Novel serotonin transporter regulators: Natural aristolane- and nardosinane- types of sesquiterpenoids from Nardostachys chinensis Batal


Serotonin transporter (SERT) is a classic target of drug discovery for neuropsychiatric and digestive disorders, and against those disorders, plants of Nardostachys genus have been valued for centuries in the systems of Traditional Chinese Medicine, Ayurvedic and Unani. Herein, chemical investigation on the roots and rhizomes of Nardostachys chinensis Batal. led to the isolation of forty sesquiterpenoids including six new aristolane-type sesquiterpenoids and six new nardosinane-type sesquiterprenoids. Their structures were elucidated by extensive spectroscopic methods, combined with analyses of circular dichroism and single-crystal X-ray diffraction data. To explore natural product scaffolds with SERT regulating activity, a high-content assay for measurement of SERT function in vitro was conducted to evaluate the SERT regulating properties of these isolates. In conclusion, eleven compounds could be potential natural product scaffolds for developing drug candidates targeting SERT. Among which, kanshone C of aristolane-type sesquiterpenoid inhibited SERT most strongly, while desoxo-nachinol A of nardosinane-type sesquiterpenoid instead enhanced SERT potently.


Serotonin transporter (SERT) is a classic target of drug discovery for neuropsychiatric and digestive disorders. At serotonin synapses in the central nervous system, SERT is responsible for the reuptake of 5-hydroxytryptamine into presynaptic neurons, and it is implicated in the occurrence of mood disorders, for instance, depression, anxiety or obsessive-compulsive disorder1. At enterochromaffin cells in the digestive system, SERT inactivates the stimulant effects of 5-hydroxytryptamine on gastrointestinal tract mucosa, and it plays important roles in the pathophysiology of digestive disorders such as irritable bowel syndrome, slow transit constipation and functional abdominal bloating2,3. To screen SERT activity of the candidate compounds, the high-content assay for measurement of SERT function based on human embryonic kidney 293 cell line stably expressing human SERT (hSERT-HEK) and the fluorescent substrate 4-[4-(dimethylamino)phenyl]-1-methylpyridinium (APP+) has been established4,5, and this novel method is more feasible in practice than the traditional isotope labeling uptake assay.

To identify novel SERT regulators from natural products, Nardostachys chinensis Batal. (NCB) has been studied. NCB is mainly distributed in Sichuan, Gansu, Qinghai and Xizang areas in China. The root and rhizome of NCB have been used as both herbal drugs and functional foods for centuries to treat digestive disorders in traditional Chinese medicine6. Modern pharmacological studies demonstrated that NCB show bioactivities in against depression, arrhythmia, convulsion, myocardial ischemia and hypertension7. This plant was enriched with bioactive sesquiterpenoids, among which aristolane-, nardosinane-, and guaiane- types of sesquiterpenoids were the representative constituents8,9. Herein, we report the isolation, structure elucidation and effects on SERT function of six new and twelve known aristolane-type sesquiterpenoids (Fig. 1), together with six new and sixteen known nardosinane-type sesquiterpenoids (Fig. 2) from NCB.

Figure 1

Aristolane-type sesquiterpenoids from N. chinensis Batal.

Figure 2

Nardosinone-type sesquiterpenoids from N. chinensis Batal.

Results and Discussion

Structure identification

The 70% aqueous ethanol extract of the air-dried roots and rhizomes of Nardostachys chinensis Batal. was subjected to various modern chromatographic isolation (including preparative thin layer chromatography, silica gel/Sephadex LH-20 column chromatography, and reversed-phase C18 preparative/semipreparative high performance liquid chromatography) to give six new (compounds 3, 6, 7, 11, 14 and 18) and twelve known aristolane-type sesquiterpenoids (Fig. 1), together with six new (compounds 19, 2224, 26, and 30) and sixteen known nardosinane-type sesquiterpenoids (Fig. 2). Based on the comparison of spectroscopic data with those previously reported, those known compounds were identified as nardoaristolone C (1)10, nardoaristolone B (2)11, 1(10)-aristolen-9β-ol (4)12, kanshone C (5)13, kanshone H (8)14, (−)-aristolone (9)15, (−)-(14β,15β)-aristolone (10)16, kanshone F (12)8, kanshone G (13)8, debilon (15)17, nardostachone (16)18, 1(10)-aristolen-2-one (17)19, nardosinone (20)17, nardosinonediol (21)17, kanshone A (25)17, kanshone E (27)20, isonardosinone (28)20, nardosinanone H (29)9, (4 S,4aR,5 R)-4a,5-dimethyl-4-(1-methylethenyl)-4a,5,6,7-tetrahydronaphthalen-1(4 H)-one (31)21, nardosinanone I (32)9, nardosinanone B (33)22, nardosinanone G (34)9, nardosinanone M (35)10, nardonoxide (36)12, nardosinanone F (37)9, desoxo-nachinol A (38)17, narchinol B (39)23 and narchinol A (40)24, respectively. The structures of the new compounds were deduced by analysis of extensive spectroscopic data [including HRESIMS, 1D/2D NMR, optical rotation and circular dichroism (CD) data].

Nardoaristolone C (1) was isolated as colorless needles. The 1H and 13C NMR data (Supplementary Table S3) revealed that the structure of 1 was similar to that of nardoaristolone B (2), except that the two olefinic carbons (C-1 and C-9) in 2 were replaced by the two oxygenated carbons in 1 [δ C 60.1 (C-1) and δ C 69.4 (C-9), δ H 3.90 (1 H, s, H-1)], indicating the existence of an oxiran ring adjacent to carbonyl in 1. Analysis of the HSQC and HMBC spectroscopic data supported the assignments of its 1D NMR data, which were identical with those data recently reported by other groups in this April.

The relative configuration of compound 1 was revealed by single-crystal X-ray diffraction (Mo Kα) data, together with the NOESY correlations between δ H 1.97 (H-6) and δ H 1.04 (H3-13), and between δ H 1.56 (H3-12) and δ H 2.57 (H-4)/1.20 (H3-11). For X-ray crystallographic analysis of light atom structures (those containing only C, H, N or O), if a heavier element (such as chlorine, bromine or sulfur) is not contained in the crystals, the anomalous scattering with molybdenum (Mo) radiation is always too small to assign the absolute configuration25. Herein, the absolute configuration of 1 was then assigned by comparing its CD data with those of 2 (Supplementary Fig. S79), both sharing a positive Cotton effect at 215 ± 5 nm and a negative Cotton effect at 255 ± 2 nm, which were identical with the reported data10.

Nardoaristol (3) was isolated as a colorless oil, and epoxynardosinone (19) was isolated as colorless crystals. Their molecular formulas were confirmed respectively to be C15H24O2 and C15H22O4 by analysis of their NMR and ESIMS data. The 1H and 13C NMR data (Supplementary Table S3 and Table 1) demonstrated that the structures of 3 and 19 were similar to those of 1(10)-aristolen-9β-ol (4) and nardosinone (20), except that the olefinic carbons (C-1 and C-10) in 4 and 20 were replaced by the oxygenated carbons of two oxiran rings in 3 [δ C 57.4 (C-1) and δ C 66.7 (C-9), δ H 3.47 (1 H, d, J = 2.4 Hz, H-1)] and 19 [δ C 60.2 (C-1) and δ C 61.9 (C-10), δ H 3.92 (1 H, br s, H-1)], respectively. The deduction was further confirmed by analysis of the HSQC and HMBC spectroscopic data and the relative configurations of 3 and 19 were assigned by the 2D NOESY experiments (Fig. 3). Similarly, the structure of epoxynardosinanone H (30) was established by comparison of its 1D/2D NMR spectroscopic data with those of nardosinanone H (29).

Table 1 13C NMR (100 MHz, CDCl3) data for sesquiterpenoids 3, 6, 7, 14, 18–19, 22–24, 26 and 30.
Figure 3

Key HMBC and NOESY correlations for the new sesquiterpenoids from N. chinensis Batal.

3-Hydroxylkanshone H (6) was isolated as a colorless oil, and 3-oxokanshone H (7) was isolated as a white amorphous powder. Analysis of their ESIMS and NMR data established the molecular formulas to be C15H20O2 and C15H18O2. On the basis of the HSQC and HMBC spectra, the structures of 6 and 7 were elucidated as 3-hydroxylaristol-1,9-dien-8-one [δ H 4.04 (1 H, d, J = 9.6 Hz)/δ C 71.4, H-3/C-3 in 6] and aristol-1,9-dien-3,8-dione (δ C 199.6, C-3 in 7), indicating that they were 3-hydroxyl and 3-oxo derivatives of kanshone H (8), respectively. The β configuration of the 3-hydroxyl group in 6 was deduced from the key NOESY correlations (Fig. 3) between H-3 and H3-14 (δ H 1.14)/H3-15 (δ H 1.27).

Based on the above deduced relative configurations, the absolute configurations of the new compounds 6, 7, 19 and 30 were proposed by comparing their experimental CD data with those of the known aristolane-type and nardosinane-type sesquiterpenoids isolated from NCB (Supplementary Fig. S79), which normally share the same intrinsic absolute configurations for stereogenic carbons C-4, 5, 6 and 7, owing to the relatively conservative biogenic pathway of sesquiterpenoids in NCB. The structures of compounds 68 possess similar chromophores26, and then they showed similar patterns of CD curves as shown in Supplementary Fig. S79. When compared with 8, 3-hydroxyl substitution seemed to weaken the positive cotton effect around 245 nm in 6, while 3-ketone substitution was prone to reverse the shoulder peak around 300 nm to be a “valley”-style curve in 7. Assignments of the absolute configurations of 3 was fairly challenging owing to its structure without any cyclic ketone group. Considering that 3 was a 1(10)-epoxidation derivative of the known aristolane-type sesquiterpenoid (4) and its CD spectrum also showed negative Cotton effects (<215 nm), the absolute configuration of 3 was proposed as shown in Fig. 1.

Based on the comparisons of 1D/2D NMR data, 1-hydroxylaristolone (11) and 9β-debilon (14) were found out to be two new stereoisomers of the reported structures axinysone B27 and debilon (15), respectively. 1D NMR data of 11 was exactly the same as those reported for axinysone B, but the optical rotation and CD data of 11 was opposite to those of axinysone B, indicating that the two compounds are a pair of enantiomers. The experimental CD spectrum of 11 was further compared with the computational ECD spectra of the (1 S,4 R,5 R,6 S,7 R)-11a and (1 R,4 S,5 S,6 R,7 S)-11b obtained by time-dependent density functional theory (TDDFT) quantum mechanics [B3LYP/6–31 G(d)]28. The excellent agreement of the experimental and computational ECD spectra (see Supplementary Fig. S80) confirmed the assignment of the absolute configuration of 11 as (1 S,4 R,5 R,6 S,7 R). 14 was elucidated as a 9β-hydroxyl epimer of 15 (debilon) by analysis and comparison of the spectroscopic data (including 1D/2D NMR, ORD and CD data) for compounds 14 and 15.

Aristolanhydride (18) was isolated as a white amorphous powder with a molecular formula of C15H20O3, deduced by analysis of the HRESIMS (m/z 249.1469 [M + H]+, calcd for C15H21O3 +, 249.1491) and NMR data. The 1H NMR spectrum of 18 revealed the existences of four methines [δ H 6.65 (1 H, t, J = 3.9 Hz), 2.10 (1 H, m), 1.37 (1 H, d, J = 9.9 Hz) and 1.42 (1 H, d, J = 9.9 Hz)], two methylenes [δ H 2.15 (2 H, m), 1.82 (1 H, td, J = 6.7, 3.1 Hz) and 1.45 (1 H, m)] and four methyls [δ H 1.18 (3 H, s), 1.33 (3 H, s), 1.44 (3 H, s) and 0.89 (3 H, d, J = 6.9 Hz)] in 18. Furthermore, the 13C NMR spectrum of 18 suggested a double bond (δ C 141.2 and 137.4) and two ester carbonyls (δ C 176.1 and 171.8) were included in the structure of 18. HMBC correlations from δ H 6.65 (H-1) to δ C 171.8 (C-10), 141.2 (C-11), 41.4 (C-5) and 24.1 (C-2), from δ H 0.89 (H3-16) to δ C 41.4, 37.4 (C-4) and 26.5 (C-3), from δ H 1.44/1.18 (H3-13/14) to δ C 46.4 (C-6) and 30.3 (C-12), from δ H 1.37 (H-6) to δ C 176.1 (C-8) and from δ H 1.33 (H3-15) to δ C 46.4 then established the planar structure of 18, a rare 8,9-cleavage anhydride derivative of aristolane-type sesquiterpenoid. The stereochemistry problem of 18 was solved by NOESY experiment and comparison of its experimental ECD spectrum with the computational ones (see Supplementary Fig. S81).

The structures of other four new nardosinane-type sesquiterpenoids, nardosinonetriol (22), 7-oxonardosinone (23), 7-oxonardosinoperoxide (24) and 2-oxokanshone A (26), were established by comparing their NMR data (Supplementary Tables S3S4 and Table 1) with those of known compounds 21 (nardosinonediol) and 25 (kanshone A). Compound 24 was deduced as a peroxide compound on the basis of its positive HRESIMS (m/z 267.1589 [M + H]+, cald for C15H23O4 +, 267.1591) and NMR data. Furthermore, the absolute configurations of these compounds were all proposed as shown in Fig. 2 based on the consideration of conservative biogenic pathway for nardosinane-type sesquiterpenoids, assisted by 2D NOESY experiments as shown in Fig. 3. The plausible biosynthetic pathways for aristolane- and nardosinane- types of sesquiterpenoids were proposed as shown in Supplementary Figures S82S83.

SERT regulating activities

As shown in Table 2, compounds 2, 4, 68, 11, 16, 19, 2324, 2729, 3233, 36, 38 and 40 enhanced SERT activity while compounds 5, 1213, 17, 2021, 30, 35 and 37 inhibited SERT activity. Compounds 1, 9, 15, 18, 22, 25, 26, 31 and 34 did not show any SERT activity meanwhile compounds 3, 10, 14 and 39 were not tested due to insufficient amount. For the SERT enhancers, nardoaristolone B (2), nardonoxide (36) and desoxo-nachinol A (38) showed potent effects, among which a 4,11-O-briged nardosinane-type sesquiterpenoid (nardonoxide, 36) with a 5/6/6 tricyclic ring system showed the strongest effect; 1(10)-aristolen-9β-ol (4), kanshone H (8), nardostachone (16), 7-oxonardosinoperoxide (24), kanshone E (27) and nardosinanone H (29) were in the middle; and 3-hydroxylkanshone H (6), 3-oxokanshone H (7), 1-hydroxylaristolone (11), epoxynardosinone (19), 7-oxonardosinone (23), isonardosinone (28), nardosinanone I (32) and nardosinanone B (33) showed weak effects, however, 8 exhibited stronger effect than its 3-hydroxyl derivative (6) and 3-oxo derivative (7), 23 exhibited weaker effect than its 9-peroxide derivative (24), and 28 showed weaker effect than its 2-oxo derivative (27). For the SERT inhibitors, kanshone C (5), nardosinone (20) and nardosinonediol (21) significantly inhibited SERT activity. Further analysis suggested that 1(10) or 1(9)-epoxidation of the double bond (as shown in cases of 1/2, 19/20 and 30/31) seems to inactivate or reverse the aristolane-type and nardosinane-type sesquiterpenoids’ regulation effects on SERT activity.

Table 2 Effects of the compounds identified from NCB on SERT activity.

Both enhancer and inhibitor of SERT, interestingly, were identified in the two main types of sesquiterpenoids (subdivided into five subtypes I, II, III, IV and V) from NCB, and the SERT enhancers are richer in quantity than the SERT inhibitors. According to the proposed biosynthetic pathways (see Supplementary Figures S82S83), it suggests that there exist conversions between SERT enhancers and SERT inhibitors in NCB. For instance, the SERT inhibitor kanshone C (5), and the SERT enhancers 1(10)-aristolen-9β-ol (4) and nardostachone (16), are derived from compound 4.

Reported or marketed clinical antidepressants of the SERT mechanism, as known, include SERT inhibitors as well as SERT enhancers. SERT inhibitors (SIs) are dominant containing tricyclic antidepressants, selective serotonin reuptake inhibitors and selective noradrenaline reuptake inhibitors29, while the selective enhancer of SERT is rare and the only one in clinical use is tianeptine30. So far, hundreds of known SIs31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52, including 3-(aminoalkyl)-5-fluoroindole53, have been developed by chemical modification or hybridization of the known clinical antidepressants, based on the molecular templates shown in Fig. 4. Although the molecular structures of known SIs are diverse, they usually contain heterochains and/or heterocycles of N, O, and/or S, with halogen-substituted phenyls (F or Cl) as the common substituents. In spite of this, few SIs of natural products have been discovered, the only example, cyclo(L-Phe-L-Phe) found in chicken essence, is a dual inhibitor of SERT and acetylcholinesterase54. As far as we know, this is the first report on SERT activities of the natural aristolane-type and nardosinane-type sesquiterpenoids. Further, according to the above structure-activity discussion of known SIs, it suggests that eleven compounds (Fig. 5) including eight SERT enhancers (nardonoxide, nardoaristolone B, desoxo-nachinol A, 1(10)-aristolen-9β-ol, kanshone E, nardosinanone H, kanshone H, nardostachone, S1S8) and three SERT inhibitors (kanshone C, nardosinonediol, nardosinone, S9S11) may provide novel potential scaffolds for synthesis of SERT regulators.

Figure 4

Molecular scaffolds of reported or marketed antidepressants targeting SERT.

Figure 5

Potential natural sesquiterpenoid scaffolds targeting SERT from N. chinensis Batal.

In conclusion, forty sesquiterpenoids were isolated from roots and rhizomes of Nardostachys chinensis Batal., and their structures were identified by combined modern spectroscopic methods. Among these compounds, eleven natural scaffolds bidirectionally regulate SERT activity. They are potential lead compounds for regulation of SERT activity in drug discovery and provide novel molecular templates for synthesis of SERT enhancers and inhibitors, especially enhancer of SERT, which is rare so far in drug discovery. Presently, we are exploiting more SERT regulation lead structures from NCB and investigating antidepressant effect of SERT regulators in vivo assays to be reported in due time.


General experimental procedures

Optical rotations were measured using a Rudolph AUTOPOL V polarimeter (Rudolph Research Analytical, Hackettstown, USA). UV and electronic circular dichroism spectra were obtained on a Jasco J-815 circular dichroism spectropolarimeter (JASCO Corporation, Tokyo, Japan). ESIMS was performed on a Waters Synapt G2 mass spectrometer (Waters, Milford, MA, USA), and HRESIMS spectra was obtained on a quadrupole time-of-flight mass spectrometer QSTARTM Elite system (Applied Biosystems/MDS Sciex, Foster, CA, USA; Concord, ON, Canada). 1D and 2D NMR spectra were recorded on a Bruker AV-III spectrometer (1H/13C, 400 MHz/100 MHz, 600 MHz/150 MHz, Bruker, Zurich, Switzerland) using TMS as an internal standard. Chemical shifts (δ) were expressed in ppm. X-ray crystal data were analyzed on Rigaku MicroMax-007HF (Rigaku Corporation, Japan), preformed with Mo Kα. Preparative HPLC was performed on Waters 2489 HPLC system (Waters, Milford, MA) using an Agilent Zorbax SB-C18 ODS column (21.2 mm × 250 mm, 7 μm). Column chromatography (CC) was carried out with silica gel (200–300/400–500 mesh, Qingdao Marine Chemical, Inc., Qingdao, China), sephadex LH-20 (GE Healthcare UK Ltd, Buckinghamshire, England) and D101 macroporous resin (Tianjin Chemical Co., Ltd., Tianjin, China). Column fractions were monitored by TLC (silica gel 60 GF254, 15 µm, Merck, Darmstadt, Germany), and the spots were visualized by heating the plates after spraying with 10% H2SO4 in ethanol. All reagents of HPLC or analytical grade were purchased from Tianjin Damao Reagent Co., Ltd., Tianjin, China.

Plant material

The air-dried plant materials were purchased from Anhui Jiren Pharmacy Co. Ltd. (Bozhou, China) in July, 2011, and were identified as roots and rhizomes of Nardostachys Chinensis Batal. by Prof. Tian-xiang Li, Tianjin University of Traditional Chinese Medicine. A voucher specimen (No. B20604126) was deposited in the Tianjin Key Laboratory of Modern Chinese Medicine, Institute of Traditional Chinese Medicine, Tianjin University of Traditional Chinese Medicine, Tianjin, China.

Extraction and isolation

The air-dried roots and rhizomes (20.0 kg) of NCB were first percolated 3 times by 8 times amount of 70% ethanol (v/v) at a speed of 10 ml/min under room temperature, and the residues were then reflux extracted 3 times (2 hours each time) by 8 times amount of 70% ethanol at 80–90 °C. The combined 70% ethanol solution was concentrated to dryness in vacuo (3.4 kg) and resuspended in water before being partitioned successively with petroleum ether, ethyl acetate, and n-butanol to give petroleum ether (PE), ethyl acetate (EA), and n-butanol (BU) extracts. The BU extract, combined with the extract afforded by alcohol precipitation of the rest water solution, was chromatographed on D101 macroporous resin column, gradiently eluted with EtOH/H2O (0:100−95:5) to obtain 5 fractions (H2O, 1000 g; 30%, 190 g; 50%, 175 g; 70%, 140 g; 95%, 70 g). Among them, the 95% EtOH fraction was concentrated to dryness in vacuo before combined with the EA extract. The PE extract (320 g) was fractionated by silica gel CC, gradiently eluted with PE/EA (100:0–0:100) to obtain 22 primary fractions Fr.1–Fr.22. Fr.5 (120.0 g) was successively subjected to a sephadex LH-20 CC (CH2Cl2/MeOH, 1:1) and a silica gel CC (PE/EA, 90:10) to afford 6 subfractions Fr.5–1–5–6, among which, Fr.5–1 gave compound 30 (4 mg, Rf = 0.3, CH2Cl2:MeOH = 1:1) and Fr.5–5 gave compound 29 (8 mg, Rf = 0.3, PE:EA:MeOH = 7:2:1) after purifications by preparative thin layer chromatography (pTLC) over silica gel. Fr. 6 (57.0 g) and Fr. 9 (40 g) were then subjected to repeated silica gel CC (PE/EA, 100:0 to 70:30) to afford compounds 2 (13 mg), 4 (280 mg), 5 (10 mg), 16 (250 mg), 20 (240 mg), 26 (3 mg) and 36 (4 mg). One subfraction Fr.6–4–5 from Fr.6 was further subjected to semi-preparative HPLC (MeOH:H2O = 80:20) to give compound 13 (t R = 7.0 min, 4.7 mg). Fr.7 (70 g) was separated by reverse phase ODS CC (MeOH/H2O, 65:35 to 100:0) to give 6 subfractions Fr.7–1–7–6. Fr.7–1 was then subjected to repeated silica gel CC (PE/EA, 100:0 to 70:30) to give compounds 1 (8 mg), 19 (6 mg) and 32 (8 mg). Similarly, Fr.7–2–7–4 were subjected to repeated silica gel CC (PE/EA, 100:0 to 70:30) to give compounds 23 (5 mg), 24 (6 mg) and 27 (3 mg). The remain subfractions from Fr.7–2–7–4 were further purified by silica gel pTLC eluted with PE/EA/methanol (9:0.5:0.5 for 3, 7, and 15, 7:2:1 for 6, 8:1.5:0.5 for 17, 5:3:2 for 38, and 2:1:2 for 35 and 40) to afford compounds 3 (2 mg, Rf = 0.2), 6 (5 mg, Rf = 0.3), 7 (6.5 mg, Rf = 0.5), 15 (9 mg, Rf = 0.5), 17 (5 mg, Rf = 0.6), 38 (8 mg, Rf = 0.5), 35 (4 mg, Rf = 0.2) and 40 (3 mg, Rf = 0.3). Compounds 8 (15 mg) and 9 (12 mg) were obtained by recrystallization of the subfraction Fr.7–2–2–1, and compound 10 (3 mg) was given by recrystallization of the subfraction Fr.7–2–2–9. Fr.10 (35 g) was successively subjected to a sephadex LH-20 CC (CH2Cl2/MeOH, 1:1) and a silica gel CC (PE/EA, 70:30) to afford 28 (7 mg). Fr.13 (54 g) was fractionated by silica gel CC (CH2Cl2/MeOH, 90:10) to subfrations Fr.13–1–Fr.13–6. Fr.13–2 was then subjected to silica gel pTLC to give compound 25 (4 mg, Rf = 0.3, PE:EA:MeOH = 7:2:1). Fr. 15 (7.0 g) was subjected to silica gel CC (PE/EA, 100:0 to 0:100) to obtain 18 primary fractions Fr.15–1–Fr.15–18. Subfraction Fr.15–1 was then subjected to semi-preparative HPLC (MeOH:H2O = 60:40) to afford compound 37 (t R = 10.0 min, 2.5 mg). Fr. 17 (3.0 g) was then subjected to silica gel CC (PE/EA, 100:0 to 0:100) to obtain 16 sufractions Fr.17–1–Fr.17–16. Subfraction Fr.17–9 was further isolated by semi-preparative HPLC (MeOH:H2O = 65:35) to yield compound 34 (t R = 5.0 min, 4.2 mg).

The EA extract (1200 g) was fractionated by silica gel CC, gradiently eluted with CH2Cl2/MeOH (100:0–0:100) to obtain 15 primary fractions Fr.1–Fr.15. Fr.10 (40 g) was subjected to silica gel CC eluted with 100% EA to get 8 subfratcions Fr.10–1–Fr.10–8. And then, subfraction Fr.10–3 was successively subjected to a sephadex LH-20 CC (100% MeOH) and a silica gel CC (CH2Cl2/MeOH, 100:0 to 0:100) to give compound 21 (17.5 mg). Fr.1 (782 g) was separated into 12 subfractions Fr.1–1–Fr.1–12 by normal-phase silica gel CC (PE/EA, 100:0 to 0:100). Subfraction Fr.1–1 (25 g) was then subjected to repeated silica gel CC to give 15 fractions Fr.1–1–1–Fr.1–1–15. Fraction Fr.1–1–5 was fractionated to 6 subfractions Fr.1–1–5–1–Fr.1–1–5–5 by preparative HPLC (MeOH/H2O, 70:30 to 95:5, 70.0 min), and then subfraction Fr.1–1–5–1 was further purified by preparative HPLC (MeCN:H2O = 35:65) to yield compound 31 (t R = 23.0 min, 11.0 mg). Similarly, subfraction Fr.4–2–3–2 afforded by repeated silica gel CC of Fr.4, was subjected to semi-preparative HPLC (MeOH:H2O = 65:35) to yield compound 12 (t R = 56.3 min, 10 mg). Fr.1–5 (16 g) was isolated by preparative HPLC (MeOH/H2O, 30:70 to 95:5, 80.0 min) to afford 6 fractions Fr.1–5–1–Fr.1–5–6, among which fracion Fr.1–5–5 was purified to yield compound 18 (t R = 37.5 min, 15.6 mg). Fr.1–9 (18 g) was separated into 10 fractions Fr.1–9–1–Fr.1–9–10 by preparative HPLC (MeOH/H2O, 30:70 to 95:5, 80.0 min), and fraction Fr.1–9–6 was further isolated by preparative HPLC (MeCN:H2O = 45:55) to yield compound 11 (t R = 24.5 min, 11.3 mg). Subfraction Fr.1–2 (80 g) was chromatographed on silica gel CC eluted with PE/EA (from 50:1 to 0:100) to obtain 11 fractions Fr.1–2–1–Fr.1–2–11. Fraction Fr.1–2–11 was then purified to yield compound 14 (t R = 51.0 min, 10.6 mg) by preparative HPLC (MeOH/H2O, 30:70 to 95:5, 65 min).

The 30% aqueous ethanol fraction (190 g) from the BU extract (1800 g) was then subjected to D101 macroporous resin column, gradiently eluted with MeOH/H2O (10:90–100:0) to obtain 5 subfractions (H2O, 39 g; 10%, 20 g; 20%, 17 g; 30%, 21 g; 100%, 70 g). The 10% MeOH subfraction was subjected to repeated sephadex LH-20 CC (50% aqueous MeOH or 100% MeOH) to afford compounds 22 (11.8 mg) and 33 (11.0 mg), and the rest samples were further isolated by preparative HPLC (MeOH:H2O = 28:72) to afford compound 39 (t R = 16.1 min, 43.1 mg).

Chemical structure data

The NMR spectra of the new compounds are provided in the Supplementary Information.

Nardoaristol (3)

Colorless oil (EtOAc); [α]20 D =  +18.00 (c 0.1, MeOH); UV (MeOH) λ max 197.5; CD (c 0.05, MeOH) λε) 209.5 (−0.03), 215.5 (+0.012), 219 (−0.005), 221.5 (+0.02); 1H NMR (CDCl3, 400.13 MHz) and 13C NMR data (CDCl3, 100.61 MHz), see Supplementary Table S1 and Table 1; (−)-ESIMS m/z 235.19 [M−H], ( + )-ESIMS m/z 237.27 [M + H]+.

3-Hydroxylkanshone H (6)

Colorless oil (EtOAc); [α]20 D = −50.13 (c 0.5, MeOH); UV (MeOH) λ max 283.0; CD (c 0.05, MeOH) λε) 213.5 (+1.32), 275.5 (+0.08), 296.0 (+0.66), 336.5 (−3.55); 1H NMR (CDCl3, 400.13 MHz) and 13C NMR data (CDCl3, 100.61 MHz), see Supplementary Table S1 and Table 1; (−)-HRESIMS m/z 231.1385 [M−H] (calcd for C15H19O2 , 231.1391); (−)-ESIMS m/z 231.30 [M−H], (+)-ESIMS m/z 255.28 [M + Na]+.

3-Oxokanshone H (7)

White amorphous powder (CH2Cl2); [α]20 D = −391.88 (c 0.32, MeOH); UV (MeOH) λ max 198, 290; CD (c 0.05, MeOH) λε) 197 (−1.11), 221 (9.40), 307 (−3.12), 348 (−2.24); 1H NMR (CDCl3, 400.13 MHz) and 13C NMR data (CDCl3, 100.61 MHz), see Supplementary Table S2 and Table 1; (+)-HRESIMS m/z 231.1378 [M + H]+ (calcd for C15H19O2 +, 231.1380); (−)-ESIMS m/z 229.97 [M−H], (+)-ESIMS m/z 231.33 [M + H]+.

1-Hydroxylaristolone (11)

Colorless needles (MeOH); [α]20 D = −186.31 (c 0.52, CHCl3); UV (MeOH) λ max 195, 230; CD (c 0.06, MeOH) λε) 201 (+3.20), 257 (−4.74), 294 (−0.71), 328 (−2.23); (−)-HRESIMS m/z 233.1544 [M−H] (calcd for C15H21O2 , 233.1542)., (+)-HRESIMS m/z 235.1670 [M + H]+ (calcd for C15H23O2 +, 235.1693).

9β-Debilon (14)

Colorless crystals (CH2Cl2); [α]20 D = −33.31 (c 0.73, MeOH); UV (MeOH) λ max 198, 290; CD (c 0.1, MeOH) λε) 200 (−1.85), 218.9 (−0.09), 243.2 (−0.83), 260.8 (−0.45), 318.5 (−0.51), 379.4 (+0.07); 1H NMR (CDCl3, 400.13 MHz) and 13C NMR data (CDCl3, 100.62 MHz), Supplementary Table S2 and Table 1; (−)-HRESIMS m/z 251.1646 [M + OH] (calcd for C15H23O3 , 251.1647). (+)-HRESIMS m/z 257.1498 [M + Na]+ (calcd for C15H22O2Na+, 257.1512).

Aristolanhydride (18)

White amorphous powder (CH2Cl2); [α]20 D = −67.00 (c 0.32, MeOH); UV (MeOH) λ max 210, 285; CD (c 0.07, MeOH) λε) 200 (−1.44), 209 (−0.62), 224 (−2.15); 1H NMR (CD3OD, 400.13 MHz) and 13C NMR data (CD3OD, 100.62 MHz), see Supplementary Table S2 and Table 1; (+)-HRESIMS m/z 249.1469 [M + H]+ (calcd for C15H21O3 +, 249.1491); (−)-HRESIMS m/z 265.1441 [M + OH] (calcd for C15H21O4 , 265.1449).

Epoxynardosinone (19)

Colorless crystals (EtOAc); [α]20 D = +1.27 (c 1.0, MeOH); UV (MeOH) λ max 197.0; 1H NMR (CDCl3, 400.13 MHz) and 13C NMR data (CDCl3, 100.61 MHz), see Supplementary Table S3 and Table 1; (−)-HRESIMS m/z 265.1453 [M - H] (calcd for C15H21O4 , 265.1445); (−)-ESIMS m/z 265.25 [M−H], ( + )-ESIMS m/z 267.33 [M + H]+.

Nardosinonetriol (22)

Colorless needles (MeOH); [α]20 D =  +58.89 (c 0.3, MeOH); UV (MeOH) λ max 193, 243; CD (c 0.05, MeOH) λ(Δε): 208 (−1.57), 241 (+1.76), 284 (−2.29), 303 (−2.18), 335 (−2.27); 1H NMR (DMSO-d 6, 600.23 MHz) and 13C NMR data (DMSO-d 6, 150.94 MHz), see Supplementary Table S3 and Table 1; (−)-HRESIMS m/z 313.1649 [M + COOH] (calcd for C16H25O6 , 313.1657), (+)-HRESIMS m/z 291.1569 [M + Na]+ (cald for C15H24O4Na+, 291.1567).

7-Oxonardosinone (23)

Colorless oil (EtOAc); [α]20 D = +4.67 (c 0.1, MeOH); UV (MeOH) λ max 198.0, 295.5; CD (c 0.05, MeOH) λ(Δε): 203.0 (−0.67), 209 (−0.78), 225.0 (−0.14), 245.0 (−0.39), 293.5 (+1.57), 329.0 (−1.51); 1H NMR (CDCl3, 600.25 MHz) and 13C NMR data (CDCl3, 150.95 MHz), see Supplementary Table S3 and Table 1; (+)-HRESIMS m/z 251.1637 [M + H]+ (calcd for C15H23O3 +, 251.1642); (−)-ESI-MS m/z 249.25 [M−H], (+)-ESI-MS: m/z 251.37 [M + H]+.

7-Oxonardosinoperoxide (24)

Colorless oil (EtOAc); [α]20 D = +55.33 (c 0.3, MeOH); UV (MeOH) λ max 197.5, 269.5; CD (c 0.05, MeOH) λ(Δε): 279.5 (+1.13), 326.5 (−0.91); 1H NMR (DMSO-d 6, 400.13 MHz) and 13C NMR data (DMSO-d 6, 100.61 MHz), see Supplementary Table S3 and Table 1; ( + )-HRESIMS m/z 267.1589 [M + H]+ (calcd for C15H23O4 +, 267.1591); (−)-ESI-MS m/z 265.19 [M−H], (+)-ESI-MS m/z 267.36 [M + H]+.

2-Oxokanshone A (26)

Colorless oil (EtOAc); [α]20 D = −153.85 (c 0.13, MeOH); UV (MeOH) λ max 205, 266; CD (c 0.05, MeOH) λ(Δε): 205 (+2.16), 221 (+4.40), 295 (−3.05), 334 (−0.25); 1H NMR (CDCl3, 400.13 MHz) and 13C NMR data (CDCl3, 100.61 MHz), see Supplementary Table S4 and Table 1; (+)-HRESIMS m/z 249.1485 [M + H]+ (calcd for C15H21O3 +, 249.1485).

Epoxynardosinanone H (30)

Colorless oil (EtOAc); [α]20 D = −160.56 (c 0.24, MeOH); UV (MeOH) λ max 198, 231; CD (c 0.05, MeOH) λ(Δε): 216 (+7.24), 252 (−8.05), 312 (+3.92); 1H NMR (CDCl3, 400.13 MHz) and 13C NMR data (CDCl3, 100.61 MHz), see Supplementary Table S4 and Table 1; (+)-HRESIMS m/z 233.1535 [M + H]+ (calcd for C15H21O2 +, 233.1536); (+)-ESI-MS: m/z 233.47 [M + H]+.

X-ray crystallographic analysis of 1

The structure were solved by SHELXS-97 and refined by full-matrix least-squares techniques using the SHELXL-97 program. Crystallographic data for the structure of 1 reported in this study has been deposited with the Cambridge Crystallographic Data Centre under the reference number CCDC 1058758. Copies of the data can be obtained, free of charge, on application to the Director, CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: +44-(0)1223–336033 or e-mail:

Crystal data for 1

Colorless prism, C14H18O3, M = 234.28, 0.20 × 0.18 × 0.12 mm3, monoclinic, space group P21; a = 9.6605(19) Å, b = 6.9639(14) Å, c = 9.7024(19) Å, α = γ = 90°, β = 108.44(3)°, V = 619.2(2) Å3, Z = 2, D calcd = 1.257 g/cm3, F 000 = 252. Mo Kα radiation, λ = 0.71073 Å, T = 113(2) K, μ (Mo Kα) = 0.087 mm−1. The final R 1 was 0.0443 and wR 2 was 0.0852.

Computational methods

Relative configurations of 11 and 18 were deduced by analysis of their 1D/2D NMR data. Stochastic conformational searches were firstly conducted under MMFF94 force field to give one and three possible conformers for 11 and 18, respectively. Conformers were then optimized and the frequencies were calculated by further time-dependent density functional theory (TDDFT) method to verify their stability. Using the conformers at the B3LYP/6–31 G(d) basis set level in methanol, sixty excitation states at the B3LYP/6–31 G(d) basis set level were calculated, and finally the calculation results were Boltzmann averaged to yield the depicted electronic circular dichroism (ECD) spectra of 11 and 18 with half bandwidths of 0.45 eV and 0.40 eV, respectively. All calculations were performed by Gaussian 09 program package (Version C.01).

SERT activity assay

The SERT activity assay was based on the reported method4,5 with modifications. To evaluate SERT activity of these compounds, a validated stably transfected hSERT-HEK293 cell line were used in the high content assay. The fluorescent substrate 4-[4-(Dimethylamino)phenyl]-1-methyl pyridinium (APP+) was used to examine SERT activity and the fluorescent dye Hoechst 33342 stain cellular nuclei. The effects of test compounds on SERT function were calculated by the following equation: Relative fluorescent intensity (RFI) = (Intracellular APP+ fluorescent intensity treatment/Intracellular APP+ fluorescent intensity control). The positive control drugs in testing SERT function include SSRI fluoxetine 2.0 μM and SSRE tianeptine 1.0 μM. All test compounds were run in triplicates with three repeat times. The data from the SERT function assays were analyzed by using SPSS software (Version 11.5, IBM Company). The RFI values under different treatments were evaluated by one-way analysis of variance (ANOVA), followed by post hoc testing using Dunnett’s multiple comparisons tests. The results were expressed in Mean ± S.E.M.55. Some available compounds were re-purified to make sure their purities ≥95% (HPLC, wavelengths = 230, 254, 280, 360 nm) before bioactivity evaluation.


  1. 1.

    Lin, S. H. et al. Serotonin and Mental Disorders: A Concise Review on Molecular Neuroimaging Evidence. Clin. Psychopharmacol. Neurosci. 12, 196–202 (2014).

  2. 2.

    Sikander, A. et al. Role of Serotonin in Gastrointestinal Motility and Irritable Bowel Syndrome. Clin. Chim. Acta. 403, 47–55 (2009).

  3. 3.

    Chen, J. X. et al. Guinea Pig 5-HT Transporter: Cloning, Expression, Distribution, and Function in Intestinal Sensory Reception. Am. J. Physiol: Gastrointest Liver Physiol. 275, 433–448 (1998).

  4. 4.

    Fowler, A. et al. A Nonradioactive High-throughput/High-content Assay for Measurement of the Human Serotonin Reuptake Transporter Function in vitro. J. Biomol. Screen 11, 1027–1034 (2006).

  5. 5.

    Solis, E. J. et al. 4-(4-(Dimethylamino)phenyl)-1-methylpyridinium (APP+) is a Fluorescent Substrate for the Human Serotonin Transporter. J. Biol. Chem. 287, 8852–8863 (2012).

  6. 6.

    Wu, M. H. et al. An Ethnobotanical Survey of Medicinal Spices Used in Chinese Hotpot. Food Res. Int. 48, 226–232 (2012).

  7. 7.

    Wu, J. et al. The Medicinal Textual Research of “Gansong”. J. Chin. Med. Mat. 34, 1459–1461 (2011).

  8. 8.

    Tanitsu, M. et al. Guaiane- and Aristolane-type Sesquiterpenoids of Nardostachys chinensis Roots. Phytochemistry 59, 845–849 (2002).

  9. 9.

    Zhang, J. B. et al. Nardosinane-type Sesquiterpenoids of Nardostachys chinensis Batal. Fitoterapia 100, 195–200 (2015).

  10. 10.

    Shen, X. Y. et al. Six new sesquiterpenoids from Nardostachys chinensis Batal. Fitoterapia 119, 75–82 (2017).

  11. 11.

    Liu, M. L. et al. Nardoaristolones A and B, Two Terpenoids with Unusual Skeletons from Nardostachys chinensis Batal. Org. Lett. 15, 1000–1003 (2013).

  12. 12.

    Shide, L. et al. Gansongon, a New Aristolane Ketone from Nardostachys chinesis Batalin and Structure Revision of an Aristolenol. Planta Medica 53, 556–558 (1987).

  13. 13.

    Bagchi, A. et al. Kanshone C, a Sesquiterpenoid of Nardostachys chinensis Roots. Phytochemistry 27, 2877–2879 (1988).

  14. 14.

    Liu, M. L. et al. Novel Sesquiterpenes from Nardostachys chinensis Batal. Tetrahedron 69, 6574–6578 (2013).

  15. 15.

    Su, H. et al. Sesquiterpenes from Laurencia similis. Molecules 14, 1889–1897 (2009).

  16. 16.

    Furusawa, M. et al. Biotransformation of Aristolane- and 2,3-Secoaromadendrane-Type Sesquiterpenoids Having a 1,1-Dimethylcyclopropane Ring by Chlorella fusca var. vacuolata, Mucor Species, and Aspergillus niger. Chem. Pharm. Bull. 54, 861–868 (2006).

  17. 17.

    Itokawa, H. et al. Cytotoxic Sesquiterpenes from Nardostachys chinensis. Chem. Pharm. Bull. 41, 1183–1184 (1993).

  18. 18.

    Sastry, S. D. et al. Terpenoids—CVII: The Structure of Nardostachone. Tetrahedron 23, 2491–2493 (1967).

  19. 19.

    Dellar, J. E. et al. Antimicrobial Sesquiterpenes from Prostanthera aff. melissifolia and P. rotundifolia. Phytochemistry 36, 957–960 (1994).

  20. 20.

    Bagchi, A., Oshima., Y. & Hikino, H. Kanshones D and E, Sesquiterpenoids of Nardostachys chinensis Roots. Phytochemistry 27, 3667–3669 (1988).

  21. 21.

    Bagchi, A., Oshima., Y. & Hikino, H. Kanshones A and B Sesquiterpenes of Nardostachys chinensis. Phytochemistry 27, 1199–1201 (1988).

  22. 22.

    Zhang, J. B. et al. Novel Nardosinane Type Sesquiterpenoids from Nardostachys chinensis Batal. Tetrahedron 70, 4507–4511 (2014).

  23. 23.

    Hwang, J. S. et al. Inhibitory Constituents of Nardostachys chinensis on Nitric Oxide Production in RAW 264.7 Macrophages. Bioorg. Med. Chem. Lett. 22, 706–708 (2012).

  24. 24.

    Hiroshi, H. et al. Structure and Absolute Configuration of Narchinol A. Phytochemistry 11, 2097–2099 (1972).

  25. 25.

    Thompson, A. L. & Watkin, D. J. X-ray Crystallography and Chirality: Understanding the Limitations. Tetrahedron: Asymmetry 20, 712–717 (2009).

  26. 26.

    Berova, N., Di Bari, L. & Pescitelli, G. Application of Electronic Circular Dichroism in Configurational and Conformational Analysis of Organic Compounds. Chem. Soc. Rev. 36, 914–931 (2007).

  27. 27.

    Zubia, E., Ortega, M. J. & Carballo, J. L. Sesquiterpenes from the Sponge Axinyssa isabela. J. Nat. Prod. 71, 2004–2010 (2008).

  28. 28.

    Pescitelli, G. & Bruhn, T. Good Computational Practice in the Assignment of Absolute Configurations by TDDFT Calculations of ECD Spectra. Chirality 28, 466–474 (2016).

  29. 29.

    Tran, P. V. et al. Dual Monoamine Modulation for Improved Treatment of Major Depressive Disorder. J. Clin. Psychopharmcol. 23, 78–86 (2003).

  30. 30.

    Cooper, C. M. et al. Tianeptine in an Experimental Medicine Model of Antidepressant Action. J. Psychopharmcol. 29, 582–590 (2015).

  31. 31.

    Santra, S. et al. Development of Potent Dopamine-Norepinephrine Uptake Inhibitors (DNRIs) Based on a (2S,4R,5R)-2-Benzhydryl-5-((4-methoxybenzyl)amino)tetrahydro-2H-pyran-4-ol Molecular Template. Bioorg. Med. Chem. 23, 821–828 (2015).

  32. 32.

    Wu, Y. J. et al. Discovery of a Cyclopentylamine as an Orally Active Dual NK1 Receptor Antagonist-Serotonin Reuptake Transporter Inhibitor. Bioorg. Med. Chem. Lett. 24, 1611–1614 (2014).

  33. 33.

    Sharma, H. et al. Flexible and Biomimetic Analogs of Triple Uptake Inhibitor 4-((((3S,6S)-6-Benzhydryltetrahydro-2H-pyran-3-yl)amino)methyl)phenol: Synthesis, Biological Characterization, and Development of a Pharmacophore Model. Bioorg. Med. Chem. 22, 311–324 (2014).

  34. 34.

    Liu, S. et al. Design and Synthesis of 4-Heteroaryl 1,2,3,4-tetrahydroisoquinolines as Triple Reuptake Inhibitors. ACS Med. Chem. Lett. 5, 760–765 (2014).

  35. 35.

    Elias, O. et al. Design of Novel Multiple-Acting Ligands Towards SERT and 5-HT2C Receptors. Bioorg. Med. Chem. Lett. 24, 2118–2122 (2014).

  36. 36.

    Van Orden, L. J. et al. A Novel Class of 3-(Phenoxy-phenyl-methyl)-pyrrolidines as Potent and Balanced Norepinephrine and Serotonin Reuptake Inhibitors: Synthesis and Structure-Activity Relationships. Bioorg. Med. Chem. Lett. 23, 1456–1461 (2013).

  37. 37.

    Dreyfus, N. et al. Discovery of a Potent, Dual Serotonin and Norepinephrine Reuptake Inhibitor. ACS Med. Chem. Lett. 4, 560–564 (2013).

  38. 38.

    Shao, L. et al. Discovery of N-Methyl-1-(1-phenylcyclohexyl)methanamine, a Novel Triple Serotonin, Norepinephrine, and Dopamine Reuptake Inhibitor. Bioorg. Med. Chem. Lett. 21, 1438–1441 (2011).

  39. 39.

    Shao, L. et al. Synthesis and Pharmacological Characterization of Bicyclic Triple Reuptake Inhibitor 3-Aryl Octahydrocyclopenta[c]pyrrole Analogues. J. Med. Chem. 54, 5283–5295 (2011).

  40. 40.

    Nolan, T. L. et al. Identification of a Novel Selective Serotonin Reuptake Inhibitor by Coupling Monoamine Transporter-Based Virtual Screening and Rational Molecular Hybridization. ACS Chem. Neurosci. 2, 544–552 (2011).

  41. 41.

    Micheli, F. et al. 1-Heteroaryl-6-(3,4-dichlorophenyl)-3-azabicyclo[4.1.0]heptane: Further Insights into a Class of Triple Re-Uptake Inhibitors. Bioorg. Med. Chem. 19, 3451–3461 (2011).

  42. 42.

    Gopishetty, B. et al. Further Structure-Activity Relationship Studies on 4-((((3S,6S)-6-Benzhydryltetrahydro-2H-pyran-3-yl)amino)methyl)phenol: Identification of Compounds with Triple Uptake Inhibitory Activity as Potential Antidepressant Agents. J. Med. Chem. 54, 2924–2932 (2011).

  43. 43.

    Angus, D. et al. The Identification, and Optimisation of hERG Selectivity, of a Mixed NET/SERT Re-Uptake Inhibitor for the Treatment of Pain. Bioorg. Med. Chem. Lett. 21, 271–275 (2011).

  44. 44.

    Micheli, F. et al. 1-(Aryl)-6-[alkoxyalkyl]-3-azabicyclo[3.1.0]hexanes and 6-(Aryl)-6-[alkoxyalkyl]-3-azabicyclo[3.1.0]hexanes: a New Series of Potent and Selective Triple Reuptake Inhibitors. J. Med. Chem. 53, 2534–2551 (2010).

  45. 45.

    Micheli, F. et al. 6-(3,4-Dichlorophenyl)-1-[(methyloxy)methyl]-3-azabicyclo[4.1.0]heptane: a New Potent and Selective Triple Reuptake Inhibitor. J. Med. Chem. 53, 4989–5001 (2010).

  46. 46.

    Bismuth-Evenzal, Y. et al. N-Methyl-Citalopram: A Quaternary Selective Serotonin Reuptake Inhibitor. Biochem. Pharmacol. 80, 1546–1552 (2010).

  47. 47.

    Roggen, H. et al. Synthesis of Enantiomerically Pure Milnacipran Analogs and Inhibition of Dopamine, Serotonin, and Norepinephrine Transporters. Bioorg. Med. Chem. Lett. 17, 2834–2837 (2007).

  48. 48.

    Zhang, S. et al. Further Structural Exploration of Trisubstituted Asymmetric Pyran Derivatives (2S,4R,5R)-2-Benzhydryl-5-benzylamino-tetrahydropyran-4-ol and Their Corresponding Disubstituted (3S,6S) Pyran Derivatives: a Proposed Pharmacophore Model for High-Affinity Interaction with the Dopamine, Serotonin, and Norepinephrine Transporters. J. Med. Chem. 49, 4239–4247 (2006).

  49. 49.

    Toda, N. et al. A Conformational Restriction Approach to the Development of Dual Inhibitors of Acetylcholinesterase and Serotonin Transporter as Potential Agents for Alzheimer’s Disease. Bioorg. Med. Chem. 11, 4389–4415 (2003).

  50. 50.

    Toda, N. et al. Design, Synthesis and Structure-Activity Relationships of Dual Inhibitors of Acetylcholinesterase and Serotonin Transporter as Potential Agents for Alzheimer’s Disease. Bioorg. Med. Chem. 11, 1935–1955 (2003).

  51. 51.

    Zhang, A. et al. Thiophene Derivatives: a New Series of Potent Norepinephrine and Serotonin Reuptake Inhibitors. Bioorg. Med. Chem. Lett. 12, 993–995 (2002).

  52. 52.

    Hoepping, A. et al. Novel Conformationally Constrained Tropane Analogues by 6-Endo-Trig Radical Cyclization and Stille Coupling - Switch of Activity Toward the Serotonin and/or Norepinephrine Transporter. J. Med. Chem. 43, 2064–2071 (2000).

  53. 53.

    Soubhye, J. et al. Hybrid Molecules Inhibiting Myeloperoxidase Activity and Serotonin Reuptake: a Possible New Approach of Major Depressive Disorders with Inflammatory Syndrome. J. Pharm. Pharmacol. 66, 1122–1132 (2014).

  54. 54.

    Tsuruoka, N. et al. A DKP Cyclo(L-Phe-L-Phe) Found in Chicken Essence is a Dual Inhibitor of the Serotonin Transporter and Acetylcholinesterase. PLoS One 7, e50824 (2012).

  55. 55.

    Liang, S. et al. A Novel Sesquiterpene and Three New Phenolic Compounds from the Rhizomes of Acorus tatarinowii Schott. Bioorg. Med. Chem. Lett. 25, 4214–4218 (2015).

Download references


The authors are grateful to Prof. Yi Zhang and Li-Feng Han at Institute of Traditional Chinese Medicine, Tianjin University of Traditional Chinese Medicine, China for measuring the optical rotation and CD data. ECD experiment and calculation were accomplished by Dr. Li Li in Institute of Materia Medica, Chinese Academy of Medical Science Peking Union Medical College. This work was financially supported by the National Natural Science Foundation of China (No. 81603250), the International Science & Technology Cooperation Program of China (Grant No. 2013DFA31620) and National Major Scientific and Technological Special Project for “Significant New Drugs Development” (No. 2012ZX09304007).

Author information

H.-H. Wu and Y.-T. Xu designed experiments. Y.-P. Chen, H.-H. Zheng, Z.-P. Wang, H. Zhang, X. Deng and H.-H. Wu performed the isolation of compounds and analyzed the spectroscopic data. S.-S. Ying, Y.-T. Liu and Y.-J. Wu contributed to the biological activity evaluation. T.-X. Li identified the herbal material. The manuscript was prepared by H.-H. Wu. and Y.-T. Xu. The research work was supported by the projects of H.-H. Wu, Y.-T. Xu, X.-M. Gao and Y. Zhu. All the authors reviewed the manuscript.

Correspondence to Yan-Tong Xu or Hong-Hua Wu.

Ethics declarations

Competing Interests

The authors declare that they have no competing interests.

Additional information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.