Diterpenoid alkaloids from Aconitum anthoroideum that offer protection against MPP+–Induced apoptosis of SH–SY5Y cells and acetylcholinesterase inhibitory activity

Article abstract of DOI:10.1016/j.phytochem.2020.112459

Nine unprecedented diterpenoid alkaloid, including a diterpenoid alkaloid featuring a diterpenoid moiety, anthoroidine A; one bisditerpenoid alkaloid joined with a carbon–carbon single bond, anthoroidine B; three racemulosine-type C₂₀-diterpenoid alkaloids, anthoroidines C–E; one hetidine-type C₂₀-diterpenoid alkaloid, anthoroidine F; and three hetisine-type C₂₀-diterpenoid alkaloids, anthoroidines G-I, together with ten known diterpenoid alkaloids were isolated from Aconitum anthoroideum DC. Their structures were established via detailed spectroscopic analyses. Most of the isolated compounds along with five known diterpenoid alkaloids obtained in a previous study were screened for neuroprotective activities and acetylcholinesterase inhibitory effects. Nominine showed potent protective activity against MPP⁺–induced apoptosis in SH–SY5Y cells, with a rescue rate of 34.4percent (50 μM). Rotundifosine F showed a significant inhibitory activity against AChE (IC₅₀ = 0.3 μM). The structure–activity relationship of these alkaloids is also briefly discussed.

Full text of DOI:10.1016/j.phytochem.2020.112459

Phytochemistry 178 (2020) 112459
Contents lists available at ScienceDirect
Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem
Diterpenoid alkaloids from Aconitum anthoroideum that offer protection
against MPPþ–Induced apoptosis of SH–SY5Y cells and acetylcholinesterase
inhibitory activity
Shuai Huang, Ji-Fa Zhang, Lin Chen, Feng Gao, Xian-Li Zhou*
School of Life Science and Engineering, Southwest Jiaotong University, Chengdu, 610031, Sichuan, PR China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Nine unprecedented diterpenoid alkaloid, including a diterpenoid alkaloid featuring a diterpenoid moiety,
anthoroidine A; one bisditerpenoid alkaloid joined with a carbon–carbon single bond, anthoroidine B; three
racemulosine-type C20-diterpenoid alkaloids, anthoroidines C–E; one hetidine-type C20-diterpenoid alkaloid,
anthoroidine F; and three hetisine-type C20-diterpenoid alkaloids, anthoroidines G-I, together with ten known
diterpenoid alkaloids were isolated from Aconitum anthoroideum DC. Their structures were established via
detailed spectroscopic analyses. Most of the isolated compounds along with five known diterpenoid alkaloids
obtained in a previous study were screened for neuroprotective activities and acetylcholinesterase inhibitory
effects. Nominine showed potent protective activity against MPPþ–induced apoptosis in SH–SY5Y cells, with a
Aconitum anthoroideum DC.
Ranunculaceae
Diterpenoid alkaloids
Anthoroidine a
Anthoroidine B
Neuroprotective activities
Acetylcholinesterase inhibitory effects
rescue rate of 34.4% (50
μM). Rotundifosine F showed a significant inhibitory activity against AChE (IC50 ¼ 0.3
μ
M). The structure–activity relationship of these alkaloids is also briefly discussed.
1. Introduction
their neuroprotective activities and cholinesterase inhibitory effects
(Ahmad et al., 2017a, 2017b,Atta–ur–Rahman et al., 2000).
Alzheimer’s disease (AD) and Parkinson’s disease (PD), the most
common progressive neurodegenerative diseases (NDs) impairing
memory, cognition and movement, are an important modern health
problem because they affect millions of people (primarily in the aging
population) worldwide, and the number of patients is growing every day
(Tarasoff-Conway et al., 2015; Shihabuddin et al., 2018). These condi-
tions have an enormous global impact estimated to be more than $604
billion, affecting not only patients but also caregivers, who are usually
family members (Alzheimer’s Association, 2018; Betarbet et al., 2000).
There is an urgent need to discover efficient therapeutic drugs that can
block or delay the progressive loss of neurons.
Aconitum anthoroideum DC., belonging to the family Ranunculaceae,
is mainly distributed in the Xinjiang Uygur region (Wang, 1979), and its
bioactive constituents have not yet been reported. As part of our ongoing
research program toward discovering structurally intriguing and bio-
logically important diterpenoid alkaloids from the Aconitum and
Delphinium genera, we studied the whole plants of A. anthoroideum,
which led to the isolation of nine unprecedented diterpenoid alkaloids,
namely anthoroidines A–I (1–9), together with ten known compounds,
tanguticuline E (10) (Fan et al., 2019), hetidine hydrochloride (11)
(Pelletier et al., 1970), hetisinone (12) (Gonzalez et al., 1986), hetisine
(13) (Jiang and Pelletier, 1991), tadzhaconine (14) (Yusupova
Natural herbal products play a vital role in drug discovery (Ibrahim
et al., 2013; Malak et al., 2018). Diterpenoid alkaloids, the main char-
acteristic constituents of the genera Aconitum and Delphinium, possess
interesting bioactivities, such as antiepileptiform, antioxidant, anti-
–inflammatory, analgesic, antiarrhythmia, antifungal, and cytotoxic
properties, as well as being an antagonist of the neuronal nicotinic
acetylcholine receptor (Wang et al., 2010; McKay et al., 2007). How-
ever, only a small proportion of alkaloids have been investigated for
et al., 1992), (þ)–(13R,19S)–1β,11α–diacetoxy–2α–benzoyloxy–13,
19–dihydroxyhetisan (15) (Jiang et al., 2012), trifoliolasine E (16)
(Zhou et al., 2005), atisinium chloride (17) (Pelletier and Mody, 1979),
nominine (18) (He et al., 1997), and aconicarmicharcutinium A hy-
drochloride (19) (Meng et al., 2017a, 2017b). The protective effects
against MPPþ–induced apoptosis in SH–SY5Y cells and the acetylcho-
linesterase (AChE) inhibitory activities of these isolated compounds
along with five known diterpenoid alkaloids, rotundifosine E (20),
* Corresponding author. Natural Products Laboratory of School of Life Science and Engineering, Southwest Jiaotong University, Chengdu, 610031, Sichuan, PR
China.
E-mail address: zhouxl@swjtu.edu.cn (X.-L. Zhou).
https://doi.org/10.1016/j.phytochem.2020.112459
Received 2 March 2020; Received in revised form 8 July 2020; Accepted 12 July 2020
Available online 23 July 2020
0031-9422/© 2020 Elsevier Ltd. All rights reserved.

S. Huang et al.
Phytochemistry 178 (2020) 112459
rotundifosine F (21), rotundifosine G (22), atidine (23), and ajaconine
(24), obtained in our previous phytochemical studies of Aconitum
rotundifolium (Zhang et al., 2019), were evaluated. Herein, the isolation
and structural elucidation of these alkaloids as well as the biological
evaluation of these compounds are described.
Table 1
NMR Data for Compounds 1 and 2 (1H: 600 MHz,13C: 150 MHz; δ in ppm, J in
Hz).
No.
1 (in CDCl3)
2 (in CD3OD)
δH
δC, type
δH
δC, type
1
2
3
α
1.67 m
28.9, CH2
α
1.73 m
29.2, CH2
2. Results and discussion
β 3.17 d (15.6)
β 1.51 m
3.71 br. s
72.8, CH
α
1.20 m
21.9, CH2
31.6, CH2
β 1.53 a
2.1. Structure elucidation and identification
α
1.97 m
37.1, CH2
α 1.78 m
β 1.50 a
–
β 1.24 m
Compound 1, an amorphous solid, displayed a protonated molecular
ion at m/z 626.3845 [M þ H]þ in its HRESIMS spectrum (calcd.
626.3845), corresponding to the molecular formula C40H51NO5. Its IR
spectrum showed absorption bands for hydroxy (3393 cmÀ 1), carbonyl
(1723 cmÀ 1) and carbon–carbon double bond (1666 cmÀ 1) functional
groups. The NMR data (Table 1) of 1 exhibited signals characteristic of a
4
5
6
36.1, C
–
–
46.4, C
1.91 s
3.92 br. s
58.3, CH
65.1, CH
73.0, C
α
1.46 m
32.6, CH2
β 1.69 a
7
α
1.74 d (14.4)
34.2, CH2
α
1.98 m
32.3, CH2
β 2.15 d (14.4)
β 1.69 a
8
–
43.1, C
–
45.0, C
1.63 a
48.3, CH
46.5, C
trisubstituted double bond [δ 5.63 (s); δ 128.0 (d), 146.3 (s)], an
exocyclic double bond [δH 4.70 (s), 4.89 (s);CδC 109.1 (t), 143.6 (s)], two
tertiary methyl groups [δH 0.98 (s) and 1.15 (s); δ 24.2 (q), 29.5 (q)],
one aldehyde moiety [δH 9.85 (s); δC 206.7 (d)],Ca ketone group [δC
227.3 (s)], and an oxygenated methylene carbon [δH 3.87 (d, J ¼ 14.2
H
9
2.08 m
–
55.1, CH
51.2, C
10
11
–
4.26 s
76.0, CH
α
1.36 m
28.8, CH2
β 1.63 a
12
13
2.49 s
4.28 s
50.5, CH
71.4, CH
2.34 br. s
35.6, CH
α
1.81 m
44.3, CH2
Hz), and 4.17 (d, J ¼ 14.2 Hz); δC 68.8 (t)]. Further analysis of the 13
C
β 1.49 m
14
15
2.70 d (8.4)
50.1, CH
1.53 a
45.7, CH
NMR, DEPT and HSQC data of 1 revealed that the unassigned carbon
signals could be attributed to eleven methylenes, fourteen sp3 methines
(three oxygenated) and six sp3 quaternary carbons. These features sug-
gested that 1 consists of a C20–diterpenoid alkaloid moiety (in black) and
a diterpenoid (in blue) fragment (Wang and Liang, 2002).
α
2.16 m
33.2, CH2
5.28 d (0.8)
128.4, CH
β 2.05 m
–
16
17
143.6, C
–
151.6, C
a 4.89 s
b 4.70 s
1.15 s
109.1, CH2
2.27 m
33.2, CH2
18
19
29.5, CH3
60.0, CH2
1.05 s
19.3, CH3
172.8, CH
The diterpenoid alkaloid moiety was deduced as hetisine (Jiang and
Pelletier, 1991) by the similarities in their 1D and 2D NMR data. The
ketone group (δC 227.3) in the diterpenoid moiety was assigned to C–200
by the H–10/C–200 and H–130/C–200 HMBC correlations. The correla-
tions observed in the HMBC experiment between the signals at δH 1.20
(H–30), 1.54 (1H, m, H–50), and 0.98 (3H, s, H–180) with δ 206.8
allowed the assignment of the aldehyde group at C–190. CompaCring the
spectroscopic data of the diterpenoid moiety in 1 with those of campy-
lopin (Wang and Yan, 2007), a hetidane–type diterpene isolated from
Delphinium campylocentrum, indicated that they have the same molecular
skeleton. The main difference was that the acetyl and hydroxyl groups at
C–70 and C–150 in campylopin were both substituted by hydrogen atoms
in 1. In addition, the expected exocyclic double bond was isomerized to
the C–150–C–160 endocyclic double bond, and this is strongly supported
by the HMBC cross–peaks of H–150/C–70, H–150/C–170, and
H–170/C–120. Additionally, the HMBC correlations of H–170 with C–2,
C–150 and C–120, and of H–2 with C–10 and C–4 implied that the non-
–alkaloid part was linked with the alkaloid part via a C–170–O–C–2 ether
linkage. Accordingly, the complete planar structure of 1 was further
verified by analysis of the HMBC, HMQC, and 1H–1H COSY spectra
(Fig. S1).
α
3.81 d (11.4)
7.38 d (3.2)
β 2.99 d (11.4)
20
10
4.53 s
68.0, CH
3.48 br. s
81.4, CH
α
1.63 m
28.4, CH2
α
1.79 a
34.0, CH2
β 1.32 m
2.07 m
β 2.27 m
β 2.74 d (15.2)
20
30
α
19.7, CH2
35.6, CH2
4.14 a
67.8, CH
α
1.20 m
α
1.51 m
40.4, CH2
β 1.98 a
–
β 1.79 a
–
40
50
60
49.0, C
37.8, C
1.54 m
53.7, CH
21.9, CH2
1.60 a
3.65 s
59.6, CH
71.0, CH
α
1.42 m
β 1.98 m
70
α
2.17 m
33.7, CH2
α
1.80 m
40.6, CH2
β 2.03 m
β 1.71 m
80
–
42.3, C
–
40.5, C
90
2.24 m
53.9, CH
53.7, C
1.60 a
–
59.6, CH
46.4, C
100
110
–
α
1.60 m
30.4, CH2
4.98 s
92.9, CH
β 2.06 m
120
130
2.40 br. s
31.8, CH
5.65 d (6.4)
4.14 a
121.2,CH
70.9, CH
α
1.32 m
35.9, CH2
β 1.64 m
140
150
1.98 a
52.6, CH
1.75 m
48.8, CH
5.63 s
128.0, CH
α
1.88 d (19.2)
37.4, CH2
The relative configuration of 1 was established through analysis of
the coupling constants and key NOESY correlations (Fig. S2). The
180–methyl group was assigned as β–oriented due to the correlations of
H–180 with H–60β, H–30β and H–50β in the NOESY spectrum. As a
β 2.23 d (19.2)
160
170
–
146.3, C
–
145.8, C
a 4.17 d (14.2)
b 3.87 d (14.2)
0.98 s
68.8, CH2
2.16 m
35.5, CH2
180
190
24.2, CH3
206.7, CH
0.99 s
30.0, CH3
64.3, CH2
consequence, H–190 was assigned as being
α–oriented, which was also
9.85 s
α 3.34 d (11.2)
supported by the correlations between H–190 and H–20α. The correla-
tions of H–30β/H–10β, H–90/H–10β, H–70β/H–90, and H–50/H–70β indi-
cated that H–50 and H–90 were also β–oriented. Comparison of the
chemical shift and peak shape of H–2 (δH 4.16, br. S) with those of
known hetisine–type alkaloids (Bessonova and Saidkhodzhaeva, 2000)
suggested that the hydrogen at C–2 should be β–oriented. The large
β 2.46 d (11.2)
200
a
–
227.3, C
3.30 m
66.4, CH
indicates overlapping signals.
the structure of compound 1 was determined as shown in Fig. 1, and it
was named anthoroidine A.
coupling constant of H–13 (J ¼ 8.4 Hz) with H–14
α
revealed that the
A plausible biogenetic pathway for 1 was proposed and is shown in
Fig. 2. A hetidane–type diterpene (I) could be oxidized to corresponding
epoxide II. Subsequently, a nucleophilic primary hydroxyl group of
hetisine could attack the oxirane moiety of II, generating corresponding
conjugated diterpenoid alkaloid III, which could be converted to
anthoroidine A by the elimination of water.
dihedral angle between these two H–atoms was ca. 0 �C, which implied
that H–13 was in an
α–orientation. The NOESY correlation of H–11 with
H–15β indicated the β–orientation of H–11. Since the absolute config-
uration of hetisine was confirmed by X–ray crystallographic analysis
(Tashkhodzhaev et al., 1992), it was proposed that the diterpenoid
alkaloid substructure in 1 had the same absolute configuration. Hence,
2

S. Huang et al.
Phytochemistry 178 (2020) 112459
Fig. 1. Structures of compounds 1–24.
The molecular formula of anthoroidine B (2), which was isolated as a
moiety of trichocarpinine, which was further verified by 2D NMR ex-
periments. Apart from the hetidine–type alkaloid skeleton, further
comparison of the NMR data with those of the known compound heti-
sine (Jiang et al., 2012) indicated that they have the same molecular
skeleton; the main differences were the hemiacetal group at C–110 in 2,
and that the typical exocyclic double bond was isomerized to a
C–120–C–160 endocyclic double bond. These changes were confirmed by
the HMBC cross–peaks of H–110/C–80 and C–90, and of H–120/C–140,
C–160 and C–170, and the 1H–1H COSY correlations of H–110/H–90, and
H–120/H–130. Furthermore, the hetidine–type alkaloid moiety (in black)
was linked with the hetisine fragment (in blue) via a single bond be-
tween C–170 and C–17 based on the HMBC correlations of H–17/C–160
and C–170 and of H–170/C–16 and C–17 as well as the 1H–1H COSY
correlation of H–170/H–17.
white powder, was inferred to be C40H52N2O4 from its HRESIMS ([M þ
H]þ at m/z 625.3997, calcd for C40
H
53
N2O4: 625.4005) and 13C NMR
data. Analysis of its NMR data (Table 1) revealed the existence of two
angular methyl groups [δH 1.05 (3H, s), 0.99 (3H, s); δ 19.3 (q), 30.0
(q)], two trisubstituted double bonds [δH 5.28 (1H, d, CJ ¼ 0.8 Hz); δC
128.4 (d), 151.6 (s); δ 5.65 (1H, d, J ¼ 6.4 Hz); δ 121.2 (d), 145.8 (s)],
H
C
and two diagnostic sets of three non–oxygenated quaternary carbons (δC
46.4/37.8, 45.0/40.5, 46.5/46.4) and an oxygenated methylene carbon
(δ 73.0). The aforementioned evidence suggested that compound 2
C
consisted of two C20–diterpenoid alkaloid moieties (Wang and Liang,
2002). Comparison of the 13C NMR, HMQC, and DEPT data with those of
the known bis–diterpenoid alkaloid trichocarpinine (Lin et al., 2009)
indicated that one moiety of 2 was connected to the hetidine–type
3

S. Huang et al.
Phytochemistry 178 (2020) 112459
Fig. 2. Plausible biosynthetic pathway of anthoroidine a (1).
In the NOESY spectrum of 2 (Fig. S2), the cross–peak between H–10β/
anthoroidine B by a retro–aldol followed by an intramolecular nucleo-
philic addition reaction of a hydroxyl group and a formyl group.
The molecular formula of compound 3 was determined to be
C24H33NO5 from its HRESIMS signal at m/z 416.2432 [M þ H]þ (calcd
for C24H34NO5, 416.2437). The IR data showed the presence of hydroxy
groups (3353 cmÀ 1) and carbonyl groups (1716 cmÀ 1). The NMR data
(Table 2) showed the presence of an N–ethyl group [δH 1.11 (3H, t, J ¼
H–130, and H-110/H-150β, proved that H–110 and H–130 were β–oriented.
Since the absolute configurations of hetitidine and hetisine were
confirmed by X–ray crystallographic analysis (Tashkhodzhaev et al.,
1992; Tang et al., 2008), the absolute configuration in 2 was proposed to
be the same, which was further supported by comparison of its experi-
mental and calculated electronic circular dichroism (ECD) spectra using
the quantum chemical time-dependent density functionaltheory
(TDDFT) method (Cheng et al., 2019) (Fig. S84). Hence, the structure of
anthoroidine B (2) was determined as shown.
7.2 Hz); δ 13.3 (q), 48.1 (t)], a methine carbon bearing a hydroxyl
C
group [δH 3.88 (1H, d, J ¼ 9.6 Hz); δC 75.3 (d)], an acetoxy group [δH
1.88 (3H, s), δ 22.5 (q), 170.0 (s)], a carboxyl group (δ 178.6) and a
C
A possible biosynthetic pathway for 2 was proposed (Fig. 3), and two
coexisting alkaloids, naviculine B (a) and guanfu base V (b), were
considered to be the monomeric biosynthetic precursors of 2. An
intermolecular [2 þ 2] cycloaddition of two diterpenoid alkaloids would
give intermediate I, which could then be transformed into key inter-
mediate III by oxidation and elimination to furnish the C–170–C–17
single bond of 2. Intermediate III would be finally converted to
vinyl group [δHC 5.61 (1H, dd, J ¼ 18.0, 11.4 Hz), 4.88 (1H, d, J ¼ 18.0
Hz), 4.96 (1H, d, J ¼ 18.0 Hz); δ 141.9 (d), 113.1 (t)]. Further analyses
of its 13C NMR and DEPT spectraC, which showed three diagnostic carbon
signals of two quaternary carbons (δ 47.3 and 57.3) and one oxygen-
C
ated tertiary carbon (δC 88.5), suggested that
3 had a rare
C
20–diterpenoid alkaloid skeleton similar to that of racemulosine (Wang
et al., 2000). Comparison of NMR data to racemulosine and HRESIMS,
Fig. 3. Plausible biosynthetic pathway of anthoroidine B (2).
4

S. Huang et al.
Phytochemistry 178 (2020) 112459
the major differences included 3 bearing an acetoxy group at C-8 instead
of a hydroxyl group, and the CONH2 group (δ 179.5) at C-14 replaced
by a COOH group (δC 178.6) in racemulosine.CThe carbon signal of C–8
at (δ 70.6) in racemulosine was shifted downfield to (δ 88.5) in
Table 2
NMR Data for Compounds 3–5 (1H: 600 MHz,13C: 150 MHz; δ in ppm, J in Hz; in
CDCl3).
No.
3
4
5
C
C
compound 3, suggesting that the acetoxy group in 3 might be located at
C–8, and which was further confirmed by the HMBC correlations from
H–6/H–7/H–9/H–14 to C–8. In addition, the carboxyl group was located
at C–14 base on the HMBC correlations from H–14 to C–17 and the
molecular formula deduced from HRESIMS.
δH
δC, type
δH
δC, type
δH
δC, type
1
2
3.88
75.3,
CH
3.85
75.3,
CH
3.88
75.3,
CH
d (9.6)
d (9.6)
d (9.6)
α
1.64
49.5,
CH2
α
1.64
49.5,
CH2
α
1.66
49.5,
CH2
d (13.8)
β 2.24 m
5.61 dd
(18.0,
11.4)
d (13.8)
β 2.23 m
5.62 dd
(18.0,
11.4)
d (13.8)
β 2.27 m
5.65 dd
(18.0,
11.4)
The relative configuration of compound 3 (Fig. S2) was deduced
from a NOESY experiment. The NOESY cross–peaks between H–1/H–5,
3
141.9,
CH
141.8,
CH
141.8,
CH
H–1/H–10 indicated the
α–orientation of OH–1. Additionally, the
cross–peak between H–14/H–10 showed that H–14 is β–oriented. Since
the absolute configuration of the skeleton of racemulosine was
confirmed by X–ray crystallography (Wang et al., 2000), the absolute
configuration of this skeleton was proposed to be retained in 3. There-
fore, the structure and absolute configuration of anthoroisine C (3) were
defined as shown in Fig. 1.
4
5
–
47.3, C
53.7,
CH
–
47.2, C
53.5,
CH
–
57.4, C
53.4,
CH
1.94
1.90 a
1.95
d (7.2)
d (7.8)
6
α
1.43 dd
24.8,
CH2
α
1.41 m
24.8,
CH2
α
1.42 dd
23.9,
CH2
(15.0, 7.2)
(15.0, 7.2)
β 1.81 dd
(15.0, 7.2)
2.53
β 1.78 m
β 1.75 m
The HRESIMS signal at m/z 578.2968 [M þ H]þ (calcd for
7
3.28
42.8,
CH
3.21
42.8,
CH
41.1,
CH
d (7.2)
–
d (6.6)
–
d (7.2)
–
C
30H44NO10, 578.2965) implied that the molecular formula of 4 was
8
9
88.5, C
44.7,
CH
88.4, C
44.5,
CH
82.4, C
42.6,
CH
C
30H43NO10. A comparison of the 13C NMR data of 3 and 4 indicated the
2.66 dd
(6.6, 4.8)
2.19 m
2.66 t
(6.0)
2.18 m
2.34 a
presence of resonances for a sugar moiety (δ 94.1, 76.8, 76.6, 72.9, 69.9
C
10
46.0,
CH
45.8,
CH
2.17 m
46.2,
CH
and 61.7) (Huang et al., 2013), which was identified as β–D–glucose by
gas chromatography of the hydrolyzed product (Fig. S39) and based on
the coupling constant of the anomeric proton [δH 5.58 (1H, d, J ¼ 7.8
Hz)]. Relative to those of 3, the 13C NMR signal of C–17 of 4 was shifted
11
12
–
57.3, C
34.4,
CH2
–
57.2, C
34.2,
CH2
–
57.4, C
33.3,
CH2
α
1.75 m
α
1.76 m
α
1.76 m
β 1.95 m
β 1.90 a
β 2.34 a
upfield (Δδ 6.7 ppm), suggesting that the glucose moiety was located at
13
14
15
16
17
18
2.47 m
32.6,
CH
2.49 m
32.8,
CH
2.71 m
35.6,
CH
C
C–17, which was supported by the HMBC correlation of H–10 with C–17.
All of the available evidence suggests that the structure of anthoroisine D
(4) is as depicted in Fig. 1.
2.53 m
48.2,
CH
2.63 m
48.1,
CH
2.73 m
50.8,
CH
α
1.79 m
29.4,
CH2
α
1.86 m
29.0,
CH2
α
1.82 m
26.3,
CH2
The protonated molecular ion in the HRESIMS data of compound 5
at m/z 388.2482 (calcd for C23H34NO4: 388.2488, [M þ H]þ) provided a
molecular formula of C23H33NO4. Combined analyses of the spectro-
scopic data suggested that 5 is structurally similar to 3, indicated that 5
also possesses a racemulosine–type diterpenoid alkaloid skeleton, the
main difference was that the acetoxy at C–8 in 3 was replaced by a
methoxy group in 5, which was validated by the loss of 28 mass units in
mass spectrometry. This replacement was supported by the HMBC cor-
relations (Fig. S1) of the protons of OCH3–8 with C–8. Thus, the struc-
ture of anthoroisine E (5) was confirmed by extensive analyses of its 1D
and 2D NMR spectra.
β 2.45 m
β 2.43 m
β 2.38 m
α
1.27 m
28.1,
CH2
α
1.28 m
28.1,
CH2
α
1.26 m
26.7,
CH2
β 2.23 m
β 2.15 m
β 2.26 m
–
178.6,
C
–
171.9,
C
–
174.0,
C
a 4.96 br.
d (11.4)
113.1,
CH2
a 4.97 br.
d (11.4)
113.1,
CH2
a 5.00 br.
d (11.4)
113.1,
CH2
b 4.88 br.
d (18.0)
b 4.88 br.
d (18.0)
b 4.91 br.
d (18.0)
19
α
2.31
55.5,
CH2
α
2.30
55.4,
CH2
α
2.63
55.6,
CH2
d (10.8)
β 2.57 a
2.98 s
d (10.8)
β 2.59 m
2.97 s
d (10.8)
2.96 s
β 2.34 a
62.4,
CH
20
62.4,
CH
62.5,
CH
Compound 6 possessed the molecular formula C20H27NO2, which
was determined from its HRESIMS signals at m/z 314.2114 ([M þ H]þ,
21
a 2.54 m
b 2.52 m
1.11 t
48.1,
CH2
a 2.55 m
b 2.52 m
1.12 t
48.1,
CH2
13.3,
CH3
22.7,
CH3
170.9,
C
a 2.57 m
b 2.54 m
1.14 t
48.1,
CH2
13.3,
CH3
calcd 314.2120).
6
was determined to be
a
hetidine–type
22
13.3,
CH3
C
20–diterpenoid alkaloid based on its NMR spectra (Tables 3 and 4),
which showed the presence of an exocyclic double bond [(δH 4.99 (s)
(7.2)
(7.2)
(7.2)
OAc
1.88 s
22.5,
CH3
1.93 s
and 5.02 (s); δ 156.3 (s), 109.1 (t)], a methyl group [δ 0.92 (s); δ 21.6
C
C
(q)], and three diagnostic non–oxygenated quaternary cHarbon signals (δC
38.9, 45.2, and 50.2) (Wang and Liang, 2002). The signal at δ 183.6 in
the 13C NMR data indicated that C–20 is connected to a nitrogeCn through
a double bond (Wang and Liang, 2002), which was supported by the
HMBC correlations of H–1, H–9, and H–19 with C–20. Along with the
above–mentioned signals, the 13C NMR spectrum of 6 displayed signals
for two oxygenated carbons [δC 77.0 (s) and 71.5 (d)], suggesting that
this compound possessed two additional hydroxy groups. These two
hydroxy groups were assigned at C–5 and C–15 based on the HMBC
correlations from H–6, H–7, H–18, and H–19 to C–5 and from H–9 and
H–17 to C–15. In the NOESY spectrum of 6, the cross–peak between
170.0,
C
OMe
10
3.22 s
48.6,
CH3
5.58
94.1,
CH
d (7.8)
3.48 m
20
72.9,
CH
30
3.59 m
3.56 m
3.46 m
3.79 br s
76.8,
CH
40
69.9,
CH
50
76.6,
CH
60
61.6,
CH2
H–15 and H–7α, H–7α and H–14 proved that H–15 was α–oriented.
Therefore, the structure of anthoroisine F (6) was determined as shown
in Fig. 1, and the full assignment of its spectroscopic data was achieved
based on the 1D–and 2D NMR analysis.
a
indicates overlapping signals.
The molecular formula of compound 7 was established as C27H31NO3
based on its HRESIMS and 13C NMR data. NMR spectroscopic data
(Tables 3 and 4) of compound 7 in conjunction with the HSQC data
indicated the presence of an angular methyl group [δH 0.97 (3H, s); δ
C
5

S. Huang et al.
Phytochemistry 178 (2020) 112459
Table 3
Table 4
1H NMR Data for Compounds 6–9 (600 MHz, in CDCl3, δH in ppm, J in Hz).
13C NMR Data for Compounds 6–9 (150 MHz, in CDCl3, δC in ppm).
No.
6
7
8
9
No.
6
7
8
9
1
α
1.89 m
α
1.84 m
4.07 a
α
1.94 d (16.2)
1
26.7, CH2
20.2, CH2
35.0, CH2
38.9, C
26.4, CH2
19.7, CH2
34.0, CH2
38.1, C
65.9, CH
31.2, CH2
38.0, CH2
35.6, C
31.4, CH2
70.4, CH
36.7, CH2
36.7, C
β 1.42 m
β 1.20 m
β 2.95 d (16.2)
2
2
3
α
1.44 m
β 1.63 m
1.71 m
α
1.47 m
β 1.64 m
1.44 m
α
3.13 d (14.4)
a 5.19 m
3
β 1.40 m
4
α
α
α
2.19 a
α
1.54 dd (15.6,
5
77.0, C
61.1, CH
65.2, CH
32.4, CH2
46.1, C
58.0, CH
65.0, CH
34.3, CH2
43.5, C
61.5, CH
64.3, CH
36.3, CH2
44.1, C
4.2)
6
27.3, CH2
30.6, CH2
45.2, C
β 1.13 m
β 1.24 m
β 1.37 m
β 1.84 br.
d (15.6)
1.60 s
7
8
5
6
–
α
1.44 s
1.94 s
9
51.3, CH
50.2, C
43.1, CH
49.8, C
53.3, CH
50.4, C
55.4, CH
50.7, C
1.81 m
3.22 br. s
3.89 br. s
3.22 s
10
11
12
13
14
15
16
17
18
19
20
10
β 1.72 m
26.3, CH2
33.8, CH
31.0, CH2
36.2, CH
71.5, CH
156.3, C
109.1, CH2
21.6, CH3
65.0, CH2
183.6, C
23.3, CH2
38.5, CH
76.3, CH
52.9, CH
71.3, CH
150.6, C
112.8, CH2
28.9, CH3
63.0, CH2
72.0, CH
166.0, C
130.5, C
129.6, CH
128.4, CH
133.0, CH
128.4, CH
129.6, CH
75.9, CH
47.9, CH
70.2, CH
50.4, CH
33.1, CH2
143.6, C
109.8, CH2
29.4, CH3
60.4, CH2
68.7, CH
166.8, C
130.1, C
128.7, CH
130.3, CH
133.4, CH
130.3, CH
128.7, CH
75.5, CH
48.8, CH
74.0, CH
50.4, CH
34.0, CH2
144.9, C
108.9, CH2
29.7, CH3
63.5, CH2
68.9, CH
176.5, C
34.0, CH
19.6, CH3
18.9, CH3
7
α
1.32 dt (9.6,
α
1.70 dd (13.6,
α
1.85 dd (15.0,
α 1.75 dd (15.6,
1.2)
2.8)
7.8)
2.4)
β 2.35 a
β 2.09 dd (13.6,
2.8)
β 2.15 m
β 1.62 dd (15.6,
2.4)
9
2.32 m
1.84 a
2.44 m
1.98 dd (10.2,
2.4)
11
α
1.75 m
α
1.62 m
5.47 d (9.0)
a 4.28 d (9.0)
β 2.03 dd (14.4,
4.2)
β 1.93 dd (14.4,
4.0)
12
13
2.34 m
2.55 m
2.68 d (2.4)
4.30 d (9.0)
2.43 d (2.4)
5.15 dt (10.2,
2.4)
α
1.72 a
4.56 d (2.4)
20
β 1.41 m
30
14
15
17
2.25 d (10.2)
1.84 a
4.10 s
3.03 d (9.0)
2.28 dd (10.2,
2.4)
40
50
4.03 s
α
2.19 a
α
2.22 d (17.4)
60
β 2.42 m
a 4.99 s
b 4.82 s
1.11 s
β 2.04 d (17.4)
a 4.91 s
70
a 5.02 s
b 4.99 s
0.92 s
a 5.18 s
b 5.03 s
0.97 s
OAc
170.3, C
b 4.72 s
22.0, CH3
18
19
0.98 s
α
3.79 ABq
α
2.43 ABq
α
4.09 ABq
α 2.86 ABq
(19.2)
(12.4)
(12.0)
(12.0)
and two hydroxy groups. Two hydroxy groups were assigned at C–1 (δ
C
β 3.56 ABq
(19.2)
β 2.51 ABq
(12.4)
β 2.99 ABq
(12.0)
β 2.50 ABq
(12.0)
65.9 d) and C–13 (δC 70.2 d) based on the HMBC correlations of H–3 and
H–20 with C–1 and of H–11 and H–14 with C–13, respectively. Addi-
tionally, the location of the benzoyl group was established based on the
HMBC correlation between H–11 and the carbonyl carbon (δ 166.8).
The large coupling constant of H–13 (J ¼ 9.0 Hz) with H–14αCrevealed
20
20
2.87 s
4.78 s
3.46 s
2.51 m
30
7.94 d (7.6)
7.39 t (7.6)
7.52 t (7.6)
7.39 t (7.6)
7.94 d (7.6)
8.15 d (7.2)
7.44 t (7.8)
7.56 t (7.8)
7.44 t (7.8)
8.15 d (7.2)
1.18 q (6.6)
1.16 q (6.6)
40
50
that H–13 was in an
α
–orientation. In the NOESY experiment, the
–orientation of H–1. The
60
cross–peaks of H–1/H–20 suggested the
α
70
coupling constant between H–11 with H–9β (J ¼ 9.0 Hz) indicated a 1,2-
diaxial relationship between them, implying that H–11 was β-oriented,
further supported by the correlations between H-11 and H–9β in the
NOESY experiment. Consequently, all of the available evidence sug-
gested the structure of anthoroisine H (8) is as depicted in Fig. 1.
Compound 9 was obtained as a white powder and had a formula of
OAc
2.03 (s)
a
indicates overlapping signals.
28.9], six methylenes (δC 26.4, 19.7, 34.0, 32.4, 23.3, 63.0), eight sp3
methines (δ 61.1, 65.2, 43.1, 38.5, 76.3, 52.9, 71.3, 72.0), and three sp3
C
quaternary carbons (δC 38.1, 46.1, 49.8), one exocyclic methylene group
C
26H35NO5 as determined from its HRESIMS (m/z 442.2590 [M þ H]þ,
[δH 5.03 (br. S) and 5.18 (br. S); δ 112.8, 150.6], and a benzoyl group
calcd 442.2593) and 13C NMR spectra. The characteristic NMR spec-
troscopic data (Tables 3 and 4) of 9 strongly indicated that it was a
hetisine–type C20–diterpenoid alkaloid possessing a hydroxy group, an
acetyl group [δH 2.03 (3H, s); δC 22.0 (q), 170.3 s)] and an isobutyryl
C
[δH 7.94 (2H, d, J ¼ 7.6 Hz), δH 7.52 (1H, t, J ¼ 7.6 Hz), δH 7.39 (2H, t, J
¼ 7.6 Hz); δ 130.5, 129.6 � 2, 128.4 � 2, 133.0, 166.0]. From the
C
above–mentioned spectroscopic data, compound 7 could be assigned as
a hetisine–type C20–diterpenoid alkaloid possessing a benzoyl group
(Wangchuk et al., 2007). Comparison of the 1D NMR data (Tables 3 and
4) of 7 with those of the known compound nominine (18) indicated that
7 has an additional benzoyl group located at C–13 and the HMBC
group [δ 1.16 (3H, q, J ¼ 6.6 Hz), 1.18 (3H, q, J ¼ 6.6 Hz); δ 18.9 (q),
19.6 (q),H34.0 (d), 176.5 (s)]. Long–range correlations observed in the
HMBC data from H–13 to the carbonyl carbon of the isobutyryloxy
group indicated that the isobutyryloxy group was at C–13. Further
analysis of the NMR spectra of 9 indicated that it was structurally related
to the known alkaloid trichodelphinine D (Tang et al., 2008). The dif-
ference was that the acetyl group in trichodelphinine D was replaced by
an isobutyryl group in 9, which was also supported by 2D NMR and
HRESIMS data. The large coupling constant in the 1H NMR spectrum of 9
C
cross–peaks between H–13 (δ 4.56, d, J ¼ 2.4 Hz) and the carbonyl
H
carbon (δC 166.0). The presence of a hydroxy group at C–15 was sup-
ported by the HMBC correlations from H–7 and H–9 to C–15 and from
H–15 to C–16 and C–17, and these correlations are listed in Fig. S1. As
shown in Fig. S2, the key NOESY correlations of H–13 with H–20, H–14
with H–15, and H–15 with H–7
and the
α indicated the β–orientation of H–13
of H–13 (J ¼ 10.2 Hz) with H–14
α revealed that H–13 was in an
α
–orientation of H–15. The relative configuration of 7 was
α
–orientation. The coupling constant of H–2 (J ¼ 4.2 Hz) with H–3
deduced base on analysis of the coupling constants, NOESY experiments
(Fig. S2) and X–ray (Fig. 4) experiments. Therefore, the structure of
anthoroisine G (7) was defined as shown in Fig. 1.
indicated that H–2 was in an equatorial position, which indicated a
β-orientation. Additionally, the coupling constant between H–11 with
H–9β (J ¼ 9.0 Hz) implying that H–11 was β-oriented. Hence, the
structure of anthoroisine I (9) was confirmed as shown in Fig. 1.
The HRESIMS data of 8 implied a molecular formula of C27H31NO4
based on the signal at m/z 434.2326 ([M þ H]þ, C27H32NO4, calcd
434.2331). The NMR spectroscopic data strongly suggested that 8 was
also a hetisine–type C20–diterpenoid alkaloid bearing a benzoyl moiety
6

S. Huang et al.
Phytochemistry 178 (2020) 112459
Fig. 4. ORTEP drawing of 7.
2.2. Biological evaluation
Table 6
Effects of compounds 6, 8–12, 14, and 15–17 against MPPþ–induced cytotox-
icity in SH–SY5Y cells a.
1–Methyl–4–phenylpyridinium (MPPþ) is widely used as a neuro-
toxin for inducing a syndrome closely resembling classic PD in animal
and cellular models (Singer and Ramsay, 1990; Przedborski and Jack-
son–Lewis, 1998). In this study, the effects of compounds 3, 4, and 6–19
on the viability of SH–SY5Y cells were determined by an MTT assay.
Treatment with compounds 6, 8–12, 14, and 17–19 resulted in more
Compound
Recovery ratio (%) b (20
μM)
Recovery ratio (%) b (50
μM)
6
À 10.2
À 11.5
À 37.4
À 6.4
À 7.5
À 4.1
2.9
À 16.7
À 9.7
À 14.5
À 10.9
À 4.5
À 12.2
À 6.8
14.9
8
9
10
11
12
14
17
18
19
than 95% cell viability at 50 μM in the SH–SY5Y cell line (Table 5).
Therefore, these compounds were chosen as candidate compounds for
testing against MPPþ–induced cytotoxicity in SH–SY5Y cells (Table 6).
Nominine (18) showed significant recovery of 34.4% at a concentration
13.6
5.1
34.4
of 50 μM, while aconicarmicharcutinium A hydrochloride (19) showed
16.8
7.8
the strongest recovery ratio of 16.8% at a concentration of 20 μM, and it
a
Recovery rate ¼ [(cell viability)þMPPþcompound – (cell viability)MþPP]/[(cell
was more potent than compounds 17, 18, and 14.
viability)þMPP] � 100%.
b
Mean value of three independent determinations.
3. Conclusions
inhibitory activity toward AChE, and huperzine A was employed as a
Although the etiologies of AD and PD are unknown, decreased
acetylcholine levels are correlated with these progressive NDs (Bartus
et al., 1982; Giacobine and Becker, 1997; Giacobini, 2000). The inhi-
bition of AChE has thus been considered a promising approach for the
treatment of AD and PD (Zhang et al., 2015). In our search for novel,
potent cholinesterase inhibitors, compounds 3 and 6–24 (>5 mg of each
of these compounds was obtained) were assayed in vitro for their
positive control (Table 7). Rotundifosine F (21) showed the strongest
AChEI activity (IC50 ¼ 0.3 μM). Compounds 22 and 17 also displayed
high potencies (IC50 < 2
10 M). The remaining alkaloids were inactive against AChE (IC50 > 10
M). Based on the aforementioned biological data, the structure–activity
μ
M), followed by 7, 14, 24, 9, and 23 (IC50
<
μ
μ
relationships (SARs) for these diterpenoid alkaloids can be briefly
summarized. (1) C–17 side-chain modifications might be an important
factor for the activity of the hetidine–type diterpenoid alkaloids
(compare 21, 22 and 6). (2) For hetisine–type diterpenoid alkaloids, the
hydroxyl substituents at C–2 and C–19 are beneficial to the activity. (3)
A benzoyl substituent at C–13 of hetisine–type alkaloids also improve
the activity.
Table 5
Effects of compounds 3, 4, and 6–19 on the viability of SH–SY5Y cells a.
Compound
Cell viability (%) b
Cell viability (%) b
(20
μM)
(50 μM)
3
91.2
86.4
84.3
90.2
99.0
64.0
In summary, nine undescribed and ten known diterpenoid alkaloids
4
6
111.4
89.0
7
Table 7
8
104.6
103.9
102.2
104.5
107.9
83.0
104.0
98.3
100.6
106.7
99.6
84.5
101.5
71.5
75.6
98.7
98.1
95.3
AChE inhibitory activities of compounds 3 and 6–24.
9
Compound
IC50
(μM) � S.E.M
Compound
IC50 (μM) � S.E.M
10
11
12
13
14
15
16
17
18
19
3
27.4 � 5.2
17.6 � 2.7
6.3 � 1.6
>100
15
16
17
18
19
20
21
22
23
24
17.8 � 9.2
>50
6
7
2.3 � 0.2
15.8 � 1.7
13.6 � 2.4
21.3 � 9.2
0.3 � 0.02
1.7 � 0.6
9.7 � 3.8
8.9 � 4.9
94.9
8
89.3
9
9.3 � 3.0
20.4 � 7.8
>50
92.8
10
99.0
11
104.2
108.8
12
17.7 � 1.2
8.4 � 0.8
6.4 � 1.2
0.065 � 0.001
13
14
a
Inhibition rate ¼ (A490, control–A490, sample)/(A490, control–A490, blank) � 100;
Cell viability ¼ 100– Inhibition rate.
huperzine A
b
Mean value of three independent determinations.
a The treatments were replicated three times.
7

S. Huang et al.
Phytochemistry 178 (2020) 112459
were isolated from the herb A. anthoroideum for the first time. More than
1500 natural diterpenoid alkaloids have been identified to date (Wang
et al., 2010; McKay et al., 2007). In most cases, C20–diterpenoid alka-
loids contain only common groups, such as OH, OAc, OBz, and OCn
(Wang and Liang, 2002). Compound 1 was the first example of a
diterpenoid alkaloid conjugated to a unique diterpenoid moiety. To the
best of our knowledge, approximately twenty naturally occurring bis-
diterpenoid alkaloids have been reported, and in all the reported bis-
diterpenoid alkaloids, the two diterpenoid alkaloid moieties were linked
with an O–ether linkage (Wang and Liang, 2002; Lin et al., 2009; Zhang
et al., 2013; Cai et al., 2010; Duan et al., 2019; He et al., 2017; Ding
et al., 1992). Compound 2 is the first bisditerpenoid alkaloid joined with
a C–170–C–17 single bond. Interestingly, the racemulosine–type diter-
penoid alkaloid is a rare C20–diterpenoid alkaloid with a unique skel-
eton, and it was proposed to originate from a denudatine–type
diterpenoid alkaloid through double Wanger–Meerwein rearrangements
of rings A and C (Wang and Liang, 2002). To date, racemulosine, isolated
from A. racemulosum Franch var. Pengzhouense, is the only member of
this subclass (Wang et al., 2000). Excitingly, three racemulosine–type
alkaloids in the current study not only enriches the chemical diversity of
diterpenoid alkaloids but also contributes to the development of po-
tential ND therapeutic drugs.
4. Experimental section
4.1. General experimental procedures
Optical rotations were determined using a PerkinElmer polarimeter
�
with a sodium lamp operating at 598 nm and 20 C. IR spectra were
obtained using a Thermo Fisher Nicolet 6700 spectrometer. The NMR
spectra were recorded on a Bruker AV 600 spectrometer. HRESIMS data
were measured using a Q–TOF micro mass spectrometer (Waters). HPLC
separations were carried out on a Waters SymmetryShield™ RP–18
column (5 μm, 4.6 � 250 mm) with a Waters 600 controller and a Waters
2487 detector.
ECD spectra were measured on an Aviv Model 420SF spec-
tropolarimeter (Aviv Biomedical Inc., Lakewood, USA). Silica gel
(Qingdao Haiyang Chemical Co., Ltd., China) and Sephadex LH–20
(Pharmacia Co.) were utilized for column chromatography (CC).
C
20–diterpenoid alkaloids, anthoroidines C–E (3–5), were isolated from
A anthoroideum DC., which may serve as a reliable taxonomic marker of
the subgenus Aconitum. Glucose is ubiquitous in plants; however, only
five diterpene alkaloids containing a glucose moiety have been reported
to date (Zhang et al., 2016; Meng et al., 2017a,b). Compound 4 is the
first diterpenoid alkaloid with a racemulosine skeleton bearing a glucose
isolated from a natural source, providing a new candidate for further
pharmacological investigations. To the best of our knowledge, aconi-
carmicharcutinium hydrochloride is the only arcutine–type
4.2. Plant material
The whole plants of A. anthoroideum DC. (Ranunculaceae) were
collected in August 2016 from Sayram Lake, Bortala, Xinjiang, People’s
Republic of China (GPS coordinates: 44�640N, 81�380E). The plant was
identified by Professor Liangke Song of the School of Life Science and
Engineering, Southwest Jiaotong University, Sichuan, P. R. China, and a
voucher specimen was deposited at the same institution (No.
ZN361520160816).
C
20–diterpenoid alkaloid with an iminium moiety, and its absolute
configuration has been unambiguously established by X–ray crystallo-
graphic analysis (Fig. 5) for the first time in this report. In addition, the
assay of the neuroprotective effects indicated for the first time that some
of these isolated compounds could protect SH–SY5Y cells from
MPPþ–induced apoptosis and increase cell viability. Nominine (18)
showed a highly potent protective effect, with a recovery rate of 34.4%
4.3. Extraction and isolation
Air–dried whole plants of A. anthoroideum DC. (4.8 kg) were
extracted with 95% EtOH five times at room temperature, and each
extraction lasting four days. The solvent was evaporated to afford the
ethanol extract (500 g). The extract was suspended in H2O (2 L) and
adjusted to pH 2–3 with 10% HCl and sequentially extracted with pe-
troleum ether (3 � 3 L) and ethyl acetate (3 � 3 L). The pH of the
aqueous layer was adjusted to 11 with aqueous ammonia solution, and it
was extracted with CH2Cl2 (3 � 3 L). The CH2Cl2 extracts were
concentrated to produce the crude alkaloid extract (18 g). The crude
alkaloid extract was separated by column chromatography over silica
gel using CH2Cl2:MeOH (100:1–1:100) mixtures with increasing polar-
ity, and fractions A–E were obtained based on TLC analysis.
(50 μM). The AChE inhibitory activities of these compounds as well as
five known diterpenoid alkaloids were also assessed. Rotundifosine F
(21) exhibited significantly higher AChE inhibitory activity relative to
those of other alkaloids, suggesting its potential as a new lead compound
for the development of AChE inhibitors. The phytochemical investiga-
tion, bioactivity examination, and SAR analysis of these diterpenoid
Fr. A (3.2 g) was subjected to a silica gel column with light petro-
leum:acetone:Et2NH (25:1:0.1–0:1:0.1) to obtain compounds 3 (20 mg),
5 (2 mg), and 9 (10 mg). Fr. B (2.8 g) was submitted to silica gel CC
eluting with light petroleum:acetone:Et2NH (20:1:0.1–0:1:0.1) to yield
compounds 6 (16 mg), 7 (60 mg), and 14 (15 mg). Fr. C (4 g) was
separated on a silica gel column (light petroleum:acetone:Et2NH,
18:1:0.1–0:1:0.1) to afford subfractions C1 (80 mg), C2 (200 mg), C3
(480 mg), C4 (230 mg), and C5 (150 mg). Subfractions C1 and C2 were
separated
on
silica
gel
(light
petroleum:acetone:Et2NH,
16:1:0.1–0:1:0.1) to afford compounds 8 (11 mg), 13 (9.2 mg) and 15
(15 mg). Compounds 12 (24 mg), 16 (9 mg), 17 (12 mg), and 18 (17 mg)
were obtained by purifying subfraction C3 by silica gel CC (light petro-
leum:acetone:Et2NH, 13:1:0.1–0:1:0.1). Fr. D (2 g) was subjected to
silica gel CC and eluted with CH2Cl2:MeOH (40:1–1:1) to obtain four
subfractions (D1–D3). Subfraction D2 was subjected to Sephadex LH–20
column chromatography (MeOH) to yield compound 9 (8 mg). Sub-
fraction D3 was further purified using an RP–18 silica gel column with
MeOH:H2O (15:85–50:50) as the mobile phase to yield compounds 4 (5
mg) and 16 (45 mg) and a mixture (23 mg) that was further purified over
RP–18 silica gel with MeOH:H2O (3:7) to afford compounds 1 (2 mg)
and 2 (2 mg). Compounds 11 (13 mg), 10 (13 mg), and 19 (19 mg) were
Fig. 5. ORTEP drawing of 19.
8

S. Huang et al.
Phytochemistry 178 (2020) 112459
isolated from fraction E (258 mg) with mixtures of light petroleum:
acetone:Et2NH of 9:1:0.1, 8:1:0.1 and 4:1:0.1, respectively.
4.5. Acid hydrolysis of 4
According to a previously described protocol (Huang et al., 2013),
compound 4 (approximately 2 mg) was dissolved in 1 N HCl-dioxane
(1:1, 2 mL) and stirred at 90 �C for 3 h. The reaction mixture was
diluted with H2O (3 mL), neutralized with 0.5 N NaOH and then
extracted twice with CHCl3. The H2O layer was concentrated under a
stream of nitrogen. Subsequently, 1–(trimethylsilyl) imidazole (0.1 mL)
and pyridine (0.2 mL) were added to the residue, and the solution was
stirred at 60 �C for 20 min. After the solvent was removed, the residue
was partitioned between H2O and CHCl3. The combined organic phase
was dried and analyzed by GC using an L-CP-Chirasil-Val column (0.32
mm � 25 m). D-Glucose was confirmed by comparison of the retention
time with that of an authentic standard (7.65 min).
4.3.1. Anthoroisine A (1)
White powder; [α]20 D ¼ À 4.5 (c 0.35, CHCl3); IR (KBr) vmax: 3393,
3349, 2919, 2850, 1723, 1666, 1460, 1384, 1084, 1047, 884, 600 cmÀ 1
;
1H NMR and 13C NMR data, see Table 1. HRESIMS (m/z): 626.3845 [M þ
H]þ (calcd For C40H52NO5: 626.3845).
4.3.2. Anthoroisine B (2)
White powder; [α]20 D ¼ þ37.2 (c 0.5, CH3OH); IR (KBr) vmax: 3399,
2926, 2870, 1644, 1457, 1368, 1344, 1315, 1180, 1080, 1025 cmÀ 1; 1H
NMR and 13C NMR data, see Table 1. HRESIMS (m/z): 625.3997 [M þ
H]þ (calcd for C40H53N2O4: 625.4005).
4.3.3. Anthoroisine C (3)
4.6. Cell viability assay
White powder; [α]20 D þ27.0 (c 0.5, CHCl3); IR (KBr) vmax 3353,
3074, 2937, 2875, 1716, 1562, 1397, 1255, 1190, 1147, 1038, 999, 755
cm À 1; 1H NMR and 13C NMR data, see Table 2; HRESIMS m/z 416.2432
[M þ H]þ (calcd for C24H34NO5: 416.2437).
SH–SY5Y (human neuroblastoma cell line) cells were obtained from
ATCC, and the cells were maintained in growth medium containing
DMEM high–sugar medium (containing 10% calf serum, 100 KU⋅LÀ 1
penicillin, and 100 mg LÀ 1 streptomycin). The cells were cultured at 37
�C with 5% CO2 (v/v). MTT and MPPþ were purchased from Sigma-
Aldrich (St. Louis, MO, USA). For experiments, SH–SY5Y cells were
4.3.4. Anthoroisine D (4)
White powder; [α]20 D þ10.6 (c 0.5, CHCl3); IR (KBr) vmax 3369,
seeded in 96-well plates at a density of 1 � 105 cells/mL in 200
μL of
2925, 2875, 1727, 1568, 1459, 1416, 1372, 1256, 1073, 1034, 964, 757
cm À 1; 1H NMR and 13C NMR data, see Table 2; HRESIMS m/z 578.2968
[M þ H]þ (calcd for C30H44NO10: 578.2965).
medium. After allowing the cells to attach and reach 70–80% conflu-
ence, they were treated with different concentrations of the test com-
pounds or MPPþ for 24 h. The cell viability was estimated by an MTT
colorimetric assay. Twenty microliters of MTT (5 mg/mL) was added to
each well, and the cells were cultured for 4 h. Subsequently, the medium
was removed, and the formazan crystals were dissolved with DMSO. The
optical densities (OD) at 490 nm were measured on a microplate reader
(TECAN SPECTRA, Wetzlar, Germany) (Wang et al., 2017).
4.3.5. Anthoroisine E (5)
White powder; [
α
]20 D þ50.0 (c 0.1, CHCl3); 1H NMR and 13C NMR
data, see Table 2; HRESIMS m/z 388.2482 [M þ H]þ (calcd for
C
23H34NO4: 388.2488).
4.3.6. Anthoroisine F (6)
White powder; [α]20 D þ68.2 (c 0.45, CHCl3); IR (KBr) vmax 3358,
4.7. Acetylcholinesterase inhibition assay
2961, 2932, 1663, 1460, 1260, 1135, 1093, 897, 803 cm À 1; 1H NMR
and 13C NMR data, see Tables 3 and 4; HRESIMS m/z 314.2114 [M þ
H]þ (calcd for C20H28NO2: 314.2120).
Electric eel acetylcholinesterase (EC 3.1.1.7), acetylthiocholine io-
dide (ATCI), and 5,50-dithiobis-2-nitrobenzoic acid (DTNB) were pur-
chased from Sigma-Aldrich (St. Louis, MO, USA). The AChE inhibitory
activities were assessed using a 96-well microplate assay modified from
Ellman’s method (Liu et al., 2013; Ellman et al., 1961). Briefly, 0.2 mM
4.3.7. Anthoroisine G (7)
White powder; [α]20 D þ19.0 (c 0.50, CHCl3); IR (KBr) vmax 3405,
3071, 2949, 2923, 1717, 1450, 1278, 1448, 1263, 1112, 1071, 748, 714
cm À 1; 1H NMR and 13C NMR data, see Tables 3 and 4; HRESIMS m/z
418.2379 [M þ H]þ (calcd for C27H32NO3: 418.2382).
Ellman’s reagent (DTNB) (150
solution of the test compound (in DMSO, 10
U.I./mL, 10
L) were mixed and incubated for 20 min at 25 �C. Subse-
quently, ATCI in buffer (10 mM, 10
μ
L) in phosphate buffer (0.1 M, pH 8.0), a
μ
L) and AChE in H2O (0.05
μ
μ
L) was added, and the cells were
4.3.8. Anthoroisine H (8)
incubated at 37 �C for 20 min. The absorbance was monitored at 405 nm
using a microplate reader (TECAN SPECTRA, Wetzlar, Germany).
Enzyme inhibitory activity (%) ¼ [1–(Asample/Acontrol)] � 100. The IC50
values were evaluated using the software package Prism V5.0 (Graph-
Pad Software, San Diego, CA, USA).
White powder; [α]20 D þ15.5 (c 0.18, CHCl3); IR (KBr) vmax 3376,
2967, 2920, 1711, 1450, 1285, 1228, 1115, 1026, 753, 714 cm À 1; 1H
NMR and 13C NMR data, see Tables 3 and 4; HRESIMS m/z 434.2326 [M
þ H]þ (calcd for C27H32NO4: 434.2331).
4.3.9. Anthoroisine I (9)
Declaration of competing interest
The authors declare no competing financial interest.
Acknowledgments
White powder; [α]20 D þ9.4 (c 0.75, CHCl3); IR (KBr) vmax 3439,
3166, 2931, 1731, 1654, 1458, 1262, 1242, 1155, 1091, 754 cm À 1; 1H
NMR and 13C NMR data, see Tables 3 and 4; HRESIMS m/z 442.2590 [M
þ H]þ (calcd for C26H36NO5: 442.2593).
4.4. X-ray crystallographic analysis
This research was supported by grants from National Natural Science
Foundation of China (81773605), the Interdisciplinary Frontier Basic
Research Project of SWJTU (2682017QY04), and the Science and
Technology Program of Sichuan, China (2018JY0077).
Crystallographic Data for Anthoroisine
G
(7) and Aconi-
carmicharcutinium A Hydrochloride (19), see the Supporting Informa-
tion. Crystal data for 7 and 19 (deposition numbers: CCDC 1559538 and
1,559,539) have been deposited with the Cambridge Crystallographic
Data Center. These crystallographic data can be obtained free of charge
via www.ccdc.cam.ac.uk/deposit (or from the CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: þ 44 1223336033; deposit@ccdc.cam.ac.
uk).
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.phytochem.2020.112459.
9

S. Huang et al.
Phytochemistry 178 (2020) 112459
Malak, L.G., Ibrahim, M.A., Moharram, A.M., Pandey, P., Tekwani, B., Doerksen, R.J.,
Ferreira, D., Ross, S.A., 2018. Antileishmanial carbasugars from geosmithia
langdonii. J. Nat. Prod. 81, 2222–2227.
References
Ahmad, H., Ahmad, S., Khan, E., Shahzad, A., Ali, M., Tahir, M.N., Shaheen, F.,
Ahmad, M., 2017a. Isolation, crystal structure determination and cholinesterase
inhibitory potential of isotalatizidine hydrate from Delphinium denudatum. Pharm.
Biol. 55, 680–686.
�
McKay, D.B., Chang, C., Gonzalez–Cestari, T.F., McKay, S.B., Ei–Hajj, R.A., Bryant, D.L.,
Zhu, M.X., Swaan, P.W., Arason, K.M., Pulipaka, A.B., Orac, C.M., Bergmeier, S.C.,
2007. Analogs of methyllycaconitine as novel noncompetitive inhibitors of nicotinic
receptors: pharmacological characterization, computational modeling, and
pharmacophore development. Mol. Pharmacol. 71, 1288–1297.
Ahmad, H., Ahmad, S., Shah, S.A.A., Latif, A., Ali, M., Khan, F.A., Tahir, M.N.,
Shaheen, F., Wadood, A., Ahmad, M., 2017b. Antioxidant and anticholinesterase
potential of diterpenoid alkaloids from Aconitum heterophyllum. Bioorg. Med. Chem.
25, 3368–3376.
Meng, X.H., Guo, Q.L., Zhu, C.G., Shi, J.G., 2017a. Unprecedented C19-diterpenoid
alkaloid glycosides from an aqueous extract of "fu zi": neoline 14-O-L-arabinosides
with four isomeric L-anabinosyls. Chin. Chem. Lett. 28, 1705–1710.
Alzheimer’s Association, 2018. Alzheimer’s disease facts and figures. Alzheimer’s
Dementia 14, 367–425.
Meng, X.H., Jiang, Z.B., Guo, Q.L., Shi, J.G., 2017b. A minor arcutine-type C20
-
diterpenoid alkaloid iminium constituent of "fu zi". Chin. Chem. Lett. 28, 588–592.
Pelletier, S.W., Iyer, K.N., Bhalla, V.K., Newton, M.G., Aneja, R., 1970. Crystal structure
of the diterpene alkaloid hetidine. J. Chem. Soc. D 7, 393.
Atta–ur–Rahman, Fatima, N., Akhtar, F., Choudhary, M.I., Khalid, A., 2000. New
norditerpenoid alkaloids from Aconitum falconeri. J. Nat. Prod. 63, 1393–1395.
Bartus, R.T., Dean, R.L., Beer, B., Lippa, A.S., 1982. The cholinergic hypothesis of
geriatric memory dysfunction. Science 217, 408–417.
Pelletier, S.W., Mody, N.V., 1979. An unusual rearrangement of ajaconine: an example of
a "disfavored" 5-endo-trigonal ring closure. J. Am. Chem. Soc. 101, 492–494.
Przedborski, S., Jackson–Lewis, V., 1998. Mechanisms of MPTP toxicity. Mov. Disord. 13,
35–38.
Bessonova, I.A., Saidkhodzhaeva, S.A., 2000. Hetisane-type diterpenoid alkaloids. Chem.
Nat. Compd. 36, 419–477.
Betarbet, R., Sherer, T.B., MacKenzie, G., Garcia–Osuna, M., Panov, A.V., Greenamyre, J.
T., 2000. Chronic systemic pesticide exposure reproduces features of Parkinson’s
disease. Nat. Neurosci. 3, 1301–1306.
Shihabuddin, L.S., Brundin, P., Greenamyre, J.T., Stephenson, D., Sardi, S.P., 2018. New
frontiers in Parkinson’s disease: from genetics to the clinic. J. Neurosci. 38,
9375–9382.
Cai, L., Song, L., Chen, Q.H., Liu, X.Y., Wang, F.P., 2010. Piepunine, A novel bis-
diterpenoid alkaloid from the roots of Aconitum piepunense. Helv. Chim. Acta 93,
2251–2255.
Singer, T.P., Ramsay, R.R., 1990. Mechanism of the neurotoxicity of MPTP an update.
FEBS Lett. 274, 1–8.
Tang, Q.F., Liu, J.H., Xue, J., Ye, W.C., Zhang, Z.J., Yang, C.H., 2008. Preparative
isolation and purification of two new isomeric diterpenoid alkaloids from Aconitum
coreanum by high-speed counter-current chromatography. J. Chromatogr. B 872,
181–185.
Cheng, Z.B., Li, Y.L., Xu, W., Liu, W., Liu, L.J., Zhu, D.G., Kang, Y., Luo, Z.H., Li, Q., 2019.
Three new cyclopiane-type diterpenes from a deep-sea derived fungus penicillium sp.
ypga11 and their effects against human esophageal carcinoma cells. Bioorg. Chem.
91, 103129.
Tarasoff-Conway, J.M., Carare, R.O., Osorio, R.S., Glodzik, L., Butler, T., Fieremans, E.,
Axel, L., Rusinek, H., Nicholson, C., Zlokovic, B.V., Frangione, B., Blennow, K.,
Ding, L.S., Wu, F.E., Chen, Y.Z., 1992. Diterpenoid alkaloids from Aconitum gymnandrum.
Acta Pharm. Sin. 27, 394–396.
�
Menard, J., Zetterberg, H., Wisniewski, T., de Leon, M.J., 2015. Clearance systems in
Duan, X.Y., Zhao, D.K., Shen, Y., 2019. Two new bis-C20-diterpenoid alkaloids with anti-
inflammation activity from Aconitum bulleyanum. J. Asian Nat. Prod. Res. 21,
323–330.
the brain-implications for Alzheimer disease. Nat. Rev. Neurol. 11, 457–470.
Tashkhodzhaev, B., Yusupova, M., Salimov, B.T., Ibragimov, B.T., 1992. Crystal and
molecular structure of the hydrochloride of the diterpene alkaloid hetisine.
J. Inclusion Phenom. Mol. Recognit. Chem. 14, 311–316.
Ellman, G.L., Courtney, K.D., Andres, V., Featherstone, R.M., 1961. A new and rapid
colorimetric determination of acetylcholinesterase activity. Biochem. Pharmacol. 7,
88–95.
Wang, F.P., Chen, Q.H., Liu, X.Y., 2010. Diterpenoid alkaloids. Nat. Prod. Rep. 27,
529–570.
Fan, X.R., Yang, L.H., Liu, Z.H., Lin, L.M., Li, C., Guo, S.S., Wang, Z.M., Wang, Z.J.,
Sui, F., 2019. Diterpenoid alkaloids from the whole plant of Aconitum tanguticum
(Maxim.) Stapf. Phytochemistry 160, 71–77.
Wang, F.P., Liang, X.T., 2002. The Alkaloids: Chemistry and Biology, vol. 59. Elsevier
Science, New York, p. 5.
Wang, F.P., Peng, C.S., Yu, K.B., 2000. Racemulosine, a novel skeletal C20-diterpenoid
alkaloid from Aconitum racemulosum Franch var. pengzhouense. Tetrahedron 56,
7443–7446.
Giacobine, E., Becker, R., 1997. Alzheimer’s Disease: Molecular Biology to Therapy.
€
Birkhauser Boston, Boston, p. 187.
Giacobini, E., 2000. Cholinesterase inhibitors stabilize alzheimer disease. Neurochem.
Wang, F.P., Yan, L.P., 2007. Campylopin from Delphinium campylocentrum, the first
hetidane C20-diterpene, suggests a new alkaloid biogenetic pathway. Tetrahedron
63, 1417–1420.
Res. 25, 1185–1190.
Gonzalez, A.G., De la Fuente, G., Reina, M., Diaz, R., Timon, L., 1986. 13C NMR
�
spectroscopy of some hetisine subtype C20-diterpenoid alkaloids and their
derivatives. Phytochemistry 25, 1971–1973.
Wang, G., Zhu, L.J., Zhao, Y.Q., Gao, S.Y., Sun, D.J., Yuan, J.Q., Yao, X.S., 2017.
A natural product from Cannabis sativa subsp sativa inhibits homeodomain-
interacting protein kinase 2 (HIPK2), attenuating MPPþ-induced apoptosis in human
neuroblastoma SH-SY5Y cells. Bioorg. Chem. 72, 64–73.
He, J.B., Luan, J., Lv, X.M., Rui, D.Y., Tao, J., Wang, B., Niu, Y.F., Ju, H.P., 2017.
Navicularines A-C: new diterpenoid alkaloids from Aconitum naviculare and their
cytotoxic activities. Fitoterapia 120, 142–145.
Wang, W.T., 1979. Flora Reipublicae Popularis Sinicae, vol. 27. Science Press, Beijing,
p. pp323.
He, L., Chen, Y.Z., Ding, L.S., Li, B.G., 1997. Alkaloids from tongolo larkspur (Delphinium
tongolese). Chin. Tradit. Herb. Drugs 28, 392–395.
Wangchuk, P., Bremner, J.B., Samosorn, S., 2007. Hetisine-type diterpenoid alkaloids
from the bhutanese medicinal plant Aconitum orochryseum. J. Nat. Prod. 70,
1808–1811.
Huang, S., Zhou, X.L., Wang, C.J., Wang, Y.S., Xiao, F., Shan, L.H., Guo, Z.Y., Weng, J.,
2013. Pyrrolizidine alkaloids from Liparis nervosa with inhibitory activities against
LPS-induced NO production in RAW 264.7 macrophages. Phytochemistry 93,
154–161.
Yusupova, I.M., Salimov, B.T., Tashkhodzhaev, B., 1992. Tadzhaconine - a new C20
diterpene alkaloid from Aconitum zeravschanicum. Chem. Nat. Compd. 28, 335–339.
Zhang, J.F., Li, Y., Gao, F., Shan, L.H., Zhou, X.L., 2019. Four new C20-diterpenoid
alkaloids from Aconitum rotundifolium. J. Asian Nat. Prod. Res. 21, 716–724.
Zhang, J.F., Dai, R.Y., Shan, L.H., Chen, L., Xu, L., Wu, M.Y., Wang, C.J., Huang, S.,
Zhou, X.L., 2016. Iliensines A and B: two new C19-diterpenoid alkaloids from
Delphinium iliense. Phytochem. Lett. 17, 299–303.
Ibrahim, M.A., Na, M., Oh, J., Schinazi, R.F., McBrayer, T.R., Whitaker, T., Doerksen, R.
J., Newman, D.J., Zachos, L.G., Hamann, M.T., 2013. Significance of endangered and
threatened plant natural products in the control of human disease. Proc. Natl. Acad.
Sci. U. S. A. 110, 16832–16837.
Jiang, B.Y., Lin, S., Zhu, C.G., Wang, S.J., Wang, Y.N., Chen, M.H., Zhang, J.J., Hu, J.F.,
Chen, N.H., Yang, Y.C., Shi, J.G., 2012. Diterpenoid alkaloids from the lateral root of
Aconitum carmichaelii. J. Nat. Prod. 75, 1145–1159.
Zhang, S.S., Ma, Q.Y., Huang, S.Z., Dai, H.F., Guo, Z.K., Yu, Z.F., Zhao, Y.X., 2015.
Lanostanoids with acetylcholinesterase inhibitory activity from the mushroom
Haddowia longipes. Phytochemistry 110, 133–139.
Jiang, Q., Pelletier, S.W., 1991. Two new diterpenoid alkaloids from Aconitum palmatum.
J. Nat. Prod. 54, 525–531.
Zhang, Z.T., Chen, D.L., Chen, Q.H., Wang, F.P., 2013. Bis-diterpenoid alkaloids from
Aconitum tanguticum var. trichocarpum. Helv. Chim. Acta 96, 710–718.
Zhou, X.L., Chen, D.L., Chen, Q.H., Wang, F.P., 2005. C20-diterpenoid alkaloids from
Delphinium trifoliolatum. J. Nat. Prod. 68, 1076–1079.
Lin, L., Chen, D.L., Liu, X.Y., Chen, Q.H., Wang, F.P., Yang, C.Y., 2009. Bis-diterpenoid
alkaloids from Aconitum tanguticum var. trichocarpum. Nat. prod. Commun. 4,
897–901.
Liu, J.Q., Peng, X.R., Li, X.Y., Li, T.Z., Zhang, W.M., Shi, L., Han, J., Qiu, M.H., 2013.
Norfriedelins A–C with acetylcholinesterase inhibitory activity from Acerola Tree
(Malpighia emarginata). Org. Lett. 15, 1580–1583.
10