A rule to distinguish diastereomeric bianthrones by ¹H NMR

Abstract

Bianthrones have occurred widely in drug and material developments, but the stereochemistry remains in chaos so far due to the meso, racemic, and axis-rotating phenomena. Based on the computational and experimental results, an effective rule to distinguish between cis and trans bianthrones by ¹H NMR is concluded, which applies to both homo- and heterobianthrones.

---

Introduction

The 10(sp³), 10′(sp³)-bianthrones representing a family of highly-conjugated dimers have occurred widely in terrestrial and marine plants, animals, and fungi as secondary metabolites, which exhibit high molecular diversity and potent biological activities, such as cytotoxicity against Jurkat, HL60, HeLa S3, B16-F1, and KB tumor cell lines, anti-plasmodial activity against Plasmodium falciparum, and purgation.^1–10 They are also regarded as important intermediates in the biosyntheses of photosensory pigments with ingestion-deterrent, antiviral, and sedative properties, and some have been obtained as intermediates in the chemical syntheses of these pigments.^11,12 Additionally, the photochromic derivatives with a double bond between C-10 and C-10′ have attracted much attention for molecular electronics, and they may be taken as potential structural elements in supramolecular assemblages due to the conductance-switching property.^13–15 Given the high occurrence of bianthrones in drug and material developments, the detailed structure elucidation is necessary and will contribute well to understanding their function and application.

The bianthrones from natural sources and chemical syntheses are often substituted by methyl, hydroxyl, methoxyl, and other groups at different positions, which make the configurations at C-10/10′ complicated when arranged asymmetrically. Among the bianthrones with only achiral groups, a mesomer (cis/erythro/syn) and a (±) racemate (trans/threo, 10S/10′S and 10 R/10′R) are present if two anthronyls are substituted uniformly (Fig. 1). Otherwise, each planar structure corresponds to two pairs of racemates with cis and trans H-10/10′ (10S/10′R, 10R/10′S, 10S/10′S, and 10R/10′R) (Fig. 2). If there are chiral groups, at most four diastereomers may arise from the varying configurations of C-10/10′.⁶ These diastereomeric bianthrones can be separated without chiral means, although some have been obtained as a mixture. However, it is difficult to distinguish these diastereomers by achiral spectra due to the similar or even overlapped signals.⁶ Even the coupling constants of H-10/10′ and NOE spectra are untenable to establish the relative configurations without the predominant conformations being investigated.^1,8 The specific optical rotation ([α]_D) has once been used to identify them, but the possible impurity for mesomers or 1 : 1 ratio for enantiomers are neglected.^2,4,8 Additionally, the epimerization and photooxidation under acidic conditions preclude hydrolyses and derivations by some chemical methods.^6,16 Thus, the configurations for most of the substituted bianthrones have been assigned almost in chaos so far due to the meso, racemic, and axis-rotating phenomena, which may restrict the application and development of these molecules to some extent. Based on the calculations by the Dreiding force field in MarvinSketch¹⁷ and density function theory (DFT) in Gaussian 09 (ref. 18) as well as experimental and reported data, a simple rule to distinguish the diastereomeric bianthrones by ¹H NMR is described in this paper.

Fig. 1 Homobianthrones with cis (meso, (a) H_α-10/10′) and trans ((b) H_α-10/H_β-10′, (c) H_β-10/H_α-10′) H-10/10′.

Fig. 2 Heterobianthrones with cis ((a) H_α-10/10′, (b) H_β-10/10′) and trans ((c) H_α-10/H_β-10′, (d) H_β-10/H_α-10′) H-10/10′.

Results and discussion

Conformational analyses

As indicated by theoretical simulation and X-ray determination, the symmetric 10(sp³), 10′(sp³)-bianthrones prefer energy-minimized gauche forms with a face-stacked conformation rather than anti ones.¹³ Based on the quantum chemical calculations, 1,1′,8,8′-tetrahydroxybianthrone as a core of naturally occurring bianthrones (1–24) features more gauche (97.9%) than anti form (2.1%) (Fig. 3), and the anti ones decrease to 1.4% in the mesomer and 0.1% in the racemate of chrysophanol bianthrone (1). Thus, the anti conformer is suggested to be neglected relative to two predominant gauche ones with crossed rings A/B′ (I) and A′/B (II) for a cis or crossed A/A′ (III) and B/B′ (IV) for a trans H-10/10′ bianthrone (Fig. 4). It is interesting that most of the gauche conformers without any substitution at C-4/4′ and C-5/5′ feature the close dihedral angles ∠H-10–C-10–C-10′–H-10′, which may be mainly controlled by the vicinal groups in view of the smaller angle (54.6°) in III of 5-chlorochrysophanol bianthrone. The conformational analyses of these substituted bianthrones preclude any assignment of relative configurations at C-10/10′ by coupling constants or NOE experiments.

Fig. 3 Lowest-energy gauche and anti forms of 1,1′,8,8′-tetrahydroxybianthrone and their populations.

Fig. 4 Lowest-energy gauche forms of cis (H_α-10/10′; (I) crossed rings A/B, (II) crossed A′/B) and trans (H_α-10/H_β-10′; (III) crossed A/A′, (IV) crossed B/B′) diastereomers exemplified by chrysophanol bianthrone (1).

¹H NMR characteristics

As discussed above, a cis isomer prefers the crossed rings A/B′ (I) or A′/B (II) and a trans isomer favors the crossed A/A′ (III) or B/B′ (IV), which make ¹H NMR data for the overlapped parts shifting towards upfield due to their mutually-shielding effect. Concretely, the resonances relating to rings A and A′ or rings B and B′ move synchronously in each trans conformer, but oppositely in cis conformers I and II. Thus, each of rings A and B′ exhibits the lower-frequency ¹H NMR signals in I and higher-frequency ones in II than its relative A′ or B ring, but the corresponding moieties in III or IV feature the same or similar signals (taken as deduction D1). The predicted ¹H NMR data for conformers I–IV of 1 support this deduction,¹⁹ which is helpful to distinguish between cis and trans bianthrones along with the ratios of I/II and III/IV. Based on calculated results (Table 1), the OCH₃-1/1′, CH₃-3/3′, and Cl-5/5′ groups are advantageous to conformer III, and the COOH-2/2′, COOH-3/3′, OH-6/6′, OCH₃-6/6′, CH₃-7/7′, OCH₃-7/7′, and OCH₃-8/8′ groups can improve the populations of IV. These groups act roughly on the basis of their sizes and distances to C-10/10′.

Table 1 The Boltzmann populations of conformers I–IV for 10(sp³), 10′(sp³)-bianthrones^a

1,1′	2,2′	3,3′	5,5′	6,6′	7,7′	8,8′	I/%	II/%	III/%	IV/%
a Without groups at 4/4′ and 10/10′ positions.
OCH3	H	CH3	H	OH	H	OH	50.0	50.0	63.8	36.2
OCH3	H	CH3	Cl	OH	H	OH	50.0	50.0	99.7	0.3
OAc	H	CH3	H	OCH3	H	OCH3	50.0	50.0	3.3	96.7
OH	COOH	CH3	H	OH	H	OH	50.0	50.0	6.1	93.9
OH	H	CH3	H	H	H	OH	50.0	50.0	66.9	33.1
OH	H	CH3	Cl	H	H	OH	50.0	50.0	99.9	0.1
OH	H	CH3	H	OH	H	OH	50.0	50.0	52.5	47.5
OH	H	CH3	H	OH	CH3	OH	50.0	50.0	29.9	70.1
OH	H	CH3	H	OCH3	H	OH	50.0	50.0	36.8	63.2
OH	H	CH3	H	OCH3	H	OCH3	50.0	50.0	8.2	91.8
OH	H	CH3	H	H	CH3	OH	50.0	50.0	60.7	39.3
OH	H	CH3	H	H	OCH3	OH	50.0	50.0	38.6	61.4
OH	H	COOH	H	H	H	OH	50.0	50.0	45.3	54.7
OH	H	CH3	Cl, H	H	H	OH	99.8	0.2	99.6	0.4
OH	H	CH3	H	H, OH	H	OH	58.7	41.3	71.8	28.2
OH	H	CH3	H	H, OCH3	H	OH	50.9	49.1	4.9	95.1
OH	H	CH3	H	OH, OCH3	H	OH	58.8	41.2	42.6	57.4
OH	H	CH3	H	H	H, OCH3	OH	53.7	46.3	69.9	30.1

---

Homobianthrone differentiation

Among these, two enantiomorphous anthronyls of the occurring mesomers have been reported to feature the same ¹H NMR data,^20,21 which require an equal population of conformers I and II to match deduction D1. The same amount of them is further supported by quantum chemical calculations. On the other hand, the racemates have also exhibited the same resonances for two equal anthronyls,^20,21 but they do not necessitate a 1 : 1 ratio of III/IV based on deduction D1. Thus, it can be concluded that a cis diastereomer possesses the higher-frequency ¹H NMR signals for rings A and A′ and lower-frequency ones for B and B′ than its trans derivative if III is more than IV, but to the contrary if III is less (taken as deduction D2).

As references to deduce the ratios of III/IV, the Boltzmann populations of conformers III and IV for model bianthrones with CH₃, OH, OCH₃, COOH, OAc, or/and Cl groups at possible positions are shown in Table 1. The conformer III of 1 is more than IV, and the predicted ¹H NMR values for 1 match well with deduction D2 (Fig. S2 and S3†). Then, the reported ¹H NMR data for 1 may be successfully reassigned to its mesomer and racemate, respectively.^6,22–24 The mesomer and racemate of emodin bianthrone (2) have been identified by addition of an asymmetric shift reagent into their hexaacetates.²⁰ The experimental and calculated data for 2 also match each other and agree with deduction D2 regardless of a lower-frequency signal for H-4/4′ and a higher-frequency signal for H-5/5′ (in acetone-d₆) in its mesomer,^6,20,25 which probably arise from the similar populations of III and IV. When the integral equation formalism variant polarizable continuum model (IEF-PCM) is used, the calculated ¹H NMR differences between mesomer and racemate of 2 are almost reverse to the experimental ones (Fig. S4†). Thus, the calculations at the gas-phase level are reliable. As a result of our continuous work on an algicolous fungus,²⁶ the mesomer and racemate of physcion bianthrone (3)^3,22,27,28 have been separated in almost pure form for the first time, and they also display regular ¹H NMR differences. The diastereomer with lower-frequency ¹H NMR signals for rings A and A′ and higher-frequency ones for B and B′ is regarded as a mesomer due to the lower population of III. Neobulgarones A and B (4) are deduced to be racemic and meso, respectively, by their ¹H NMR differences and the higher population of III.⁵ The 8,8′-di-O-methyl derivative (5) of physcion bianthrone has the lower population of III, which suggests that its mesomer possesses lower-frequency ¹H NMR signals for rings A and A′ and higher-frequency ones for B and B′ than racemate.² Additionally, the diacetate of 5 has been differentiated into mesomer and racemate by addition of an asymmetric shift reagent.²¹ The results also agree with deduction D2 except for the slightly lower-frequency signal for CH₃-3/3′ in the mesomer, which may be influenced by the unsaturated acetyl groups. Thus, the gas-phase level is further validated. The mesomer and racemate of bianthrone A1 (6) have been elucidated by comparison of their ¹H NMR data with those of 2,^7,29 but the results are reversed according to the lower population of III. Then, bianthrone A2b (7) with no optical rotation isolated from Psorospermum glaberrimum is deduced to be a mesomer by its similar ¹H NMR data to the mesomer of 6.^7,8 Isophyscion bianthrone (9),²³ 10,⁷ and bivismiaquinone (11)³⁰ also feature the lower populations of III by comparison with the simplified model compounds, and then each of them can be divided into mesomer and racemate by deduction D2. Thus, the original configurations of 10 have been identified improperly. The mesomer of endocrocin bianthrone (12) has been established by NOE correlations, but the anti-form-based analysis is not authentic.¹ Its higher-frequency ¹H NMR signals for rings A and A′ and lower-frequency ones for B and B′ than the racemate with more IV are not in agreement with deduction D2 except for OH-1/1′ and OH-8/8′, which may be interfered by the vicinal unsaturated carboxyl groups. The chlorination at C-5/5′ results in the dominant population of III, and the regular ¹H NMR deviations are extremely represented by neobulgarones E (13b/c) and F (13a).⁵ Overall, deduction D2 is useful to distinguish between the diastereomeric mesomer and racemate of each homobianthrone, and the positions far from chiral centers and unsaturated groups are more reliable.

Prinoidin bianthrone (8) features the same chiral group at C-6/6′,⁶ which is deduced to consist of one cis (not meso) and two trans H-10/10′ diastereomers. The cis isomer can be easily identified by the different ¹H and ¹³C NMR data for two anthronyls due to their incomplete equality or enantiotropy, and the resonances for protons (OH-1/1′, H-2/2′, and CH₃-3/3′) far from chiral centers are also in accordance with deduction D2. Similarly, sennoside A with a trans H-10/10′ should comprise two diastereomers, which are supported by the broad HPLC peak.⁹ Thus, a chiral group can facilitate the identification of cis and trans homobianthrones, which further corroborates deduction D2.

Heterobianthrone differentiation

The two anthronyl moieties of these bianthrones are inclined to display nonoverlapped ¹H NMR signals due to the presence of different groups. Fortunately, most of them (14–24) feature the same A/A′ or B/B′ rings with comparable resonances.^{3–8,22,23,29} Unlike mesomers, the cis heterobianthrones exhibit different populations of I and II, which may also result in the divergence of resonances by deduction D1. Extremely, conformer I dominates the cis diastereomer of 24 by comparison with model compounds (Table 1), which should cause remarkable deviations of ¹H NMR data for rings A and A′. Thus, neobulgarone D can be identified to possess the cis H-10/10′ (24a/b).⁵ If the corresponding protons far from different groups in each diastereomer exhibit almost the same resonances, the similar populations of I and II are suggested to be present in a cis diastereomer. Then, these heterobianthrones can be easily distinguished by deduction D2 along with populations of III and IV. However, only the protons far from different groups may be considered, such as OH-1/1′, H-2/2′, and CH₃-3/3′ in 14–22. If both racemic dimers of each anthronyl favor more III or IV, the results can be referred by the corresponding heterobianthrones, such as 14, 18, 21–23. Thus, the original differentiation of 22 is unacceptable,⁷ and bianthrone A2a (18) from Psorospermum tenuifolium and (+)-crinemodin–rhodoptilometrin from Himerometra magnipinna are deduced to feature the trans and cis H-10/10′, respectively, by comparison with their analogues (6b/c and 2a).^7,10,29 On the other hand, the ratios of III/IV for naturally-occurring chrysophanol-physcion bianthrone (15), chrysophanol–isophyscion bianthrone (16), and physcion-emodin bianthrone (17) need to be confirmed by calculations (Table 1), which may be taken as valuable references for their analogues. As a result, the cis diastereomer originally assigned for 17 should be trans.³ Furthermore, a chiral group in prinoidin–emodin bianthrone (19) has resulted in four diastereomers,⁶ which can also be assigned to possess a cis (originally-labeled as B and C with lower-frequency ¹H NMR signals of OH-1/1′, H-2/2′, and CH₃-3/3′) or a trans (higher-frequency ones) relationship of H-10/10′ by deduction D2. Rhamnepalin (20) may also be divided into cis (A) and trans (B and C) diastereomers in the same way, but one of the cis isomers has not be obtained therein.⁶ In summary, the chemical shifts of protons far from different groups are suitable for the differentiation of a heterobianthrone, and the identification of a cis diastereomer requires these data to be diverged or applied in deduction D2.

Conclusions

Based on the above analyses, it can be concluded that the cis and trans 10(sp³), 10′(sp³)-bianthrones can be completely distinguished by ¹H NMR data for the positions far from different and unsaturated groups as well as chiral centers. If the relative data are diverged due to the presence of different or chiral groups, they should correspond to a cis diastereomer. Otherwise, a cis bianthrone features the higher-frequency ¹H NMR signals for rings A and A′ and lower-frequency ones for B and B′ than its trans derivative if conformer III is more than IV, and quite the contrary if III is less. The ratios of III/IV can be deduced by the sizes of groups and their distances to C-10/10′ as shown in Table 1. This rule has been validated by the calculated and experimental results and applies to both homo- and heterobianthrones. It will greatly contribute to locating the bioactivity in pharmacological assays and intermediacy in chemical and biological syntheses of photosensory pigments and photochromic agents. On the other hand, this rule may be beneficial to distinguish the other highly-conjugated dimmers, such as bifluorene derivatives,¹³ and further interprets the beauty of quantum chemical calculations in structure elucidation.³¹⁻³⁵

Experimental section

Experimental

The fungus strain Aspergillus wentii pt-1, fermentation, extraction, and preliminary isolation have been described previously.²⁶ As the continuous work, Fr. 13 eluted with petroleum ether (PE)/EtOAc (5 : 1) and was further purified by column chromatography on silica gel (PE/EtOAc, 5 : 1) and Sephadex LH-20 (CHCl₃/MeOH, 1 : 1) as well as semi-preparative HPLC (MeOH–H₂O, 85 : 15) to give 3a (3 mg) and 3b/c (1 mg) with a little in each other for the accurate comparison of ¹H NMR signals. The HSQC, HMBC, ¹H–¹H COSY, and NOESY spectra of 3a have been determined to assign the ¹H NMR data, which were further used to revise the improper ¹H NMR reassignments for some bianthrones along with coupling constants and the other correct data assigned by 2D NMR. It was noticeable that only the ¹H NMR data determined in the same deuterated solvent may be used for comparison.

Experimental ¹H NMR data for cis (meso) and trans physcion bianthrone (3): δ_cis (in CDCl₃) 11.82 (s, OH-1,1′), 6.68 (s, H-2,2′), 2.29 (s, CH₃-3,3′), 6.11 (s, H-4,4′), 6.01 (d, 2.2, H-5,5′), 3.84 (s, OCH₃-6,6′), 6.38 (d, 2.2, H-7,7′), 12.18 (s, OH-8,8′), 4.36 (s, H-10,10′); δ_trans (in CDCl₃) 11.88 (s, OH-1,1′), 6.70 (s, H-2,2′), 2.31 (s, CH₃-3,3′), 6.13 (s, H-4,4′), 5.97 (d, 2.2, H-5,5′), 3.82 (s, OCH₃-6,6′), 6.36 (d, 2.2, H-7,7′), 12.13 (s, OH-8,8′), 4.35 (s, H-10,10′).

Reassigned ¹H NMR data for cis (meso) and trans chrysophanol bianthrone (1): δ_cis (in CDCl₃) 11.72 (s, OH-1,1′), 6.73 (s, H-2,2′), 2.37 (s, CH₃-3,3′), 6.18 (s, H-4,4′), 6.36 (d, 8.0, H-5,5′), 7.36 (t, 8.0, H-6,6′), 6.88 (d, 8.0, H-7,7′), 11.80 (s, OH-8,8′), 4.49 (s, H-10,10′);²⁴ δ_cis (in acetone-d₆) 11.63 (s, OH-1,1′), 6.68 (d, 1.0, H-2,2′), 2.25 (s, CH₃-3,3′), 6.00 (d, 1.0, H-4,4′), 6.29 (dd, 1.0, 8.0, H-5,5′), 7.29 (t, 8.0, H-6,6′), 6.81 (dd, 1.0, 8.0, H-7,7′), 11.75 (s, OH-8,8′), 4.49 (s, H-10,10′);⁶ δ_trans (in CDCl₃) 11.65 (s, OH-1,1′), 6.67 (s, H-2,2′), 2.24 (s, CH₃-3,3′), 5.76 (s, H-4,4′), 6.75 (d, 7.9, H-5,5′), 7.49 (t, 7.9, H-6,6′), 6.95 (d, 7.9, H-7,7′), 11.90 (s, OH-8,8′), 4.49 (s, H-10,10′);²⁴ δ_trans (in acetone-d₆) 11.58 (s, OH-1,1′), 6.60 (d, 1.0, H-2,2′), 2.15 (s, CH₃-3,3′), 5.70 (d, 1.0, H-4,4′), 6.68 (dd, 1.0, 8.0, H-5,5′), 7.41 (t, 8.0, H-6,6′), 6.89 (dd, 1.0, 8.0, H-7,7′), 11.85 (s, OH-8,8′), 4.51 (s, H-10,10′).⁶ The signals of H-2,2′and H-4,4′ should exhibit a dq peak or s peak seemingly regardless of the effect of H-10,10′.

Reassigned ¹H NMR data for cis (meso) and trans emodin bianthrone (2): δ_cis (in CDCl₃) 11.78 (s, OH-1,1′), 6.67 (s, H-2,2′), 2.23 (s, CH₃-3,3′), 6.21 (s, H-4,4′), 6.03 (d, 2, H-5,5′), 10.71 (s, OH-6,6′), 6.18 (d, 2, H-7,7′), 11.86 (s, OH-8,8′), 4.47 (s, H-10,10′);²⁰ δ_cis (in acetone-d₆) 11.85 (s, OH-1,1′), 6.68 (s, H-2,2′), 2.27 (s, CH₃-3,3′), 6.30 (s, H-4,4′), 6.15 (d, 2.0, H-5,5′), 6.25 (d, 2.0, H-7,7′), 11.97 (s, OH-8,8′), 4.55 (s, H-10,10′);⁶ δ_trans (in CDCl₃) 11.68 (s, OH-1,1′), 6.63 (s, H-2,2′), 2.18 (s, CH₃-3,3′), 6.23 (s, H-4,4′), 6.03 (d, 2, H-5,5′), 10.73 (s, OH-6,6′), 6.23 (d, 2, H-7,7′), 11.92 (s, OH-8,8′), 4.47 (s, H-10,10′);²⁰ δ_trans (in acetone-d₆) 11.75 (s, OH-1,1′), 6.59 (s, H-2,2′), 2.20 (s, CH₃-3,3′), 6.37 (s, H-4,4′), 6.07 (brs, H-5,5′), 6.33 (d, 2.0, H-7,7′), 12.05 (s, OH-8,8′), 4.55 (s, H-10,10′).⁶

Reassigned ¹H NMR data for cis (meso) and trans 1-O-acetyl-8-O-methylphyscion bianthrone: δ_cis (in pyridine-d₅) 2.35 (s, AcO-1,1′), 6.92 (s, H-2,2′), 2.24 (s, CH₃-3,3′), 6.57 (s, H-4,4′), 6.10 (d, 2, H-5,5′), 3.74 (s, OCH₃-6,6′), 6.63 (d, 2, H-7,7′), 3.80 (s, OCH₃-8,8′), 4.36 (s, H-10,10′); δ_trans (in pyridine-d₅) 2.36 (s, AcO-1,1′), 6.92 (s, H-2,2′), 2.20 (s, CH₃-3,3′), 6.71 (s, H-4,4′), 5.97 (d, 2, H-5,5′), 3.67 (s, OCH₃-6,6′), 6.56 (d, 2, H-7,7′), 3.63 (s, OCH₃-8,8′), 4.20 (s, H-10,10′).²¹

Reassigned ¹H NMR data for cis and trans physcion-emodin bianthrone (17): δ_cis (in acetone-d₆) 11.75/11.70 (s, OH-1,1′), 6.62/6.62 (s, H-2,2′), 2.23/2.24 (s, CH₃-3,3′), 6.38/6.31 (s, H-4,4′), 6.10/6.14 (brs, H-5,5′), 9.79 (s, OH-6), 3.89 (s, OCH₃-6′), 6.31 (brs)/6.40 (d, 2.2, H-7,7′), 12.05/12.02 (s, OH-8,8′), 4.56/4.56 (s, H-10,10′); δ_trans (in acetone-d₆) 11.82/11.79 (s, OH-1,1′), 6.66/6.69 (s, H-2,2′), 2.27/2.29 (s, CH₃-3,3′), 6.34/6.21 (s, H-4,4′), 6.15/6.18 (brs, H-5,5′), 9.75 (s, OH-6), 3.84 (s, OCH₃-6′), 6.26/6.34 (d, 2.3, H-7,7′), 11.97/11.97 (s, OH-8,8′), 4.56/4.56 (s, H-10,10′).³

Computational

Only the conformers with cis H_α-10/10′ and trans H_α-10/H_β-10′ were considered due to the same ¹H NMR data for enantiomers. Conformational searches were performed via the Dreiding force field in MarvinSketch¹⁷ regardless of rotations of methyl, hydroxyl, carboxyl, and acetoxyl groups, the geometries of which were further optimized at the gas-phase B3LYP/6-31G(d) level via Gaussian 09 software¹⁸ to give conformers I–IV for each bianthrone without vibrational imaginary frequencies. Some optimized conformers were subjected to the theoretical calculations of ¹H NMR data using the gauge-independent atomic orbital (GIAO) method at the gas-phase B3LYP/6-311 + G(2d,p) level with tetramethylsilane (TMS) as a reference,¹⁹ and then the calculated values were weighted by Boltzmann distribution.

Acknowledgements

This work was financially supported by the National Natural Science Foundations of China (41106136 and 41106137), 12th Five-Year Science and Technology Plan for Agriculture (2012BAD32B09), and Chinese Academy of Sciences for Key Topics in Innovation Engineering (KZCX2-YW-QN209).

Notes and references

1. T. Awakawa, K. Yokota, N. Funa, F. Doi, N. Mori, H. Watanable and S. Horinouchi, Chem. Biol., 2009, 16, 613–623.
2. H. F. Sun, X. M. Li, L. H. Meng, C. M. Cui, S. S. Gao, C. S. Li and B. G. Wang, Helv. Chim. Acta, 2013, 96, 458–462; 1H NMR and specific optical rotation data included in the dissertation of S.H.F. were published online (http://www.cnki.net).
3. L. Du, T. Zhu, H. Liu, Y. Fang, W. Zhu and Q. Gu, J. Nat. Prod., 2008, 71, 1837–1842.
4. A. H. Aly, A. Debbab, C. Clements, R. Edrada-Ebel, B. Orlikova, M. Diederich, V. Wray, W. Lin and P. Proksch, Bioorg. Med. Chem., 2011, 19, 414–421.
5. F. Bilbert, H. Anke and O. Sterner, J. Antibiot., 2000, 53, 1123–1129.
6. L. P. Mai, F. Guéritte, V. Dumontet, M. V. Tri, B. Hill, O. Thoison, D. Guénard and T. Sévenet, J. Nat. Prod., 2001, 64, 1162–1168.
7. M. Politi, R. Sanogo, K. Ndjoko, D. Guilet, J. L. Wolfender, K. Hostettmann and I. Morelli, Phytochem. Anal., 2004, 15, 355–364.
8. B. N. Lenta, K. P. Devkota, S. Ngouela, F. F. Boyom, Q. Naz, M. I. Choudhary, E. Tsamo, P. J. Rosenthal and N. Sewald, Chem. Pharm. Bull., 2008, 56, 222–226.
9. K. Yamasaki, M. Kawaguchi, T. Tagami, Y. Sawabe and S. Takatori, J. Nat. Med., 2010, 64, 126–132.
10. N. Shao, G. Yao and L. C. Chang, J. Nat. Prod., 2007, 70, 869–871.
11. H. Falk, Angew. Chem., Int. Ed., 1999, 38, 3116–3136.
12. H. Falk and G. Schoppel, Mon. Chem., 1992, 123, 931–938.
13. P. C. Li, T. S. Wang, G. H. Lee, Y. H. Liu, Y. Wang, C. T. Chen and I. Chao, J. Org. Chem., 2002, 67, 8002–8009.
14. J. K. Gimzewski and C. Joachim, Science, 1999, 283, 1683–1688.
15. S. Lara-Avila, A. Danilov, V. Geskin, S. Bouzakraoui, S. Kubatkin, J. Cornil and T. Bjørnholm, J. Phys. Chem. C, 2010, 114, 20686–20695.
16. N. Filipescu, E. Avram and K. D. Welk, J. Org. Chem., 1977, 42, 507–512.
17. Calculator Plugins were used for structure property prediction and calculation, Marvin 5.9.2, 2012, ChemAxon (http://www.chemaxon.com).
18. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, Revision C.01, Gaussian, Inc., Wallingford CT, 2010.
19. K. W. Wiitala, T. R. Hoye and C. J. Cramer, J. Chem. Theory Comput., 2006, 2, 1085–1092.
20. D. W. Cameron, J. S. Edmonds and W. D. Raverty, Aust. J. Chem., 1976, 29, 1535–1548.
21. G. Assante, L. Camarda and G. Nasini, Gazz. Chim. Ital., 1980, 110, 629–631.
22. G. D. Monache, M. C. de Rosa, R. Scurria, B. Monacelli, G. Pasque, G. Dall′Olio and B. Bota, Phytochemistry, 1991, 30, 1849–1854.
23. G. Alemayehu, B. Abegaz, G. Snatzke and H. Duddeck, Phytochemistry, 1993, 32, 1273–1277.
24. R. N. dos Santos, M. G. de Vasconcelos Silva and R. B. Filho, Quim. Nova, 2008, 31, 1979–1981.
25. B. Botta, F. D. Monache, G. D. Monache, G. B. M. Bettolo and J. D. Msonthi, Phytochemistry, 1985, 24, 827–830.
26. R. R. Sun, F. P. Miao, J. Zhang, G. Wang, X. L. Yin and N. Y. Ji, Magn. Reson. Chem., 2013, 51, 65–68.
27. S. Kitanaka and M. Takido, Phytochemistry, 1982, 21, 2103–2106.
28. M. Bachmann, J. Lüthy and C. Schlatter, J. Agric. Food Chem., 1979, 27, 1342–1347.
29. G. D. Monache, F. D. Monache, R. di Benedetto and J. U. Oguakwa, Phytochemistry, 1987, 26, 2611–2613.
30. A. A. Hussein, B. Bozzi, M. Correa, T. L. Capson, T. A. Kursar, P. D. Coley, P. N. Solis and M. Gupta, J. Nat. Prod., 2003, 66, 858–860.
31. F. P. Miao, X. R. Liang, X. L. Yin, G. Wang and N. Y. Ji, Org. Lett., 2012, 14, 3815–3817.
32. N. Y. Ji, X. H. Liu, F. P. Miao and M. F. Qiao, Org. Lett., 2013, 15, 2327–2329.
33. X. H. Liu, F. P. Miao, M. F. Qiao, R. H. Cichewicz and N. Y. Ji, RSC Adv., 2013, 3, 588–595.
34. X. D. Li, F. P. Miao, X. R. Liang, B. G. Wang and N. Y. Ji, RSC Adv., 2013, 3, 1953–1956.
35. K. Li, C. O. Brant, M. Huertas, S. K. Hur and W. Li, Org. Lett., 2013, 5924–5927.

---

Footnote

† Electronic supplementary information (ESI) available: Additional calculated results in Fig. S1–S4 and Tables S1 and S2, NMR spectra of 3a and 3b/c, and Cartesian coordinates. See DOI: 10.1039/c3ra47055e

---