1	64.2	29.2
3	69.5	31.8	65.7	36.7	27.6	69.6	31.3	65.1	39.8	27.0
4	46.3	43.0	48.1	41.9	42.3	47.3	43.6	48.6	43.1	42.8
5	48.5	45.0	43.5	41.9	43.8	52.7	49.2	48.0	46.3	47.8
6	72.2	73.6	72.3	72.6	72.6	71.7	72.6	72.2	72.4	71.6
7	53.3	53.5	53.4	53.2	53.6	122.4	122.2	121.9	122.5	122.1
8	55.6	55.4	55.4	54.7	60.0	132.6	132.3	131.8	132.2	132.6
9	154.4	154.4	159.0	158.1	156.3	155.8	155.1	159.0	159.4	157.6
10	37.2	37.0	35.6	36.5	40.9	36.9	36.5	34.4	36.3	40.2
11	117.4	117.9	117.1	118.1	120.2	111.7	111.9	111.9	112.6	113.8
12	163.1	162.6	164.1	163.3	164.3	164.8	163.6	164.3	165.1	165.1
13	72.2	72.0	72.3	72.0	72.3	69.9	69.7	69.7	70.3	69.9
14	178.9	181.6	181.2	181.3	181.2	179.6	181.9	182.1	182.4	181.4
15	24.5	25.0	15.9	22.5	23.6	25.0	25.4	15.9	23.4	23.9
16	28.4	24.9	27.2	29.1	17.8	26.8	22.9	26.9	27.9	16.7

Figure 1.

Single-crystal X-ray structures of compounds 1, 3, 6, and 10.

The molecular formula of compound 2 was determined to be C₁₆H₁₆O₅ by HRESIMS. Comparison of ¹H NMR spectra of compounds 2 and 1 indicated that the signal due to the oxygenated methine at C-3 of the latter had been replaced by up-field signals at 2.23 and 2.79 in the former. Comparison of the ¹³C NMR spectra of 2 and 1 showed the replacement of the signal due to C-3 (δ 69.4) of the latter by an up-field signal at δ 31.8 in the former, and also up-field shifts of the carbon signals that were β and γ to C-3. These observations suggested that compound 2 was the 3-dehydroxy analog of 1. HMBC correlations between C-3 methylene signals (δ 2.23 and 2.79) and C-1 (δ 128.7), C-2 (δ 127.7), C-4 (δ 43.0), C-5 (δ 45.0), C-14 (δ 181.6), and C-15 (δ 25.0) supported this structure. The COSY, HMQC, and HMBC correlations of the rest of the molecule were similar to those observed for compound 1, further confirming this structure for compound 2. ROESY correlations of CH₃-15 to H-5, H-6 indicated that they were on the same face, and correlations of CH₃-16 to H-7 showed that they were on the opposite face. These results established that 2 had the same relative configuration as compound 1.

HRESIMS data indicated the molecular formula C₁₆H₁₈O₆ for compound 3. The ¹H NMR and ¹³C NMR spectra of 3 were also similar to those of compound 1, except for signals due to protons in ring A. The major difference was the replacement of signals due to two olefinic protons by two methylene groups [2H-1: δ 1.68 (m), and 1.93 (m), 2H-2: δ 1.75 (m), and 1.89 (m)]. Correlations in the COSY spectrum showed that these two methylene groups were attached to an oxygenated methine (δ 4.25) to form a -CH₂CH₂CHOH- moiety. Although no HMBC cross-peaks were observed between the oxygenated methine proton at δ 4.25 and any carbon, strong correlations between 15-CH₃ (δ 1.30) and the carbon signal of this methine (δ 65.7), in addition to C-2 (δ 26.7), C-4 (δ 48.1), C-5 (δ 43.5), and C-14 (δ 181.2) indicated that it was located at C-3. This observation, in combination with the molecular formula, suggested that 3 was the 1,2-dihydro- analog of 1. Additional HMBC correlations from 2H-1 (δ 1.68 and 1.93) to C-2 (δ 26.7), C-3 (δ 65.7), C-5 (δ 43.5), C-9 (δ 159.0), C-10 (δ 35.6), and C-16 (δ 27.2), as well as 2H-2 (δ 1.75 and 1.89) to C-1 (δ 27.7), C-3 (δ 65.7), C-4 (δ 48.1), and C-10 (δ 35.6) were supportive of this structure. The HMBC correlations of the rest of the molecule were identical to those of 1. The coupling constants of H-3 (J = 8.8, 3.2 Hz) indicated that it was axial. When the OH group at C-3 is equatorial it shows a strong interaction with the 15-CH₃ group. This was evident from the significant up-field shift of the ¹³C signal (δ 15.9) compared to that of compounds 2, or 1 (δ 24.5 and δ 25.0, respectively). ROESY correlation of CH₃-16 to H-3 and H-7 showed that they were on the same face, β. ROESY correlations of 15-CH₃ to H-5 and H-6 indicated that the relative configurations at C-4, C-5, C-6, C-7, C-8, and C-10 were similar to those assigned for compound 1. An X-ray crystallography study of 3 (Figure 1) confirmed the structure and the relative configuration of compound 3.

The molecular formula of compound 4, C₁₆H₁₈O₆, determined by HRESIMS, was the same as that of 3. Comparison of ¹H NMR spectra of these two compounds indicated that the only difference was the position of the OH group in the A ring. In the COSY spectrum, the oxygenated methine at δ 3.96 showed strong cross-peaks to H-1 (δ 1.51 and 2.32) and H-3 (δ 1.86 and 2.05), indicating that the OH group was at on C-2. HMBC correlations were not observed between H-2 and adjacent carbons; however, correlations of H-1 to C-2, C-3, C-5, C-9, C-10, and C-16; and H-3 to C-1, C-2, C-4, C-5, C-14, and C-15 confirmed this structure. COSY and HMBC correlations for the rest of the molecule were identical to those observed for compound 1. Even though the 2-hydoxylated methine appeared as a broad multiplet, coupling constants of H-1 and H-3 clearly indicated that H-2 was axial. ROESY correlation of H-2 to H-5 and CH3-15 indicated that it was in α oriented. The rest of the ROESY correlations of 4 were similar to those observed for 1, suggesting that the relative configurations of the remaining asymmetric centers were the same as those of the latter.

Compound 5 was determined to have the same molecular formula (C₁₆H₁₈O₆), as those of compounds 3 and 4. Comparison of ¹H NMR and ¹³C NMR spectra of 3–5 indicated that 5 only differed in the position of the OH group on the A ring. The COSY spectrum displayed a spin system consisting of two adjacent methylene groups connected to a hydroxyl methine [δ 3.94 (t, J = 6.0 Hz)]. Since the 3-OH analog had been identified, the OH group had to be on C-1. HMBC correlations of H-1 to C-2, C-3, C-5, C-9, C-10, and C-16 confirmed its location. The other HMBC correlations were very similar to those observed for compound 1. The coupling constant of H-1 (t, J = 6.0 Hz) indicated that it was in equatorial configuration. An axial OH group on C-1 would lead to interaction between this group and the CH₃-16 group. This was evident from the up-field shift of C-16 (δ 17.8) compared to that of compounds 3 and 4 (δ 27.2 and 29.1, respectively). In the ROESY spectrum, correlations of H-5 to H-1, H-6, and CH₃-15 indicated that they were on the same face, α. All of these data suggested the structure and relative configuration depicted for compound 5.

The molecular formula of compound 6 was determined to be C₁₆H₁₆O₅; one oxygen less than that observed for compound 1. Comparison of the NMR spectra of compounds 6 and 1 indicated close similarities in the signals of the A ring. The differences of the remaining signals could be attributed to the replacement of the epoxy group in the B ring of the latter by a double bond in the former. The COSY spectrum showed, in addition to the ABX spin system due to H- 1, H-2 and H-3 as observed in compound 1, a second system due to H-5 (δ 2.29, d, J = 5.2 Hz), H-6 (δ 5.01, t, J = 5.0 Hz), H-7 (δ 6.24, m), H-11 (δ 6.00, d, J = 1.6 Hz), and H-13 [δ 4.91 (d, J = 13.6 Hz) δ 5.01 (dt, J = 13.6, 2.0 Hz)]. In the HMBC spectrum, the olefinic proton H-1 (δ 6.61) showed correlations to the carbons C-3, C-9, C-5, and C-10 and H-3 (δ 4.42) to C-1, C-2, C-4, C-5, C-14, and C-15. Additionally, H-6 showed correlations to C-5, C-7, C-8, C-10; H-7 to C-5, C-9 and C-13; H-11 to C-8, C-10, C-12, and C-16; H-13 to C-7, C-8, C-9, and C-12. The ROESY spectrum showed correlations of H-3 to H-5 and H-5 to H-6, indicating that all these protons were on the same face; however, due to the close proximity of ¹H signals of CH₃-15 and CH₃-16, the ROESY spectrum was of limited use to determine the relative configuration of all asymmetric centers of this compound. An X-ray crystallographic study of this compound (Figure 1) confirmed the structure and relative configuration of 6 as depicted.

The molecular formula of compound 7 was confirmed to be C₁₆H₁₆O₄ by HRESIMS. Comparison of the ¹H- and ¹³C NMR spectra showed that 7 was the 3-dehydroxy analog of compound 6 (see Tables 1 and 2). The COSY spectrum showed the presence of a -CH=CHCH₂- spin system very similar to that observed for compound 2. The HMBC spectrum correlations ofH-3 (δ 2.26, 2.83) to C-1, C-2, C-4, C-5, C-14, and C-15 confirmed that the methylene was at 3. The other HMBC correlations were similar to those observed for compound 6. The ROESY spectrum showed correlations between CH₃-15 and H-5, H-6 as well as to one of the H-3 (δ 2.26) protons, suggesting that they are on the same face. On the other hand, CH₃-16 showed a correlation to the other H-3 (δ 2.83) protons, indicating that they are on the opposite phase. The data indicated that the relative configurations of the asymmetric centers of compound 7 were the same as those observed for compound 6.

HRESIMS data suggested molecular formula C16H18O5 for compound 8. ¹H NMR and ¹³C NMR data of this compound and compound 6 were similar except for the signals in ring A. The differences were due to replacement of the double bond in the A ring in 6 by two methylene groups in 8. This information indicated that 8 was the 1,2-dihydro analog of 6. The COSY spectrum showed the presence of a -CH₂CH₂CHOH- spin system, very similar to that observed for compound 3. HMBC correlations of H-3 (δ 4.31) to C-1, C-2, C-14, and C-15, and H-1 (δ 1.66 and 1.83) to C-2, C-3, C-5, and C-16 further supported this structure. The other correlations were very similar to those observed for compound 6. Coupling constants of H-3 (dd, J = 11.2, 4.0 Hz) indicated that it was axial. As in the case of compound 3, the chemical shift of C-15 (δ, 15.9) in 8 had shifted up-field relative to that of 6 and 7 (δ 25.0, 25.4, respectively), probably due to its interaction with the equatorial 3-OH group. The ROESY spectrum showed correlations between CH₃-15 and H-5, H-6, suggesting that they were on the same face, α. The ROESY correlation of CH₃-16 to H-3 indicated that they were on the same face, β. The 3-epimer of 8 was synthesized previously.³⁰

Compound 9 had molecular formula, C₁₆H₁₈O₅ (HRESIMS), identical to that of 8. Comparison of ¹H NMR spectra showed that signals in the ring A of compounds 9 and 4 and signals in rings B and C of compounds 9 and 8 were superimposable. This suggested that compound 9 was the 2-OH isomer of 8. As in the case of compound 4, no HMBC correlations were observed between H-2 and adjacent carbons in compound 9. However, correlations of 2H-1 (δ 1.34, 2.20) to C-2, C-3, C-5, C-9, C10, and C-16, and 2H-3 (δ 1.68, 1.96) to C-1, C-2, C-4, C- 5, C-14, and C-15 confirmed this structure. The other HMBC correlations were identical to those observed for 8. On the basis of coupling constants, the 2-OH group was determined to be equatorial. Relative configuration of 9 was determined by ROESY.

Barrero et al.³⁰ previously revised the structure of wentilactone B¹⁹ to the 2-epimer of 9. Although the ¹H NMR data observed for 9 were different from those reported for wentilactone B by Dorner et al.,¹⁹ data reported for the same compound by Barrero et al.³⁰ had close resemblance if the assignments for H-1 and H-3 were interchanged. There is a possibility that wentilactone is the 2β-hydroxy analog and is not the 2α-hydroxy analog proposed by Barreiro et al. ³⁰ by comparison with the data reported for 2α-hydroxynagilactone³¹ In light of the current study, the conformation of the 2-OH group of both should be revised to β. The complete assignment of NMR data were not previously reported for wentilactone B¹⁹; thus, the data for 9 were included in Tables 1 and 2.

The molecular formula of compound 10 (C₁₆H₁₈O₅) was identical to those of ₈ and ₉. Comparison of ₁H NMR showed that signals in the A rings of compound 10 were superimposable with those of compound 5, and signals of the B and C rings of 10 exhibited strong similarities to those of compounds 8 and 9 except for a down-field shift of H-11, possibly due to interaction of H-11 with the 1β-hydroxy group. This indicated that 10 was the 1β-hydroxy isomer of 8. A similar shift of the H-11 signal was also observed for compound 5, relative to those of 3 and 4. HMBC correlations of H-1 (δ 3.78) to C-3, C-9, C-10, and C16 and H-3(δ 1.38, 1.13) to C-1, C-2, C-4, C-5, C-14, and C-15 confirmed the structure. The coupling constant of 1-H (t, J = 6.4 Hz) showed that it was equatorial. An axial OH group on C-1 resulted in an up-field shift of the CH₃-16 signal relative to those in 9 and 8, similar to that observed for compound 5 relative to 3 and 4. In the ROESY spectrum, correlations of CH₃-15 to H-5 and H-6, and H-5 to H-6 and H-1 suggested that they were on the same face, α. An X-ray crystallographic analysis of 10 (Figure 1) confirmed the structure and the relative configuration as shown.

Compounds 11,^20,32 12,¹⁵ and 13,¹⁸ were identified by comparing physical and spectral data with those reported in the literature.

The absolute configuration of the compounds was determined by comparing CD curves with those previously reported for this class of compounds.^17,19,25 Compounds 1–5 showed negative Cotton effects at 265–270 nm in CD curves indicating that the absolute configuration at common asymmetric centers was 4S, 5R, 6S, 7R, 8R, 10S as depicted. Similarly, compounds 6– 11 showed negative Cotton effects at 290–296 nm in CD curves indicating that the absolute configuration at common asymmetric centers was 4S, 5R, 6S, 10S as depicted.

The in vitro antiplasmodial activities of compounds 1–13 are summarized in Table 3. Compounds 6 (IC₅₀ 43 and 32 ng/mL), 9 (IC₅₀ 97 and 84 ng/mL) and 7 (IC₅₀ 170 and 130 ng/mL) demonstrated strong antiplasmodial activities whereas compounds 3, 5, 8 and 13 were inactive. Selectivity index of antiplasmodial activity was calculated based on their cytotoxicity to mammalian kidney fibroblasts (Vero cells) determined in parallel experiments. All of the compounds with strong antiplasmodial activities also were cytotoxic to Vero cells, and their low selectivity indices (<30) preclude them as antimalarial drug candidates. The cytotoxic potential of these compounds was also evaluated against a panel of solid tumor cell lines and kidney epithelial cells (Table 4). The order of cytotoxic activity was similar in all cell lines for most ofthe compounds, and was parallel to their antiplasmodial activity. These observations suggested that the antiplasmodial activity of these compounds was a manifestation of their general cytotoxicity. The presence of a lactone in ring C appeared to be essential for the activity of these compounds.

Table 3.

Antiplasmodial Activity of Compounds 1–13

Compound	D6-clone		W2-clone		cytotoxicity
	IC50 ng/mL	S.I.	IC50 ng/mL	S. I.	IC50 ng/mL
1	750	2.0	600	2.5	1500
2	2300	>2.1	2300	>2.1	NC
3	NA	-	NA	-	NC
4	740	4.5	820	4.0	3300
5	NA	-	NA	-	NC
6	43	22.1	32	29.7	950
7	170	17.6	130	23.1	3000
8	NA	-	NA	-	NC
9	97	10.3	84	11.9	1000
10	900	5.3	670	7.1	4760
11	300	12.3	200	18.5	3700
12	250	6.8	190	8.9	1700
13	NA	-	NA	-	NC
chloroquinea	10		100		NC
artemisinina	4.7		3.8		NC

^aPositive controls

NC not cytotoxic at the highest dose (4760 ng/mL) tested

NA not active at the highest dose (4760 ng/mL) tested

S. I. (selectivity index) = IC50 for antiplasmodial activity/IC50 for cytotoxicity

Table 4.

Cytotoxic Activity [IC₅₀ ( µg/mL)] of Compounds 1–13

	SK-MEL	KB	BT-549	SK-OV-3	LLC-PK11
1	0.58	4.4	4.8	8	2.8
2	2.8	NC	>10.0	NC	>10.0
3	NC	NC	NC	NC	NC
4	1.1	6.8	10.0	10.0	4.2
5	10	NC	NC	NC	NC
6	0.12	0.6	1.2	1.1	1.1
7	2.3	4.3	9.6	6.0	10.0
8	NC	NC	NC	NC	NC
9	0.41	2.1	4.7	3.2	1.2
10	3.3	>10.0	>10.0	>10.0	>10.0
11	1.1	4.2	>10.0	>10.0	5.4
12	0.72	4.5	9.0	6.3	4.8
13	NC	NC	NC	NC	NC
Doxorubicin a	0.6	0.9	1.3	0.6	0.6

^aPositive control

NC = not cytotoxic up to highest test concentration of 10 µg/mL

Compounds 1, 2, 4–7, 9–11 showed total growth inhibition of a dicot (lettuce: Lactuca sativa cv. L. Iceberg) and a monocot (bentgrass: Agrostis stolonifera) at concentrations of 1 mg/mL and 0.1 mg/mL and partial inhibition at 0.001 mg/mL (Table 6). Compounds 3 and 8 caused total inhibition of growth only at the highest concentration (1 mg/mL). Compound 12 showed moderate activity at the highest concentration, and compound 13 was inactive. It appears that the lactone in ring C is essential for the phytotoxicity. However, this activity did not strongly correlate with the cytotoxic activity.

More detailed evaluation of the herbicidal activity of compounds 1, 2, 4–7, 9–11 was made in a duckweed (Lemna pausicostata) bioassay that evaluates growth after exposure to different concentrations of the compounds (Figure 2). The most active compounds (1, 2, 6, 7, 11) were more active than more than half of 26 commercial herbicides using exactly the same bioassay under identical conditions in the same laboratory.³³ The I₅₀ value for 1 (0.39 µM) was lower than 19 of the commercial herbicides previously tested with this assay. Macías et al.²² and Barrero et al.²³ reported 11 to be a potent phytotoxin on a number of plant species. Dorner et al.¹⁹ found 9 to be a weak inhibitor of wheat coleoptile elongation. However, this assay is not valid for most herbicides. Using a duckweed assay similar to ours, John et al.¹⁵ reported two compounds, structurally related to ours, to be “strongly herbicidal.” A Japanese patent¹³ exists on herbicidal activity of this class of compounds. Compounds 1 and 3 have been included in its claims for as herbicides although the latter has not been identified.¹³

Figure 2.

Dose/response effect of 1 on growth of duckweed after seven days of exposure. Inset provides I₅₀ values for 1 and the other most active compounds (Table 3) generated from dose/response data after seven days of exposure.

Experimental Section

General Experimental Procedures

Melting points were measured in a Uni-melt, Thomas Hoover capillary melting point apparatus. NMR spectra were recorded on a Bruker 400 MHz spectrometer using CDCl₃/CD₃OD (1:1) as the solvent unless otherwise stated. MS analyses were performed on an Agilent Series 1100 SL equipped with an ESI source (Agilent Technologies, Palo Alto, CA, USA). All acquisitions were performed under positive ionization mode. Column chromatography was carried out on Merck silica gel 60 (230–400 mesh). Preparative TLC was carried out using silica gel GF plates (20 × 20 cm, thickness 0.25 mm).

Isolation of Sclerotinia homoeocarpa

Sclerotinia homoeocarpa F.T. Bennett was isolated from an infected grass blade collected near the Sardis Lake dam site, Sardis, Mississippi and identified by Dr. Maria Tomaso-Peterson, of the Department of Entomology and Plant Pathology, Mississippi State University, Mississippi State. A voucher specimen (DN5-59-1) was deposited in the fungal collection at the National Center for Natural Products Research, University of Mississippi.

Extraction and Purification

S. homoeocarpa cultures, in 40 conical flasks (1 L) containing 600 mL of potato/dextrose liquid medium, were incubated for 21 days at 27 °C on an orbital shaker at 125 rpm. The medium was separated by filtration and extracted three times with ethyl acetate. The EtOAc extract was evaporated to dryness, and the crude extract (3.31 g) was thoroughly washed with hexanes. The hexanes insoluble dark brown gum (2.24 g) was chromatographed on silica gel. The column was eluted sequentially with hexanes/CH₂Cl₂ and CH₂Cl₂/MeOH with increasing concentrations of MeOH, collecting 500 mL volumes to give eight major fractions, F-1 (hexanes:CH₂Cl₂, 4:1, 170 mg), F-2 (hexanes:CH₂Cl₂, 1:1, 147 mg), F-Page3 (hexanes:CH₂Cl₂, 3:4, 370 mg), F-4 (CH₂C_l2, 247 mg), F-5 (CH₂Cl₂:MeOH, 99:1, 285 mg), F-6 (CH₂C_l2:MeOH, 98:2, 214 mg), F-7 (CH₂Cl₂:MeOH, 95:5, 154 mg) and F-8 (CH₂Cl₂:MeOH, 90:10, 228 mg). Fractions, F-3, F-4 and F-5 showed significant antiplasmodial activities. Separation of F-3 by PTLC (CH₂Cl₂:MeOH, 97.5:2.5) afforded compounds 1 (123 mg) and 6 (224 mg) and a mixture, which was further separated by PTLC (hexanes:EtOAc, 7:3) to afford 4 (23 mg) and 9 (15 mg). Similarly, separation of F-4 by PTLC (hexanes:EtOAc, 60:40 × 2) gave compounds 2 (34 mg), 7 (37 mg), 11 (39 mg), and impure 3 (5 mg) which was further purified by PTLC (CH₂Cl₂:MeOH, 98:2). Separation of F-5 by PTLC (hexanes:EtOAc, 60:40 × 2) afforded compounds 5 (67 mg) and 8 (12 mg) and a mixture, which was further separated (PTLC, CH₂Cl₂:MeOH, 95:5) to give 10 (10 mg) and 12 (82 mg). Re-crystallization of the white solid precipitated from F-8 (CH₂Cl₂ / MeOH) afforded compound 13 (87 mg).

Compound 1

colorless needles (CH₂C_l2/hexane); mp 258–259 °C; [α] ²⁶ _D +2.1 (c 0.67, CHCl₃:CH₃OH 9:1); UV (CH₃OH) λ_max (log ε) 218 (4.12); CD (CH₃OH) [θ]₂₆₇ -21.6; ¹H NMR and ¹³C NMR data (see Tables 1 and 2); HRESIMS [M + H]⁺ m/z 305.1041 (calcd for C¹⁶H¹⁷O⁶, 305.1025).

Compound 2

colorless needles (CH₂Cl₂/hexane); mp 160–162 °C; [α]²⁶ _D -54.0 (c 0.43, CHCl₃:CH₃OH 9:1); UV (CH₃OH) λ_max (log ε) 216 (4.08); CD (CH₃OH) [θ]265 -20.6; ¹H NMR (CDCl₃) and ¹³C NMR data (see Tables 1 and 2); HRESIMS [M + H]⁺ m/z 289.1094 (calcd for C₁₆H₁₇O₅, 289.1076)

Compound 3

colorless needles (EtOAc/hexane); mp 260–261 °C; [α]²⁶ _D +34.3 (c 0.30, CHCl₃); UV (CH₃OH) λ_max (log ε) 215 (4.10); CD (CH₃OH) [θ]₂₆₅ -21.8; ¹H NMR and ¹³C NMR data (see Tables 1 and 2); HRESIMS [M + H]⁺ m/z 307.1207 (calcd for C₁₆H₁₉O₆, 307.1182).

Compound 4

colorless needles (MeOH/CH₂Cl₂); mp 248–250 °C; [α]²⁶ _D +30.6 (c 0.63, CHCl₃:CH₃OH 9:1); UV (CH₃OH) λ_max (log ε) 219 (4.2); CD (CH₃OH) [θ]₂₇₀ -19.4; ¹H NMR and ¹³C NMR data (see Tables 1 and 2); HRESIMS [M + H]⁺ m/z 307.1193 (calcd for C₁₆H₁₉O₆, 307.1182).

Compound 5

colorless needles (EtOAc/hexane); mp 240–241 °C; [α]²⁶ _D -46.2 (c 0.28, CHCl₃); UV (CH₃OH) λ_max (log ε) 215 (4.33); CD (CH₃OH) [θ]265 -35.0; ¹H NMR and ¹³C NMR data (see Tables 1 and 2); HRESIMS [M + H]⁺ m/z 307.1176 (calcd for C₁₆H₁₉O₆, 307.1182).

Compound 6

colorless prisms (MeOH); mp. 252–253 °C; [α] ²⁶ _D +262.2 (c 0.57, CHCl₃:CH₃OH 9:1); UV (CH₃OH) λ_max (log ε) 260 (4.26); CD (CH₃OH) [θ]₂₉₅ -5.0; ¹H NMR and ¹³C NMR data (see Tables 1 and 2); HRESIMS [M + H]⁺ m/z 289.1095 (calcd for C₁₆H₁₇O₅, 289.1076).

Compound 7

colorless needles (CH₂Cl₂/hexane); mp 160–162 °C; [α] ²⁶ _D -162.0 (c 0.52, CHCl₃:CH₃OH 9:1); UV (CH₃OH) λ_max (log ε) 260 (4.18); CD (CH₃OH) [θ]₂₉₂ -12.3; ¹H NMR (CDCl₃) and ¹³C NMR data (see Tables 1 and 2); HRESIMS [M + H]⁺ m/z 273.1149 (calcd for C₁₆H₁₇O₄, 273.1127).

Compound 8

colorless needles (CH₂Cl₂/hexane); mp 244–245 °C; [α] ²⁶ _D -180.1 (c 0.63, CHCl₃:CH₃OH 9:1); UV (CH₃OH) λ_max (log ε) 260 (4.40); CD (CH₃OH) [θ]₂₉₀ -24.7; ¹H NMR and ¹³C NMR data (see Tables 1 and 2); HRESIMS [M + H]⁺ m/z 291.1249 (calcd for C₁₆H₁₉O₅, 291.1232).

Compound 9

colorless needles (CHCl₃/hexane); mp 254–256 °C; [α] ²⁶ _D -236.5 (c 0.58, CHCl₃:CH₃OH 9:1); UV (CH₃OH) λ_max (log ε) 260 (4.3); CD (CH₃OH) [θ]₂₉₅ -3.9; ¹H NMR and ¹³C NMR data (see Tables 1 and 2); HRESIMS [M + H]⁺ m/z 291.1251 (calcd for C₁₆H₁₉O₅, 291.1232).

Compound 10

colorless needles (CHCl₃/hexane); mp 235–236 °C; [α] ²⁶ _D -56.3 (c 0.33, CHCl₃); UV (CH₃OH) λ_max (log ε) 260 (4.21); CD (CH₃OH) [θ]₂₈₈ -26.1; ¹H NMR and ¹³C NMR data (see Tables 1 and 2); HRESIMS [M + H]⁺ m/z 291.1249 (calcd for C₁₆H₁₉O₅, 291.1232).

Compound 11

colorless needles (CHCl₃/hexane); mp 194–195 °C (lit 191–193 °C);³² [α]²⁶ _D -272.1 (c 0.53, CHCl₃:CH₃OH 9:1) (lit. −332.7);³² UV (CH₃OH) λ_max (log ε) 260 (4.21); CD (CH₃OH) [θ]₂₉₆ -5.1.

Compound 12

colorless needles (CHCl₃/hexane); mp 240−241 °C (lit. 237 °C);¹⁵ [α]²⁶ _D - 33.4 (c 0.36, CHCl₃:CH₃OH 9:1) (lit. −38.5).¹⁵

Compound 13

colorless needles (CHCl₃/hexane); mp 214–216 °C (lit 200 °C);¹⁸ [α]²⁶ _D +27.8 (c 0.36, CHCl₃:CH₃OH 9:1) (lit. +22.5).¹⁸

Crystallographic Data for Compounds 1, 3, 6, and 10

All single crystal X-Ray diffraction data were collected on a Bruker Smart Apex II system, using Cu Kα radiation with a graphite monochromator, fine-focus sealed tube. The crystal was kept at 100 K under a stream of cooled nitrogen gas from a KRYO-FLEX low temperature device. Data collection, indexing and initial cell refinements were all carried out using APEX II software. Frame integration and final cell refinements were done using SAINT software. Structure solution, refinement, graphics and generation of publication materials were performed using SHELXTL, V6.12 software. Hydrogen atoms were placed their expected chemical positions using the HFIX command and were included in the final cycles of least squares with isotropic Uij‘s related to the atom’s ridden upon.

Compound 1

C₁₆H₁₆O₆, MW = 304.29, orthorhombic space group P2₁2₁2₁, a = 6.5821 (4) Å, b = 11.0090 (7) Å, c = 18.2832 (12) Å, α = β = γ = 90°, V = 1324.84 (15) ų, Z=3, crystal dimensions 0.23 × 0.18 × 0.14 mm. The final cell parameters were determined from least-squares refinement on 6799 reflections, with Rint = 0.058, and wR(F ²) = 0.076. The supplementary crystallographic data can be obtained free of charge from The Cambridge Crystallographic Data Centre, reference number CCDC 681270, via www.ccdc.cam.ac.uk/data_request/cif.

Compound 3

C₁₆H₁₈O₆, MW = 306.30, orthorhombic space group P2₁2₁2₁, a = 7.7869 (2) Å, b = 9.1234 (3) Å, c = 19.6990 (6) Å, α = β = γ = 90°, V = 1399.48 (7) ų, Z=4, crystal dimensions 0.26 × 0.23 × 0.11 mm. The final cell parameters were determined from least-squares refinement on 5987 reflections, with Rint = 0.031, and wR(F ²) = 0.069. The supplementary crystallographic data can be obtained free of charge from The Cambridge Crystallographic Data Centre, reference number CCDC 681271, via www.ccdc.cam.ac.uk/data_request/cif.

Compound 6

C₁₆H₁₅O₅, MW = 288.30, or space group P2₁2₁2₁, a = 6.29910 (10) Å, b = 6.55110 (10) Å, c = 32.5925 (6) Å, α = β = γ = 90°, V = 1344.96 (4) ų, Z=4, crystal dimensions 0.28 × 0.22 × 0.14 mm. The final cell parameters were determined from least-squares refinement on 7789 reflections, with Rint =0.022, and wR(F ²) =0.070. The supplementary crystallographic data can be obtained free of charge from The Cambridge Crystallographic Data Centre, reference number CCDC 681272, via www.ccdc.cam.ac.uk/data_request/cif.

Compound 10

C₁₆H₁₈O₅, MW = 290.30, monoclinic space group P2₁, a = 6.1595 (2) Å, b = 14.2378 (5) Å, c = 8.3456 (3) Å, α = γ = 90°, β = 107.479 (2)°, V = 698.10 (4) ų, Z=2, crystal, dimensions 0.21 × 0.14 × 0.08 mm. The final cell parameters were determined from least-squares refinement on 3166 reflections, with Rint =0.017, and wR(F ²) =0.071. The supplementary crystallographic data can be obtained free of charge from The Cambridge Crystallographic Data Centre, reference number CCDC 681273, via www.ccdc.cam.ac.uk/data_request/cif.

Phytotoxicity Assays

Initial phytotoxicity bioassays were carried out according to the procedure previously described by Dayan et al.,³⁴ using bentgrass (Agrostis stolonifera) and lettuce (Lactuca sativa cv. L. Iceberg), in 24-well plates. Test compounds (1 mg each) were dissolved in 100 µL of 5% CH₃OH in CHCl₃ and a 20 µL aliquot of each solution was pipetted onto the filter paper and dried for 30 minutes by airflow in a sterile biohazard hood. Water (200 µl) was added after placing the dried and sample impregnated filter paper in the well. The solvent controls were treated identically, using the solvent described above. Phytotoxicity was ranked visually. The ranking of phytotoxic activity was based on a scale of 0 to 5; 0 no effect; 5 no growth.

Further bioassays were conducted with the most active compounds from the initial bioassay using duckweed (Lemna pausicostata).³³ Plants were incubated with growth medium (pH 6.4) in a conical flask. Plants were grown autotrophically for a week in a growth chamber in continuous illumination (105 µmol m⁻²s⁻¹) at 25±2°C. Fifty colonies of duckweed from the incubation colony were placed in 6-cm diam. polystyrene Petri dishes with 5 mL of nutrient medium containing test compound and 0.33 % stock solution solvent (5% v/v CH₃OH in CH₂Cl₂). This level of stock solution solvent had no effect on duckweed growth. Treatments for each dose were triplicated. Data were analyzed to determine I₅₀ values using R software.³⁵

Assay for in vitro Antiplasmodial (Potential Antimalarial) Activity

In vitro antiplasmodial activity was determined against D6 (chloroquin sensitive) and W2 (chloroquin resistant) strains of P. falciparum as described earlier.³⁶ A 200 µL suspension of red blood cells, infected with P. falciparum (2% parasitemia and 2% hematocrit), in RPMI 1640 mediumsupplemented with 10% human serum and 60 µg/mL Amikacin was added to the wells of a 96- well plate containing 10 µL of serially diluted samples. The plate was flushed with a gas mixture of 90% N₂, 5% O₂, and 5% CO₂ and incubated at 37 °C for 72 h in a modular incubation chamber (Billups-Rothenberg, CA). Parasitic LDH activity was determined by mixing 20 µL of the incubation mixture with 100 µL of Malstat™ reagent (Flow Inc., Portland, OR) and incubating at room temperature for 30 min. Then 20 µL of a 1:1 mixture of NBT/PES (Sigma, St. Louis, MO) was added, and the plate further incubated in the dark for 1 h. The reaction was then stopped by adding 100 µL of acetic acid (5%). Plates were read at 650 nm. Artemisinin and chloroquine were included in each assay as the drug controls, and IC₅₀ values were computed from the dose response curves.

Assay for Cytotoxicity to Mammalian Cells

In vitro cytotoxicity was determined against a panel of mammalian cells that included kidney fibroblast (Vero), kidney epithelial (LLC-PK₁₁), malignant melanoma (SK-MEL), oral epidermal carcinoma (KB), breast ductal carcinoma (BT-549) and ovary carcinoma (SK-OV-3) cells.³⁷ The assays were performed in 96- well tissue culture-treated plates. Cells were seeded to the wells of 96-well plate at a density of 25,000 cells/well and incubated for 24 h. Samples at different concentrations were added and plates were again incubated for 48 h. The number of viable cells was determined by using Neutral Red dye, and IC₅₀ values were obtained from dose response curves. Doxorubicin was used as a positive control.

Table 5.

Phytotoxic Activity of Compounds 1–13

Compound				Concentration (mg/mL)
	1.0		0.1		0.01		0.001
	L	B	L	B	L	B	L	B
1	5	5	5	5	5	4	2	2
2	5	5	5	5	3	4	0	1
3	5	5	4	4	0	0	0	0
4	5	5	5	5	3	2	0	0
5	5	5	5	5	4	4	1	0
6	5	5	5	5	5	4	0	0
7	5	5	5	5	1	3	0	0
8	5	5	2	3	0	0	0	0
9	5	5	5	5	2	3	1	2
10	5	5	5	5	1	1	0	0
11	5	5	5	5	2	3	0	0
12	5	5	4	4	2	1	0	0
13	1	0	0	0	0	0	0	0

L = lettuce; B = bentgrass

Ranking based on scale of 0 to 5

0 = no effect

5 = no growth