Anti-HIV tigliane diterpenoids from Reutealis trisperma

Abstract

Bioassay-guided fractionation of the n-butanol extract from the branches and leaves of Reutealis trisperma resulted in the isolation of six undescribed (crotignoids L~Q) together with two known (12-deoxyphorbol-13-hexadecanoate and 12-deoxyphorbol-13-myristate) tigliane diterpenoids. Their structures, especially the absolute configurations, were determined from extensive spectroscopic studies, including 2D NMR spectra, CD data analysis and electronic circular dichroism (ECD) calculations. All isolates were tested for anti-HIV activity against HL4–3 virus in MT4 cells. Except for crotignoid Q, the remaining seven tigliane diterpenoids exhibited potent anti-HIV activity with IC₅₀ values ranging from 0.0023 to 4.03 μM.

Graphical Abstract

1. Introduction

Euphorbiaceae plants are rich in diterpenoids with varied biological activities (Shi et al., 2008; Vasas et al., 2014; Wang, et al., 2015), such as analgesic, acetylcholinesterase inhibitory, antifungal, antimicrobial, antimycobacterial (Wang et al., 2017), antitubercular (Zhao et al., 2016), anti-inflammatory, cytotoxic (Chen, et al., 2016;), antitumor, and antiviral (Esposito, et al., 2017; Nothias, et al., 2017; Olivon, et al., 2017; Remy, et al., 2017) effects. Particularly, some tigliane and ingenane diterpenoids from Euphorbiaceae species showed potent anti-HIV activity in vitro with impressive IC₅₀ values and low cytotoxicity (Huang et al., 2014; Pan et al., 2011; Chen, et al., 2017; Wang et al., 2017; Abreu et al., 2014; Nothias-Scaglia et al., 2015). Moreover, tigliane diterpenes inhibited the replication of wild-type HIV-1 and HIV-2 strains as well as drug-resistant strains by reducing the infectivity of the progeny virus (Chen, et al., 2017). Consequently, Euphorbiaceae plants should be an important source of new anti-HIV compounds for eradication of the AIDS virus.

Reutealis trisperma (Blanco) Airy Shaw (synonym: Aleurites trisperma Blanco), belonging to Euphorbiaceae family, has been reported as a potential non-edible source for biodiesel production based on the high content of free fatty acids in its seed oil (Kumar et al., 2015). Our preliminary study found that an extract of the branches and leaves of R. trisperma exhibited strong anti-HIV-1 activity (IC₅₀ < 1 μg/mL). The subsequent investigation led to the isolation of six undescribed (1–6) and two known (7, 8) tigliane diterpenoids. The present paper describes the isolation, structure elucidation, and anti-HIV evaluation of these tigliane diterpenoids.

2. Results and discussion

Eight tigliane diterpenoids (1–8) (Fig. 1) were obtained from the n-butyl alcohol extract of R. trisperma by anti-HIV activity-guided fractionation. Their structures were elucidated by 2D-NMR, HRESIMS and electronic circular dichroism (ECD) spectra.

Figure. 1.

Structures of compounds 1–8.

Compound 1 was purified as a light yellow oil and its molecular formula was determined as C₃₂H₄₈O₇ based on an HRESIMS ion at m/z 567.3298 [M + Na]⁺ (calcd. C₃₂H₄₈O₇Na, 567.3292), requiring nine degrees of unsaturation. The IR spectrum of 1 exhibited absorption bands at 3440 and 1714 cm⁻¹ corresponding to hydroxy and carbonyl groups. Its ¹³C NMR (Table 1) and DEPT spectra revealed 32 carbon signals, including two carbonyl groups at δ_C 205.5 (C-3) and δ_C 202.5 (C-7), one ester group at δ_C 176.1 (C-1'), two double bond pairs [δ_C 160.5 (C-1) & δ_C 135.0 (C-2), δ_C 137.1 (C-5) & δ_C 148.5 (C-6)], four oxygenated carbons [δ_C 73.6 (C-9), δ_C 63.4(C-13), δ_C 63.6 (C-20) and δ_C 73.3 (C-4)] and five methyls [δ_C 14.2 (C-12'), δ_C 10.5 (C-19), δ_C 15.5 (C-17), δ_C 18.7 (C-18) and δ_C 23.1 (C-16)]. Interpretation of the 2D NMR (Fig. 2) spectra of 1, especially HMBC data, indicated that compound 1 is a tigliane-type diterpenoid with the following characteristics. The HMBC correlations of H-1/C-3, C-4, C-10 and C-19; H-19/C-1 and C-3; and H-10/C-1 confirmed the presence of ring-A (five-membered ring). Similarly, the HMBC correlations of H-5/C-4, C-7 and C-20; H-20/C-5, C-6 and C-7; and H-8/C-7 and C-9 supported the presence of ring-B (seven-membered ring). The correlations of H₃−18/C-12 and H-12/C-9, C-11 and C-13 facilitated the construction of the ring-C. Subsequently, the cyclopropyl ring-D with gem-dimethyl substitution was also established based on the mutual correlations between both methyl groups with C-13, C-14 and C-15. The ¹³C NMR data (Table 1) of 1 and crotignoid G (Zhang et al., 2015) were highly similar except for the markedly upshifted signal of C-12 (from δ_C 77.3 to δ_C 31.8) in the former compound, indicating that a hydroxyl was not present at C-12 in 1. This absence was supported by the correlations between C-12 and H-12a/b (δ_H 2.11, 1H, m & δ_H 1.59, 1H, m) in the HSQC spectrum. In addition, the molecular weight of 1 was 12 units more than that of crotignoid G, which suggested that the decanoyl side chain in crotignoid G was replaced by a dodecanoyl group (+28–16 = 12). The location of the long side chain at C-13 (δ_C 63.4) was supported by a 4-bond HMBC correlation between H-12 and C-1'. Additionally, the observed ROESY correlations (Fig. 2) from H-14 (δ_H 1.61) to H₃-18 (δ_H 0.94) and H₃−18 (δ_H 0.94) to H-10 (δ_H 3.29) indicated that these protons were on the same side of the molecule and involved in a α configuration. Based on the above evidence, the structure of 1 (crotignoid L) was established as shown in Fig. 1.

Table 1.

¹H (600 MHz) and ¹³C (150 MHz) data of 1–3 in CDCl₃ (δ in ppm, J in Hz).

	1		2		3
Position	δC	δH	δC	δH	δC	δH
1	160.5	7.65, brs	160.3	7.66, brs	160.4	7.66, brs
2	135.0		135.0		135.0
3	205.5		205.2		205.2
4	73.3		73.3		73.3
5	137.1	6.89, brs	137.3	6.87, brs	137.3	6.88, brs
6	148.5		148.5		148.5
7	202.5		202.7		202.6
8	55.4	3.48, d (5.3)	55.4	3.46, d (5.4)	55.4	3.47, d (5.4)
9	73.6		73.6		73.5
10	58.8	3.29, m	58.8	3.31, m	58.8	3.30, m
11	38.4	2.05, m	38.4	2.04, m	38.4	2.04, m
12	31.8	2.11, m	31.8	2.10, m	31.8	2.11, m
		1.59, m		1.59, m		1.59, m
13	63.4		63.4		63.4
14	25.7	1.61, d (5.4)	25.7	1.62, d (5.5)	25.7	1.62, d (5.5)
15	22.6		22.6		22.6
16	23.1	1.17, s	23.1	1.17, s	23.1	1.17, s
17	15.5	1.02, s	15.5	1.02, s	15.5	1.02, s
18	18.7	0.94, d (6.0)	18.6	0.94, d (6.4)	18.7	0.94, d (6.3)
19	10.5	1.82, brs	10.5	1.83, brs	10.5	1.83, brs
20	63.6	4.38, d (13.9)	63.9	4.38, d (13.9)	63.9	4.38, d (13.9)
		4.26, d (13.9)		4.26, d (13.9)		4.27, d (13.9)
1'	176.1		176.1		176.1
2'	34.7	2.31, t (18.0)	34.7	2.31, t (12.0)	34.7	2.31, t (12.0)
3'	24.8	1.62, m	24.8	1.62, m	24.8	1.62, m
4'	22.6	1.26, m*	22.6	1.26, m*	22.6	1.26, m*
5'	29.2*	1.26, m*	29.3*	1.26, m*	29.3*	1.26, m*
6'	29.4*	1.26, m*	29.4*	1.26, m*	29.4*	1.26, m*
7'	29.5*	1.26, m*	29.5*	1.26, m*	29.5*	1.26, m*
8'	29.6*	1.26, m*	29.6*	1.26, m*	29.6*	1.26, m*
9'	29.7*	1.26, m*	29.8*	1.26, m*	29.8*	1.26, m*
10'	32.0	1.26, m*	29.8*	1.26, m*	29.8*	1.26, m*
11'	22.8	1.30, m	29.8*	1.26, m*	29.8*	1.26, m*
12'	14.2	0.88, t (12.0)	32.1	1.26, m*	29.8*	1.26, m*
13'			22.8	1.30, m	29.8*	1.26, m*
14'			14.3	0.88, t (12.0)	32.1	1.26, m
15'					22.8	1.30, m
16'					14.3	0.87, t (12.0)

^*signal overlapping

Figure. 2.

Key 2D-NMR correlations of 1–3.

Compounds 2 and 3 were obtained as yellow oils. Their molecular formulas were determined as C₃₄H₅₂O₇ and C₃₆H₅₆O₇, respectively, based on HRESIMS results, with the same nine degrees of unsaturation as 1. Detailed analysis of the 1D- and 2D-NMR spectroscopic data revealed that, compared with 1, compounds 2 and 3 have the same skeleton (Table 1) but, rather than a dodecanoyl ester, have myristoyl and palmitoyl esters, respectively, at C-13. All three compounds have the same relative configurations based on significant 2D NMR resemblance (Fig. 2). Consequently, the structures of 2 (crotignoid M) and 3 (crotignoid N) were unequivocally characterized as shown in Fig. 1.

Compound 4 (yellow oil) was assigned a molecular formula of C₃₂H₅₀O₇ based on the HRESIMS ion peak at m/z 569.3456 [M + Na]⁺ (calcd. C₃₂H₅₀O₇Na, 569.3449), requiring eight indices of hydrogen deficiency. The IR spectrum showed absorption bands for hydroxyl (3406 cm⁻¹) and carbonyl (1707 cm⁻¹) groups. Analysis of the NMR data (Table 2) suggested a tigliane scaffold for 4 with high similarity to 1. In a comparison of the ¹³C NMR data of the two compounds, C-7 in 4 was shifted significantly upfield (δ_C 70.6 in 4, δ_C 202.5 in 1), consistent with a reduction of the ketone at this position in 1 to a secondary alcohol in 4, which was confirmed by the HSQC correlation of C-7 with H-7 (δ_H 4.82, d, J = 8.88 Hz) and HMBC correlations of H-7 with C-5, C-6, C-8 and C-9 (Fig. 3) in 4. The NOESY correlation between H-7 and H-10 in 4 suggested that H-7 is α-oriented (Fig. 3). Thus, compound 4 (crotignoid O) was characterized fully as shown in Fig. 1.

Table 2.

¹H (600 MHz) and ¹³C (150 MHz) data of 4–6 in CDCl₃ (δ in ppm, J in Hz).

	4		5		6
Position	δC	δH	δC	δH	δC	δH
1	160.3	7.64, brs	160.2	7.59, brs	160.7	7.58, brs
2	134.1		133.5		133.6
3	206.5		207.7		208.5
4	73		75.6		72.9
5	126.7	6.11, brs	37.3	2.47, d (117.0)	34.7	2.91, d (19.7)
				2.10, d (17.0)		2.47, d (19.7)
6	153.1		62.5		143.0
7	70.6	4.82, d (8.9)	63.4	3.32, d (8.3)	158.3	6.74, dd (2.1,21)
8	48.7	2.29, m	42.5	1.95, dd (4.6,4.6)	41.6	3.39, t (12.0)
9	73.2		74.3		74.3
10	56.1	3.09, m	55.6	3.44, brs	55.9	3.11, m
11	37.9	2.05, m	36	1.90, m	36.7	1.90, m
12	31.8	2.10, m	31.6	2.03, m	31.9	2.11, m
		1.30, m		1.57, m		1.62, m
13	64.1		63		63.1
14	27.8	1.32, m	30.2	0.98, d (4.6)	32.3	0.98, d (4.6)
15	22.8		22.8		22.8
16	23.1	1.23, s	23.3	1.23, s	23.3	1.26, s
17	15.9	1.04, s	15.6	1.06, s	15.4	1.11, s
18	18.7	0.94, d (6.5)	18.1	0.88, m	18.7	0.91, m
19	10.4	1.81, brs	10.4	1.79, dd (2.7, 1.1)	10.2	1.81, dd (1.2,1.2)
20	67.5	4.37, dd (13.9,13.9)	65.6	3.53, d (7.4)	193.9	9.45, s
		4.31, dd (13.9,13.9)		3.53, d (7.4)
1'	176.3		176.3		176.3
2'	34.8	2.32, t (12.0)	34.8	2.31, t (12.0)	34.8	2.35, t (18.0)
3'	24.9	1.61, m	24.9	1.58, m	24.9	1.62, m
4'	22.8	1.31, m*	22.8	1.26, m*	22.8	1.26, m*
5'	29.3*	1.31, m*	29.3*	1.26, m*	29.3*	1.26, m*
6'	29.4*	1.31, m*	29.4*	1.26, m*	29.4*	1.26, m*
7'	29.5*	1.31, m*	29.5*	1.26, m*	29.5*	1.26, m*
8'	29.6*	1.31, m*	29.6*	1.26, m*	29.6*	1.26, m*
9'	29.7*	1.31, m*	29.7*	1.26, m*	29.7*	1.26, m*
10'	32.1	1.31, m*	29.7*	1.26, m*	32.1	1.26, m*
11'	22.8	1.22, m	29.7*	1.26, m*	22.8	1.22, m
12'	14.3	0.91, m	32.1	1.26, m*	14.3	0.91, m
13'			22.8	1.22, m
14'			14.3	0.88, m

^*signal overlapping

Figure. 3.

Key 2D-NMR correlations of 4.

Compound 5 was obtained as a yellow oil. Its molecular formula, C₃₄H₅₆O₈, with seven indices of hydrogen deficiency was determined from the negative-ion HRESIMS peak at m/z 591.3904 [M - H]⁻ (calcd. for 591.3902). The NMR data (Table 2) of 5 and 4 were comparable, indicating that these two compounds have similar carbon skeletons. However, the carbon signals at C-5 and C-6 were moved upfield from δ_C 126.7 and δ_C 153.1 in 4 to δ_C 37.3 and δ_C 62.5, respectively, in 5, consistent with hydration of the double bond between C-5 and C-6 in 4. Refer that C-6 is an oxygenated tertiary carbon. The HSQC correlations of δ_C 37.3 (C-5)/δ_H 2.47 & 2.10 (2H, d, J = 16.96 Hz, H-5ab) (see Supplementary data S40) and the HMBC correlations of H-5/C-4, C-6 and C-20 (Fig. 4) supported the above assignment. The UPLC-MS fragment ion peak at m/z 329 [M - 2H₂O - C₁₃H₂₇COOH + H]⁺ suggested that the ester side chain is a myristoyl (C₁₃H₂₇COO) moiety (see Supplementary data S46). Furthermore, a correlation between H-7 (δ_H 3.32) and H-10 (δ_H 3.44) of 5 was observed in the NOESY spectrum indicating that H-7 was α-oriented and, thus, the 7-OH was oppositely oriented. Additionally, a correlation between H-7 (δ_H 3.32) and H₂-20 (δ_H 3.53) indicated that the hydroxymethyl at C-6 was also α-oriented, while the hydroxy at C-6 was β-oriented. Thus, the structure of 5 (crotignoid P) was assigned as shown in Fig. 1.

Figure. 4.

Key 2D-NMR correlations of 5.

Compound 6 was obtained as a yellow oil. Its molecular formula was determined as C₃₂H₄₈O₆ from its positive-ion HRESIMS peak at m/z 551.3339 [M + Na]⁺ (calcd. for 551.3343), with two fewer protons than the known compound 12-deoxyphorbol-myristate (8, Zhang et al., 2015). Moreover, the ¹H and ¹³C-NMR spectroscopic data of 6 (Table 2) were extremely similar to those of 8, except that the signals [δ_C 68.3 and δ_H 3.94 (1H, d, J = 12.6 Hz) & 4.01(1H, d, J = 12.6 Hz)] for a hydroxymethyl group attached at C-6 in 8 were replaced by signals [δ_C 193.9 and δ_H 9.45, (1H, s)] for a C-20 formyl group in 6. This structural assignment was supported by key HMBC correlations from H-20 [δ_H 9.45, (1H, s)] to C-5 (δ_C 34.7) and C-6 (δ_C 143.0) and from H-7 [δ_H 6.74, 1H, dd, J = 2.11 Hz] to C-20 (δ_C 193.9) (Fig. 5). Other HMBC correlations of H-1/C-3, C-4 and C-10, H-5/C-6 and C-7, H-7/C-14 and C-20, H-12/C-11, C-13 and C-1', H-19/C-1 and C-3 as well as COSY cross-peaks of H-1/H-10, H₃-18/H-11, H-11/H-12, H-7/H-8 and H-8/H-14 further characterized the carbon skeleton of 6 (Fig. 5). The NOESY correlation between H-14 (δ_H 0.98) and H₃-18 (δ_H 0.91) suggested that H₃-18 are α-oriented, while the correlation between H-8 and H-17 indicated that H-8 and H-17 are β-oriented. Finally, the structure of 6 (crotignoid Q) was elucidated as shown in Fig. 1.

Figure. 5.

Key 2D-NMR correlations of 6.

Compounds 7 and 8 were identified as 12-deoxyphorbol-13-hexadecanoate and 12-deoxyphorbol-13-myristate, respectively, by comparing the experimental NMR data with those reported for the known compounds (Ma et al., 1997; Kotsonis et al., 1996).

Unfortunately, many attempts failed to produce a single crystal of the isolated tigliane-type compounds. Thus, the absolute configuration of 1 was determined by a comparison of theoretical and experimental ECD spectra. As illustrated in Fig. 6, the calculated and measured ECD curves matched well, leading to the assignment of the absolute configuration of 1 as 4R,8S,9S,10S,11R,13S,14R. Because compounds 1–6 have the same skeleton, the absolute configurations of 2–6 were assigned by the comparison of their ECD spectra with that of 1 (Fig. 7). The absolute configurations of 2, 3 and 6 were also assigned as 4R,8S,9S,10S,11R,13S,14R, while those of 4 and 5 were assigned as 4R,7S,8S,9S,10S,11R,13S,14R and 4R,6R,7S,8S,9S,10S,11R,13S,14R, respectively.

Figure. 6.

Calculated and experimental ECD spectra of 1.

Figure. 7.

Experimental ECD spectra of 1–6.

All isolated compounds (1–8) were evaluated for anti-HIV activity and cytotoxicity toward MT4 cells in vitro. The results were presented in Table 3. Compound 7 exhibited the most potent anti-HIV-1 activity (IC₅₀ = 0.0023 μM), similarly to previous reports (Olivon, et al., 2017). With low cytotoxicity (CC₅₀ > 1.7 μM), its selectivity index (SI) was greater than 714, about 35 times higher than that of the positive control (AZT). Compounds 2 and 8 showed lower but still potent anti-HIV activity (IC₅₀ = 0.038 and 0.022 μM, respectively); their SI values were greater than 45.5 and 83.3, respectively. Compounds 1, 3, 4 and 5 were less potent (SI > 0.47). Compound 6 did not show anti-HIV activity, which might be due to the presence of a carbonyl group at C-20. The comparison of 4 with 1, 2 and 3 indicated that the hydroxy group at C-7 had a negative impact on anti-HIV activity. Besides OH-7, OH-6 might also play a role in reducing the activity of 5 (vs 7). The same structure-activity relationship was also observed between 12-deoxy-5β-6β-7α-trihydroxy-phorbol-13-hexadecanoate (IC₅₀ = 0.257 μM) and 12-deoxy-5β-hydroxy-phorbol-13-hexadecanoate (IC₅₀ = 0.002 μM) in the literature (Olivon, et al., 2017). Besides, compound 7 was more potent than 8, which might be due to its longer ester side chain.

Table 3.

Anti-HIV-1 activity and cytotoxicity of 1–8^a.

Compounds	Anti-HIV-1	Cytotoxicity	Selectivity index
	IC50 (μM)	CC50 (μM)	SI
1	0.22	> 1.8	> 8.3
2	0.038	> 1.7	> 45.5
3	0.31	> 1.6	> 4.7
4	4.03	> 1.8	> 0.47
5	0.10	> 1.6	> 16.1
6	> 1.9	> 1.9	/
7	0.0023	> 1.7	> 714
8	0.022	> 1.8	> 83.3
AZT	0.0187	> 0.37	> 20

^aThe values are means ± SD (n = 3). AZT (zidovudine) was used as a positive control.

NF: not found.

3. Conclusions

Anti-HIV activity-guided investigation of the branches and leaves of R. trisperma led to the isolation and purification of eight tigliane diterpenoids (1–8). The stereostructures of six compounds were assigned based on extensive spectroscopic data and ECD calculations. Except for 6, the seven remaining tigliane diterpenoids exhibited anti-HIV activity with IC₅₀ values ranging from 0.0023 to 4.03 μM and SI values from 0.47 to 714. The presence of a carbonyl group at C-20, as well as a hydroxy group at C-6/C-7, might have a detrimental effect for anti-HIV activity.

4. Experimental

4.1. General experimental procedures

Optical rotation data were obtained using an Autopol V Plus instrument (Rudolph, Hackettstown, NJ, USA) at 25 °C. IR data were measured on a PE Spectrum RXI spectrophotometer (PerkinElmer) using KBr disks. The UV and ECD spectra were obtained using a Chirascan Plus (Applied Photophysics Ltd., UK). NMR spectra were recorded on Bruker AMX-400 MHz and AMX-600 MHz instruments in CDCl₃ with TMS as an internal standard. Mass spectra were obtained on a Dionex Ultimate 3000 UPLC instrument with an LTQ Velos Pro MS spectrometer (Thermo Fisher Scientific, USA). HRESIMS were acquired with a Bruker Daltonics APEXIII 7.0 TESLA FTMS system (Bruker Daltonics, Billerica, MA, USA). Preparative HPLC was performed on an Agilent 1260 (Agilent Technologies) with a JAI C18 column (20 × 240 mm, 5 μm).

4.2. Plant Material

Fresh branches and leaves of Reutealis trisperma (Blanco) Airy Shaw (Euphorbiaceae) were collected on December 19, 2010 from the campus of National Pingtung University of Science and Technology (NPUST), Neipu, Pingtung, Taiwan [22°38'33.6″N 120°36'42.4″E] by Professor Sheng-Zehn Yang. A voucher specimen is deposited at NPUST, Taiwan.

4.3. Anti-HIV and Cytotoxicity Assays

The cytotoxic and anti-HIV-1 activities of extracts and chromatographic fractions as well as compounds 1–8 were measured according to the method developed by Chen and Lee (Liu et al., 2019). The selectivity index (SI) was calculated as the ratio of peritoneal macrophage CC₅₀ and anti-HIV IC₅₀. AZT was tested as a positive control.

4.4. Extraction and isolation

Dried and powdered branches and leaves of R. trisperma (6 kg) were extracted with EtOH-H₂O (95:5; 3 × 30 L × 2 h). The crude extract (330 g, IC₅₀ < 1 μg/mL) was suspended in H₂O (5 L) and partitioned with n-butyl alcohol (3 × 5 L). The n-butyl alcohol extract (108 g, IC₅₀ < 1 μg/mL) was chromatographed on a silica gel column (500 g) eluted with CH₂Cl₂-MeOH (95:5) to give fractions Fr. 1–4. Fr. 2 (11.5 g, IC₅₀ < 1 μg/mL) was separated by silica gel column chromatography (300 g) using CH₂Cl₂-MeOH (95:5) to afford fractions Fr. 2.1–2.5. Column chromatography of Fr. 2.3 (2.7 g, IC₅₀ < 1 μg/mL) over NH₂-SiO₂ gel (100 g) eluted by CH₂Cl₂-MeOH (100:0–0:100) gave five subfractions Fr. 2.3.1–2.3.5. The mixture of Fr. 2.3.2 and Fr. 2.3.3 (1.14 g, IC₅₀ < 1 μg/mL) was treated with repeated preparative HPLC on a C18 column, eluted in gradient with acetonitrile-H₂O (80:20–100:0) to yield compounds 1 (54.9 mg), 2 (8.1 mg), 3 (52.1 mg), 4 (12.2 mg), 5 (4.2 mg), 6 (5.4 mg), 7 (22.4 mg) and 8 (19.8 mg).

4.4.1. Crotignoid L (1)

light-yellow oil, ${\left[\alpha \right]}_{\text{D}}^{25}$ +6.5 (c 0.14, MeOH); UV (MeOH) λ_max (log ε): 225 (4.1) nm; CD (MeOH, nm) λ_max (Δε): +4.14 (212), 0 (230), −3.11 (249); IR (KBr) v_max: 3440, 2923, 2853, 1714, 1383, 1050 cm⁻¹; ¹H NMR (600 MHz, CDCl₃) and ¹³C NMR (150 MHz, CDCl₃) data, see Table 1. HRESIMS m/z 567.3298 [M + Na]⁺ (calcd for C₃₂H₄₈O₇Na, 567.3292).

4.4.2. Crotignoid M (2)

light-yellow oil, ${\left[\alpha \right]}_{\text{D}}^{25}$ +11.4 (c 0.30, MeOH); UV (MeOH) λ_max (log ε): 225 (4.0) nm; CD (MeOH, nm) λ_max (Δε): + 2.05 (204), 0 (236), − 1.02 (252), 0 (278), +0.57 (295); IR (KBr) v_max: 3388, 2921, 2850, 1712, 1646 cm⁻¹; ¹H NMR (600 MHz, CDCl₃) and ¹³C NMR (150 MHz, CDCl₃) data, see Table 1. HRESIMS m/z 595.3614 [M + Na]⁺ (calcd for C₃₄H₅₂O₇Na, 595.3605).

4.4.3. Crotignoid N (3)

light-yellow oil, ${\left[\alpha \right]}_{\text{D}}^{25}$ +10.3 (c 0.185, MeOH); UV (MeOH) λ_max (log ε): 225 (4.0) nm; CD (MeOH, nm) λ_max (Δε): +7.81 (215), 0 (231), −5.24 (247); IR (KBr) v_max: 3387, 2922, 2852, 1713, 1672 cm⁻¹; ¹H NMR (600 MHz, CDCl₃) and ¹³C NMR (150 MHz, CDCl₃) data, see Table 1. HRESIMS m/z 623.3927 [M + Na]⁺ (calcd for C₃₆H₅₆O₇Na, 623.3918).

4.4.4. Crotignoid O (4)

light-yellow oil, ${\left[\alpha \right]}_{\text{D}}^{25}$ +23.3 (c 0.18, MeOH); UV (MeOH) λ_max (log ε): 248 (3.8) nm; CD (MeOH, nm) λ_max (Δε): +14.4 (205), 0 (227), −4.45 (249), 0 (278); IR (KBr) v_max: 3406, 2923, 2853, 1707, 1383, 1049 cm⁻¹; ¹H NMR (600 MHz, CDCl₃) and ¹³C NMR (150 MHz, CDCl₃) data, see Table 2. HRESIMS m/z 569.3456 [M + Na]⁺ (calcd for C₃₂H₅₀O₇Na, 569.3449).

4.4.5. Crotignoid P (5)

light-yellow oil, ${\left[\alpha \right]}_{\text{D}}^{25}$ +22.2 (c 0.18, MeOH); UV (MeOH) λ_max (log ε): 247 (3.9) nm; CD (MeOH, nm) λ_max (Δε): +6.62 (212), 0 (238), −2.06 (252); IR (KBr) v_max: 3387, 2921, 2851, 1713, 1647, 1384, 1049 cm⁻¹; ¹H NMR (600 MHz, CDCl₃) and ¹³C NMR (150 MHz, CDCl₃) data, see Table 2. HRESIMS m/z 591.3904 [M – H]⁺ (calcd for C₃₄H₅₆O₈-H, 591.3902).

4.4.6. Crotignoid Q (6)

light-yellow oil, ${\left[\alpha \right]}_{\text{D}}^{25}$ +18.8 (c 0.175, MeOH); UV (MeOH) λ_max (log ε): 242 (3.8) nm; CD (MeOH, nm) λ_max (Δε): +5.77 (203), 0 (218), −6.82 (234), 0 (247), +4.86 (228); IR (KBr) v_max: 3388, 2921, 2850, 1713, 1646 cm⁻¹; ¹H NMR (600 MHz, CDCl₃) and ¹³C NMR (150 MHz, CDCl₃) data, see Table 2. HRESIMS m/z 551.3339 [M + Na]⁺ (calcd for C₃₂H₄₈O₆Na, 551.3343).

4.5. ECD Calculations

Conformational analysis was initially performed using Confab with systematic search at MMFF94 force field for undetermined relative configurations of compound 1 (Fig. S55). Long hydrocarbon chains far from the chiral centers increase trivial conformers but have little effect on ECD spectra. Thus, shortened structures for compound 1 with the −(CH₂)₁₀- removed were used in the present study.

Highlights.

Six undescribed tigliane diterpenoids were obtained from Reutealis trisperma

Absolute configuration was determined by theoretical and experimental ECD spectra

Seven diterpenoids exhibited potent anti-HIV activity