19-nor-Pimaranes from Icacina trichantha

Abstract

Four new unusual 19-nor-pimarane-type diterpenes were isolated from the tuber of Icacina trichantha (Icacinaceae, Oliv.). The structures were elucidated based on spectroscopic and HRMS analysis. The absolute configurations were determined by electronic circular dichroism. All four compounds are structural analogues of icacinol and humirianthol, but do not demonstrate the same cytotoxic activity. A plausible biogenetic pathway is proposed.

Graphical Abstract

Introduction

Icacina trichantha (Icacinaceae, Oliv.) is a flowering shrub native to forested vegetation areas of southern Nigeria [1]. Its large tuber is well known among the natives and traditional medicine practitioners for its nutritional and medicinal applications. In rural areas, tinctures of the tuber are often kept in households as a readily available remedy for treating common ailments and wounds such as fever and snake bite [2]. One of its vernacular names, ‘unumbia’ (meaning ‘induce emesis’), conveys its primary physiological action [3]. The plant is also reportedly used to treat food poisoning and constipation [4], in addition to chronic conditions such as hypertension [5] and “soft tumors” [3].

Following reports that demonstrated the emetic and other bioactivities of this plant [4,6,7], our group has conducted a series of phytochemical studies, leading to the isolation and identification of a number of pimarane diterpenoids, in particular, the uncommon 9βH type [8–12]. This interesting class of compounds has demonstrated cytotoxic [13], antibacterial [14], and antifungal [15] activities. In addition, we have reported the potential herbicidal activity of one of its ingredients, icacinol, which demonstrated selective germination inhibitory activity in Arabidopsis [16].

In a continuing effort to explore ethnobotanicals for new and bioactive substances, we have further studied the extract of I. trichantha tuber. In this report, four new 19-nor-pimaranes are described (Fig. 1).

Figure 1.

Chemical structures of icatrichanone (1) and structural analogues (2 - 4)

Results and Discussion

The EtOAc (ethyl acetate)-soluble fraction of an acetone extract of the tubers of I. trichantha was subjected to repeated column chromatography over Si gel, MCI gel, and reverse phase HPLC to afford four new compounds (1 – 4). Their planar structures were elucidated on the basis of HRESIMS (high resolution electrospray ionization-MS) and NMR data, while the absolute configurations were determined by electronic circular dichroism (ECD) analysis.

Compound 1 was obtained as a colorless amorphous powder. A molecular formula of C₁₉H₂₆O₅ was suggested based on the protonated molecular ion at m/z 335.1837 [M+H]⁺ (calcd for C₁₉H₂₇O₅⁺) and ¹³C NMR data. The ¹³C NMR spectrum (Table 1) exhibited nineteen carbon signals, which could be assigned with the aid of the DEPT NMR experiment to be two methyls, six methylenes, six methines, one carbonyl, one dioxygenated secondary carbon, and three quaternary carbons. The ¹H NMR spectrum (Table 1) displayed an isolated olefinic proton at δ_H 5.97 (H-7, s) and two methyls at δ_H 0.99 (H-17, s) and 1.09 (H-18, d, J = 6.8 Hz), respectively. All proton signals were assignable to their corresponding carbon by an HSQC (heteronuclear single quantum correlation) experiment. Initial comparison of the NMR chemical shifts of 1 with those of humirianthol [17] revealed close similarities. Indeed, careful interpretation of the NMR data of 1 indicated the presence of the same tetracyclic structure along with a 3,20-epoxy bridge as in humirianthol. The noticeable differences were the absence of C-19 lactone carbon, the downfield shift of C-6 from δ_C 71.1 (CH) to 200.5 (C), as well as downfield shifts of the alkene group from δ_C 116.4 and 145.6 to δ_C 124.8 (C-7) and 163.2 (C-8), respectively. It became obvious that both C-19 and the γ-lactone commonly found in other Icacina pimaranes were missing in this compound. Instead, the HMBC cross-peak (Fig. 2) between H-5 (δ_H 2.27) and C-6 (δc 200.5) suggested the presence of a ketone functional group at C-6.

Table 1.

¹H (400 MHz) and ¹³C (100 MHz) NMR Spectroscopic Data of Compounds 1 and 2 (δ in ppm)

	1a			2a			Humiriantholb
position	δC, type	δH (J in Hz)	HMBCc	δC, type	δH (J in Hz)	HMBCc	δC, type
1	30.3, CH2	1.84, md 1.70, m	2	30.5, CH2	1.85, md 1.71, dd (11.7, 1.6)		28.3, CH2
2	27.9, CH2	2.03, tdd (15.7, 7.5, 4.8) 1.84, md	1	27.9, CH2	2.03, dt (12.5, 1.9) 1.83, md		28.0, CH2
3	98.7, C	--		98.7, C	--		96.2, C
4	37.4, CH	2.67, m		37.3, CH	2.69, m		49.8, C
5	53.7, CH	2.27, dd (4.0, 2.0)	6, 10	53.2, CH	2.26, dd (4.0, 2.1)	6, 20	43.8, CH
6	200.5, C	--		201.0, C	--		71.1, CH
7	124.8, CH	5.97, s	5, 8, 9, 14	124.0, CH	6.22, s	8, 9, 14	116.4, CH
8	163.2, C	--		162.9, C	--		145.6, C
9	39.4, CH	2.18, dd (12.6, 3.7)	7, 8, 10	41.3, CH	2.13, dd (12.7, 3.7)	7, 10	36.2, CH
10	36.7, C	--		37.0, C	--		29.8, C
11	26.5, CH2	1.86, md 1.63, m		26.4, CH2	1.86, md 1.66, dt (12.4, 3.6)		24.9, CH2
12	32.9, CH2	1.48, m	13, 15	34.9, CH2	1.53, dt (13.7, 3.3) 1.41, ddd (13.6, 12.6, 4.3)		31.9, CH2
13	52.5, C	--		55.1, C	--		48.1, C
14	87.3, CH	4.02, s	7, 9, 12, 15, 17	106.4, C	--		85.3, CH
15	79.8, CH	3.85, dd (10.8, 4.7)		79.6, CH	3.84, dd (5.4, 1.5)	12, 14	77.4, CH
16	76.8, CH2	4.43, dd (10.1, 4.9) 3.73, dd (10.3, 1.1)	14	76.1, CH2	4.47, dd (10.0, 5.4) 3.95, dd (10.0, 1.5)	13	75.1, CH2
17	15.2, CH3	0.99, s	12, 13, 14	13.7, CH3	1.05, s	12, 13, 14, 15	15.0, CH3
18	18.8, CH3	1.09, d (7.2)	3, 4, 5	18.8, CH3	1.11, d (7.2)	3, 4, 5	18.3, CH3
19	--	--		--	--		178.2, C
20	72.2, CH2	3.80, dd (9.0, 2.9) 3.63, dd (9.0, 2.0)	1, 3 10	72.2, CH2	3.77, dd (9.2, 2.9) 3.62, dd (9.0, 2.1)	1, 3, 9 5	71.0, CH2

^aData measured in methanol-d₄.

^bData measured in DMSO-d₆.

^cHMBC correlations are from proton(s) stated to the indicated carbon.

^dsignal partially obscured.

Figure 2.

¹H-¹H COSY and key HMBC correlations of 1 - 4

The relative configuration of 1 was determined by interpretation of NOESY results (Fig. 3). Key correlations observed include H-9/H-20, H-5/H-18, and H-7/H-14/H-17. Since all previously reported pimaranes from I. trichantha whose C-9 is a methine share the same β-orientation for its proton, the C-20/C-10 bond was assumed to be in β-orientation while H-5 and H-18 are α-oriented. On the other hand, the correlation between H-17 and H-14 provided convincing evidence that both H-14 and H-17 were also α-oriented.

Figure 3.

Key NOESY correlations of 1 - 4

The absolute configuration of 1 was then determined on the basis of ECD calculations. Using the TDDFT (time-dependent density functional theory) method, the calculated ECD spectrum was generated along with its enantiomer. The experimental ECD spectrum of 1 displayed similar Cotton effect as in the calculated ECD for (3S,4S,5S,9S,10S,13S,14S,15S)-1 (Fig. 4). All in all, the structure of 1, given the trivial name icatrichanone, is determined to be 3,20:14β,16-diepoxy-3α,15α-dihydroxy-6-oxo-19-nor-pimar-7-ene. This is the first reported 19-nor-pimarane from this species.

Figure 4.

Experimental ECD spectra for 1 - 4 and the calculated ECD spectra for 1 - 4 and their enantiomers

Compound 2 was isolated as a colorless amorphous powder. The HRESIMS result (m/z 351.1817 [M+H]⁺; calculated for C₁₉H₂₇O₆⁺) suggested a molecular formula of C₁₉H₂₆O₆ with seven indices of hydrogen deficiency. The ¹³C NMR and DEPT spectra displayed nineteen carbons corresponding to two methyls, six methylenes, five methines, a carbonyl, two dioxygenated secondary carbons, and three quaternary carbons. All proton signals were then assigned to their respective carbons (Table 1). Upon comparison with the NMR data of 1, it became clear that 2 possesses the same skeleton. The only carbon of concern was C-14 (δ_C 106.4), which could be confirmed by HMBC correlations with H-7 (δ_H 6.22), H-15 (δ_H 3.84), and H-17 (δ_H 1.05) (Fig. 2). In comparison with 1, C-14 was deshielded by 19.1 ppm, consistent with a hemiketal carbon. This led to the assignment of a 14-OH group, by which the molecular formula could be fulfilled. The structure of 2 was thus similar to that of icacinol [18]. The NOESY results of 2 (Fig. 3) were consistent with those of 1 and icacinol, suggesting the same stereostructure. Finally, the experimental ECD spectrum for 2 closely matched the calculated spectrum for (3S,4S,5S,9S,10S,13S,14S,15S)-2 (Fig. 4). Therefore 2 was determined to be 14-hydroxyicatrichanone.

Compound 3, obtained as a white amorphous powder, was assigned a molecular formula of C₂₀H₂₈O₅ with seven indices of hydrogen deficiency by HRESIMS (m/z 349.2014 [M+H]⁺; calculated for C₂₀H₂₉O₅⁺) and ¹³C and DEPT NMR data. The DEPT NMR spectrum (Table 2) displayed twenty carbons corresponding to two methyls, a methoxy, six methylenes, six methines, a carbonyl, a dioxygenated secondary carbon, and three quaternary carbons. Examination of the ¹H and ¹³C NMR spectra suggested a strong similarity with those of 1, except for an additional methoxy group, which could be assigned to C-3 by an HMBC experiment (Fig. 2). Again, the relative configuration could be determined from the NOESY results (Fig. 3). The observed Cotton effect in the ECD spectrum of 3 resembled that of (3S,4S,5S,9S,10S,13S,14S,15S)-3 (Fig. 4), and it was given a trivial name 3-O-methylicatrichanone.

Table 2.

¹H (400 MHz) and ¹³C (100 MHz) NMR Spectroscopic Data of Compounds 3 and 4 (δ in ppm)

	3a			4b
position	δC, type	δH (J in Hz)	HMBCc	δC, type	δH (J in Hz)	HMBCc
1	30.1, CH2	1.83, m		29.7, CH2	1.83, m
		1.72, dd (11.2, 2.2)			1.65, dd (9.4, 4.7)	5, 10
2	25.4, CH2	2.00, m	3	25.6, CH2	2.07–2.12, m
		1.93, dd (10.7, 2.7)			1.91 −1.98, m
3	101.9, C	-		100.2, C	--
4	34.5, CH	2.80, qdd (7.2, 3.5, 2.3)		32.1, CH	2.97, m
5	53.5, CH	2.26, dd (3.8, 2.0)	6, 10	52.3, CH	2.10, md	6
6	200.4, C	-		198.3, C	--
7	124.9, CH	5.98, s	5, 14	123.6, CH	6.29, s	5, 8, 9, 14
8	163.2, C	-		159.1, C	--
9	39.3, CH	2.16, dd (12.2, 3.7)	5, 7, 8	40.0, CH	2.10, md
10	36.7, C	-		35.9, C	--
11	26.5, CH2	1.87, m		25.6, CH2	2.07–2.12, m
		1.65, tdd (12.4, 9.5, 2.7)			1.91 −1.98, m
12	32.9, CH2	1.48, m		33.7, CH2	1.55 −1.62, md
		1.45, m			1.32, ddd (13.7, 13.1, 4.1)
13	52.5, C	-		54.1, C	--
14	87.3, CH	4.02, s	7	106.0, C	--
15	79.8, CH	3.84, dd (4.8, 1.3)		79.1, CH	3.89, dd (9.0, 5.8)	13
16	76.8, CH2	4.44, dd (10.0, 4.8)	14	76.0, CH2	4.50, dd (10.6, 5.3)	14
		3.73, dd (10.0, 1.3)			4.07, d (10.5)
17	15.2, CH3	0.99, s	12, 13, 14, 15	13.2, CH3	1.06, sd	12, 13, 14, 15
18	18.4, CH3	1.03, d (7.2)	3, 5	18.5, CH3	1.05, br d (7.1)d	3, 4, 5
19	--	--		--	--
20	72.1, CH2	3.80, dd (9.1, 3.2)	1, 3, 5, 10	71.4, CH2	3.86, dd (9.4, 3.5)	3
		3.62, dd (9.1, 2.1)			3.61, dd (9.3, 2.2)	5
OCH3	49.6, CH3	3.26, s	3	49.4, CH3	3.31, s	3
14-OH					3.10, s	8, 13
15-OH					2.79, d (9.3)

^aData measured in methanol-d₄

^bData measured in chloroform-d

^cHMBC correlations are from proton(s) stated to the indicated carbon.

^dSignal partially obscured.

Compound 4 was isolated as a colorless amorphous powder. A molecular formula of C₂₀H₂₈O₆ with seven indices of hydrogen deficiency was suggested by HRESIMS (m/z 365.1982 [M+H]⁺; calculated for C₂₀H₂₉O₆⁺). The ¹³C NMR spectrum (Table 2) displayed twenty carbons corresponding to two methyls, a methoxy, six methylenes, six methines, a carbonyl, a dioxygenated secondary carbon, and three quaternary carbons. Careful examination of the ¹H and ¹³C NMR data (Table 2 and Fig. 2) clearly showed that it is the 14-hydroxy derivative of 3. Its relative configuration was determined from the NOESY results (Fig. 3). In particular, the NOESY spectrum, being run in chloroform-d, allowed visualization of correlations of 14-OH (δ_H 3.10) with H-7 and 17-CH₃. Finally, the absolute configuration of 4 was determined to be 3S,4S,5S,9S,10S,13S,14S,15S by ECD analysis (Fig. 4), and it was given the trivial name 3-O-methyl-14-hydroxyicatrichanone.

Compounds 1 – 4 were assessed for cytotoxicity (Table 3) against the MDA-MB-435 human melanoma cell line but did not show measurable inhibitory activity. Compounds 2 and 3 were also screened against the MDA-MB-231 and OVCAR3 cell lines but were found to be inactive as well. This is in sharp contrast to the significantly stronger activity against the same cell lines by humirianthol and icacinol, which possess an γ-lactone ring between C-4 and C-6 [8]. This may suggest that the lactone moiety is essential to activity, though a possible mechanism of action cannot be proposed at this time.

Table 3.

Cytotoxic activity of Compounds 1 – 4 in Human Cancer Cell Lines

Compound	MDA-MB-435		MDA-MD-231	OVCAR3
	Cell Survivala	IC50 (μM)	IC50 (μM)	IC50 (μM)
1	100%	NTb	NTb	NT b
2	100%	>25	>25	>25
3	100%	>25	>25	>25
4	100%	NTb	NTb	NTb
Icacinol	NTb	1.25c	7.30c	7.55c
Humirianthol	NTb	1.65c	3.73c	4.12c
Taxold	8%e	0.1 nM	171.40 nM	1.45 nM

^aMeasured at 5μM.

^bNot tested.

^cIncluded for reference [8,9].

^dPositive control.

^eMeasured at 1 nM.

Several nor-pimaranes have been reported from I. trichantha, but the only compound with a similar 19-nor-structure is icacintrichanone, which is actually a 17,19-di-nor-pimarane [12]. A review of the literature shows several 19-nor-diterpenoids derived from plant species in the Erythroxylaceae [19], Annonaceae [20], Meliaceae [21], Euphorbiaceae [22], and Araliaceae [23] families. One group that reported [24] such a compound (19-nor-kauran-4α-ol-17-oic acid), coincidentally also from a West African species Annona senegalensis known to produce kaurene diterpenoids [25,26], hypothesized the nor-diterpenoid may be an artifact due to oxidation [25]. Isolated kaurenes possessed aldehyde and carboxylic acid groups at the C-4 position, which the authors suspected oxidized spontaneously to a hydroxy group, resulting in the nor-diterpenoid. However, there were no reports of structural analogues with oxidation at C-6 as observed with compounds 1 – 4. Though this natural product was similarly a 19-nor-diterpenoid, we suspect their respective biogenesis differed. However, compounds 3 and 4 are suspected to be artifacts resultant of the isolation process using methanol solvent. Previous reports of isolation of compounds with hemiketal groups determined isolation schemes can result in artifacts with converted hemiacetyl groups [27].

At initial examination, the structural similarity between compound 1 and chemotaxonomically prominent humirianthol suggests a possible modification via decarboxylation of the lactone moiety. A search of the literature did not produce examples of such a transformation except for one recent report. A closely related 19-nor-diterpenoid, norannonalide, was reported following bioconversion of annonalide by growing cells of Fusarium oxysporum f. sp. tracheiphilum (UFCM 0089) cells [28]. The proposed pathway has been applied to compound 1 (Scheme 1). The nor-annonalide structure is nearly identical to icatrichanone just as annonalide is similar to humirianthol (compound V) with the only major difference being the loss of the tetrahydrofuran D-ring due to lack of cyclization of a hydroxyacetal. The proposed pathway identified the formation of a C-4 carboxylic acid intermediate (compound V) via hydrolysis as important to the bioconversion process. This intermediate is similar to the synpimaradien-19-oic acid product (compound II) of CYP99A3 catalyzed reactions of the momilactones [29,30], lactone diterpenoids structurally similar to those from I. trichantha. Based on theoretical modeling, the intermediate could plausibly interact with the catalytic site of pyruvate decarboxylase to produce the final product. We hypothesize a similar biogenetic pathway may have resulted in compounds 1 – 4.

Scheme 1 (top) and 2 (bottom).

Proposed biogenesis for compound 1 involving decarboxylation of known compound humirianthol (V) in Scheme 1 and decarboxylation of pimarane precursor (II) followed by oxygenation in Scheme 2.

An alternative biogenetic pathway was suggested based upon two observations from the literature. First, higher plants possess step-wise mechanisms for the demethylation of C-4 dimethyl-sterols in the biosynthesis of phytosterols. There is well-established evidence these processes involve the demethylation of C-4 intermediates [31,32], decarboxylation of C-4 [33–35] or other alternative mechanisms that result in a product with only one methyl group removed. Second, another closely related 19-nor-diterpenoid, momilactone E, was previously isolated from Oryza sativa [36]. No indication of the possibility of momilactone E being an artifact has been published to our knowledge. We propose a possible biogenesis for compounds 1 – 4 combining these two concepts (Scheme 2) where the C-4 decarboxylation occurs earlier in the biogenetic pathway than Scheme 1. O. sativa is well-known to produce momilactones, and a biogenetic pathway has been proposed [29,30] that involves the mevalonate pathway to produce a (9βH)-pimara-7,15-diene precursor (compound I). Cytochrome P450s, including CYP99A2 and CYP99A3 [29], are also involved to produce syn-pimara-7,15-dien-19-oic acid (compound II) from which further lactone derivatives can be formed, such as humirianthol. Alternatively, these enzymes may not be involved and there is only C-4 demethylation of the precursor. A study of O. sativa biosynthetic gene clusters also identified hydroxylase activity due to CYP76M6 and CYP76M8, which can result in β-hydroxylation of C-6 [37]. Therefore, oxidation of a 4α-methyl intermediate (compound VI) that is also hydroxylated by CYP76M6/CYP76M8 or a similar enzyme, may results in a 19-nor-diterpenoid with oxidation at C-6 specifically (compound VII/VIII). Along this line of thought, the alternative biogenetic pathway in Scheme 2 is proposed. This scheme may also explain the unique structure of momilactone E since no biogenesis has been proposed to our knowledge. However, the biosynthetic origin for icatrichanone and related compounds is currently unclear and requires further analysis.

In summary, four new Icacina diterpenoids were isolated from the tuber of I. trichantha, all belonging to the unusual 19-nor-pimarane subclass. In comparison with other pimarane-type compounds found in Icacina, they are characterized by the lack of a γ-lactone ring between C-4 and C-6. These findings highlight the chemotaxonomic uniqueness of the Icacina genus and their capacity to produce a chemically diverse class of rare pimarane diterpenoids. While taking into consideration the unclear biogenetic origins for compounds 1 – 4, they represent an expansion of the structural diversity from an already distinctive field of 9βH-pimaranes discovered from this species and an insight into their structure-activity relationship in human cancer cells.

Materials and Methods

General Experimental Procedures

Optical rotations at the sodium D line were measured with a Perkin-Elmer 241 digital polarimeter using a quartz cell with a path length of 100 mm at room temperature. Concentrations (c) are given in g/100 mL. IR spectra were measured on a Nicolet 380 Fourier Transform Infrared Spectrometer and analyzed with OMNIC software. UV spectra were collected on a Shimadzu UFLC system with a photodiode array detector. The ECD spectra were obtained on a JASCO J-810 spectrometer. NMR spectra were recorded on a Bruker AV-400HD spectrometer. All chemical shifts were quoted on the δ scale in ppm using residual solvent as the internal standard (methanol: δ_H 3.31 for ¹H NMR, δ_C 49.00 for ¹³C NMR; chloroform: δ_H 7.26 for ¹H NMR, δ_C 77.16 for ¹³C NMR). Coupling constants (J) are reported in Hz. For HPLC purification, a C₁₈ semi-preparative HPLC column (Agilent SB-C18 column, 250 × 9.4 mm, 5 μm) and a Shimadzu UFLC system were used. HRESIMS were measured on a Waters SYNAPT hybrid quadrupole/time of flight spectroscopy using positive electrospray ionization. Human melanoma cancer cells MDA-MB-435, human breast cancer cells MDA-MB-231, and human ovarian cancer cells OVCAR3 were purchased from the American Type Culture Collection (Manassas, VA). Molecular models in Figure 2 were generated by Chem3D Pro 19.0 using MM2 force field calculation.

Plant Material

Tubers of Icacina trichantha Oliv. were collected in June 2011 from the Orba village in Nsukka of the Enugu State, Nigeria, and authenticated by Prof. B.O. Olorede of the Botany Department, University of Abuja, Nigeria, and Mr. A. Ozioko, botanist at the BDCP Laboratories, Nsukka, Nigeria. A voucher specimen (UNN/FVM 456) was deposited in the pharmacology laboratory at the University of Nigeria, Nsukka, Nigeria.

Extraction and Isolation

The dried tubers of I. trichantha (5.0 kg) were milled and extracted with 80% aqueous acetone by percolation to yield 483 g of crude extract. The crude extract was partitioned into EtOAc-soluble (98 g), n-BuOH-soluble (109 g), and H₂O-soluble (276 g) fractions. The EtOAc fraction (98 g) was separated into 40 sub-fractions on a silica gel column (10 × 60 cm) eluted with gradient hexane and acetone mixtures (from 100:0 to 0:100, v/v; 700 mL each). The combined sub-fractions 20–24 was further separated into 32 fractions on a silica gel column (3 × 35 cm) with gradient dichloromethane and acetone mixtures (from 100:0 to 50:50, v/v, 100 mL each). Fractions 21, 22, and 23 were combined and further separated using MCI Gel® CHP20P column chromatography (2.5 × 30 cm) with gradient MeOH-H₂O mixtures (from 10:90 to 100:0, v/v, 100 mL each) into 17 fractions. The 7^th and 8^th fractions were purified by semi-preparative HPLC (MeOH-H₂O, 33:67, v/v, 3.0 mL/min) to afford icatrichanone (1, 12.9 mg, t_R = 25.3 min, soluble in MeOH) and 14-hydroxyicatrichanone (2, 13.0 mg, t_R = 29.5 min, soluble in MeOH). The combined sub-fractions 25–27 from the EtOAc fraction was further separated into 67 fractions on a silica gel column (2.5 × 30 cm) eluted with gradient dichloromethane and acetone mixtures (from 100:0 to 3:7, v/v, 150 mL each). Sub-fractions 40–44 were combined and further separated into 56 fractions using MCI Gel CHP20P column chromatography (2.5 × 30 cm) with gradient MeOH-H₂O mixtures (from 5:95 to 100:0, v/v, 150 mL each). Combined sub-fractions 23–34 were purified by semi-preparative HPLC (38:62 v/v, 4.0 mL/min) to afford 3-O-methyl-icatrichanone (3, 1.3 mg, t_R = 15.3 min, soluble in MeOH) and 3-O-methyl-14-hydroxyicatrichanone (4, 2.4 mg, t_R = 16.1 min, soluble in MeOH and CHCl₃).

Icatrichanone (1):

colorless amorphous powder; ${\left[\alpha \right]}_{\text{D}}^{25}=-48$ (c 0.1, MeOH); UV_max: 199, 245 nm; IR ν_max 3361, 2921, 1670, 1632, 1465, 1378, 1088, 1047 cm⁻¹; ¹H NMR (methanol-d₄, 400 MHz) and ¹³C NMR (methanol-d₄, 100 MHz), see Table 1; (+)−HRESIMS m/z 335.1837 [M + H]⁺; calcd for C₁₉H₂₇O₅⁺, 335.1853.

14-Hydroxyicatrichanone (2):

Colorless amorphous powder; ${\left[\alpha \right]}_{\text{D}}^{25}=-165$ (c 0.33, MeOH); UV_max: 199, 242 nm; IR ν_max 3368, 2935, 1669, 1042; ¹H NMR (methanol-d₄, 400 MHz) and ¹³C NMR (methanol-d₄, 100 MHz), see Table 1; (+)−HRESIMS m/z 351.1817 [M + H]⁺ (calcd for C₁₉H₂₇O₆, 351.1802).

3-O-Methyl-icatrichanone (3):

Colorless amorphous powder; ${\left[\alpha \right]}_{\text{D}}^{25}=-63$ (c 0.03, MeOH); UV_max: 200, 242 nm; IR ν_max 3342, 2943, 1631, 1026; ¹H NMR (methanol-d₄, 400 MHz) and ¹³C NMR (methanol-d₄, 100 MHz), see Table 2; (+)−HRESIMS m/z 349.2014 [M + H]⁺ (calcd for C₂₀H₂₉O₅, 349.2010).

3-O-Methyl-14-hydroxyicatrichanone (4):

Colorless amorphous powder; ${\left[\alpha \right]}_{\text{D}}^{25}=-34$ (c 0.1, MeOH); UV_max: 203, 240 nm; IR ν_max 3424, 2933, 1673, 1044; ¹H NMR (methanol-d₄, 400 MHz) and ¹³C NMR (methanol-d₄, 100 MHz), see Table 2; (+)−HRESIMS m/z 365.1982 [M + H]⁺ (calcd for C₂₀H₂₉O₆, 365.1959).

Computational Section for ECD Calculation

The systematic random conformational analyses were performed in the SYBYL-X-2.1 program by using MMFF94s molecular force field, with an energy cutoff of 10 kcal/mol to the global minima. All of the obtained conformers were further optimized using DFT at the b3lyp/6–31+g(d) level in gas phase by using Gaussian09 software [38]. The optimized stable conformers were used for ECD calculations at the cam-b3lyp/6–31+g(d) level with the PCM solvation model of methanol, with the consideration of the first 30 excitations. The overall ECD curves were all weighted by Boltzmann distribution. The calculated ECD spectra were subsequently compared with the experimental ones. The ECD spectra were produced by SpecDis 1.71 software with the UV correction of +5 nm (2 and 3), +10 nm (1), and +8 nm (4), respectively [39,40].

Cytotoxicity Assays

Experimental procedures for the cytotoxicity assays were adapted from previous studies [8,9].

Abbreviations

DEPT

distortionless enhancement by polarization transfer

ECD

electronic circular dichroism

EtOAc

ethyl acetate

HSQC

heteronuclear single quantum correlation

HRESIMS

high-resolution electrospray ionization mass spectrometry

TDDFT

time-dependent density functional theory