Erythrina alkaloids from leaves of Erythrina arborescens

Article abstract of DOI:10.1016/j.tet.2019.130515

Continued interest in Erythrina alkaloids resulted in the isolation of 38 alkaloids including 7 undescribed ones from the leaves of Erythrina arborescens Roxburgh. Among the new compounds, erythrivarines H-I were two dimeric alkaloids, while others were Erythrina alkaloid glucosides. Dimeric Erythrina alkaloids and monomers, turcomanidine and isoboldine, showed medium xanthine oxidase inhibition.

Full text of DOI:10.1016/j.tet.2019.130515

Tetrahedron 75 (2019) 130515
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Erythrina alkaloids from leaves of Erythrina arborescens
Bing-Jie Zhang , Wen-Na Xiao , Jing Chen , Mei-Fen Bao a c, Johann Schinnerl d,
Qi Wang
a
a
b
,
1
a
,
b,
**
a, c, *
, Xiang-Hai Cai
State Key Laboratory of Phytochemistry and Plant Resources in West China, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming 650201,
China
b
c
School of Public Health, Kunming Medical University, Kunming 650500, China
Yunnan Key Laboratory of Natural Medicinal Chemistry, Kunming 650201, China
Department of Botany and Biodiversity Research, University of Vienna, Rennweg 14, A-1030 Vienna, Austria
d
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 17 April 2019
Received in revised form
Continued interest in Erythrina alkaloids resulted in the isolation of 38 alkaloids including 7 undescribed
ones from the leaves of Erythrina arborescens Roxburgh. Among the new compounds, erythrivarines H-I
were two dimeric alkaloids, while others were Erythrina alkaloid glucosides. Dimeric Erythrina alkaloids
and monomers, turcomanidine and isoboldine, showed medium xanthine oxidase inhibition.
31 July 2019
Accepted 7 August 2019
Available online 14 August 2019
©
2019 Elsevier Ltd. All rights reserved.
Keywords:
Erythrina arborescens
Erythrivarines H-I
Alkaloid glucosides
Xanthine oxidase inhibitor
1
. Introduction
alkaloid mainly could affect psychosis system, such as anxiolytic-
like activity [3]. In addition, like colchicine derived from phenyl-
alanine, does Erythrina alkaloid possess antigout activity? Our
previous studies on the alkaloids from Yunnan local plant resources
disclosed the first dimeric and trimeric Erythrina alkaloids from the
The genus Erythrina L. (Fabaceae) comprises more than 100
species and is mainly distributed in tropical and subtropical areas.
Among them, there are four local and two introduced species
occurring in China [1]. To date, more than 100 alkaloids are re-
ported from this genus but reports of glycosylated alkaloid deri-
vates are scant. The reported bioactivities and relatively
conservative structures of Erythrina alkaloids had not promoted
themselves to be a phytochemical hot field. Nevertheless, they have
gained recently attention of chemists due to their distinct spi-
rocyclic skeleton [2]. Polymerization, a special form of natural
product genesis, not only diversifies molecules but also affects the
physicochemical and pharmacological properties of compounds. In
addition, glycosidation is another important way to form diverse
natural products. As far as Erythrina alkaloids, there are rarely re-
ported glycosylated alkaloids. The known individual Erythrina
’
follower of Erythrina variegata Linn. [3] [4] Hence, this finding
attracted us to study the closely related species Erythrina arbor-
escens Roxburgh. As a result, 25 alkaloids including eight unde-
scribed ones were obtained from its flowers [5]. As part of our
contained research on E. arborescens, 7 undescribed alkaloids,
named as erythrivarines H-I (1e2), erythraline-11-O-
(3), erythrartine-11-O- -glucose (4), erythraline N-oxide-11-O-
-glucose (5), 10-oxo-erythraline-11-O- -glucose (6), 10-oxo-
erythrartine-11-O- -glucose (7), together with 31 known ones
were described from its leaves in the present work.
b-D-glucose
b
-D
b-
D
b-D
b-D
2. Results and discussion
Compounds 1e7 (Fig. 1) might be alkaloids as they showed
*
Corresponding author. State Key Laboratory of Phytochemistry and Plant Re-
positive reaction with Dragendorff's reagent on TLC plates. The UV
absorption of 1 at 202, 232, and 280 nm indicated a tetrahy-
droisoquinoline chromophore [6]. Furthermore, its IR absorption
sources in West China, Kunming Institute of Botany, Chinese Academy of Sciences,
Kunming 650201, China.
** Corresponding author.
ꢀ
1
bands at 1730, 1489, 1452 cm resulted from the carbonyls and
aromatic rings, which was consistent with the characteristics of
E-mail address: xhcai@mail.kib.ac.cn (X.-H. Cai).
These authors contributed equally.
1
https://doi.org/10.1016/j.tet.2019.130515
040-4020/© 2019 Elsevier Ltd. All rights reserved.
0

2
B.-J. Zhang et al. / Tetrahedron 75 (2019) 130515
Fig. 1. New alkaloids from leaves of Erythrina arborescens.
þ.
Erythrina alkaloid. The HRESIMS (m/z ¼ 661.2881 [M þ Na] ) and
alkaloid. The 13C NMR spectrum (Table 1) of 1 displayed 14 qua-
13
C NMR spectroscopic data (Table 1) of 1 established the molecular
ternary carbons (
132.1, 130.7, 129.9, 127.8, 127.5, 68.1, 65.9), 12 methines (
133.3, 126.3, 125.7, 124.4, 113.6, 113.2, 110.8, 109.8, 77.1, 75.7, 63.4),
six methylenes ( 42.5, 42.0, 41.8, 38.2, 27.2, 24.1) and six methoxyl
groups ( 56.4, 56.4, 56.3, 56.1, 56.0, 55.9). The above mentioned
d
C
171.4, 153.0, 149.8, 149.0, 148.3, 148.1, 142.6,
1
formula C38
H
42
N
2
O
7
. In the H NMR spectrum of 1 (Table 1), six
7.65,
d
C
136.0,
singlets ( 6.94, 6.90, 6.88, 6.70, 5.63, 4.76), four doublets (d
d
H
H
dd, J ¼ 10.8, 2.4 Hz), 6.34 (d, J ¼ 10.8 Hz), 6.59 (dd, J ¼ 10.2, 2.4 Hz),
d
C
6
.11 (d, J ¼ 10.2 Hz), and six methoxyl groups (
d
H
3.83, 3.78, 3.71,
d
C
3
.69, 3.31 ꢁ 2) indicated that 1 might be a dimeric Erythrina
data suggested that 1 was similar to the previously reported
dimeric erythrivarine A [3] except that two methylenedioxys were
in erythrivarine A, instead, four additional methoxyls were present
in 1. Those differences suggested 1 possessed two ortho-methoxy
groups which with further confirmed by the two pair of aromatic
Table 1
1
H (600 MHz) and 13C (150 MHz) NMR data of alkaloids 1 and 2 in acetone-d
6
(d in
ppm, J in Hz).
1
singlets in its H NMR spectrum. The HMBC correlations from two
Entry
d
H
(1)
d
C
(1)
d
H
(2)
d
C
(
d
C
(2))
pairs of methylene protons (
d
H
3.21 and 2.98, and
H
d 2.86 and 2.52)
to C-17/17' (
d
C
113.6 d and 113.2 d) assigned the methylenes to CH -
2
1
2
3
4
7.65, dd (10.8, 2.4)
6.34, d (10.8)
3.85, overlap
124.4 d
136.0 d
75.7 d
42.0 t
7.63, d (9.0)
6.94, d (9.0)
120.6 d
108.4 d
153.6 s
113.6 s
0
11/11 in 1, respectively, rather than methines in the erythrivarine
13
A. Further comparison with the C NMR spectra of erythrivarine A,
the newly carbonyl signal ( 171.4 s) in 1 was assigned to C-8 based
on the HMBC correlations from H-10 to C-5 and C-8. Another
methine proton 4.76 (s) showed the HMBC correlations to C-8 (
171.4 s), C-6 ( 153.0 s), C-6' ( 142.6 s), and C-10' ( 41.8 t),
assigning itself to H-8 and connectivity via C-7/8 between both
units (Fig. 2).
Two typical C-3(3 ) methoxyl groups were supported by the
HMBC correlations of d 3.85 (H-3)/d 65.9 (C-5), 124.4 (C-1), 56.4
H C
2.92, dd (10.8, 4.8)
d
C
1.57, t (10.8)
5
6
7
8
65.9 s
135.6 s
124.8 s
114.6 s
129.7 d
53.5 t
d
C
153.0 s
129.9 s
171.4 s
38.2 t
d
C
d
C
d
C
0
7.08, s
4.34, brs (2H)
1
0
1
3.87, m
.60, m
3.21, overlap
.98, overlap
3
0
1
27.2 t
3.07, t (4.8, 2H)
7.55, s
36.0 t
2
0
0
0
0
1
1
1
1
1
1
3
1
1
1
2
3
4
2
3
4
5
6
7
127.8 s
130.7 s
109.8 d
149.0 s
149.8 s
113.6 d
56.4 q
55.9 q
56.0 q
125.7 d
133.3 d
77.1 d
126.8 s
134.5 s
117.3 d
147.8 s
148.9 s
113.1 d
57.7 q
56.0 q
55.9 q
126.0 d
133.0 d
77.1 d
(OMe), and of
methoxy group at C-3 (3 ) was determined as
d
4.11 (H-3 )/
d
C
68.1 (C-5 ),125.7 (C-1 ), and 56.4. The
H
a
-oriented through
6.94, s
0
0
the ROESY correlations of H-3 (3 )/H-14 (14 ), and the large
coupling constants of H-4/4'. Though there were no NOEs with H-
0 0
6.90, s
6.86, s
8 , H-8 was deduced to be a-configuration which was its preferred
-OCH
3
3.31, s (3H)
3.83, s (3H)
3.78, s (3H)
6.59, dd (10.2, 2.4)
6.11, d (10.2)
4.11, m
3.85, s (3H)
3.86, s (3H)
3.84, s (3H)
6.64, d (10.2)
6.13, d (10.2)
4.17, m
conformation in combination with the X-ray diffraction of unit,
5-OCH
6-OCH
3
3
25
erythrinine [3]. The optical rotation of alkaloid 1 (½aꢂ ꢀ160.9), was
D
25
D
3
0
closed to that of erythrivarine A (½aꢂ ꢀ165) , indicating the iden-
0
0
0
tical stereo-configuration. In addition, the configuration of C-5 was
2.63, dd (10.8, 6.0)
42.5 t
2.49, dd (12.0, 6.0)
1.95, t (12.0)
4 42.7 t
1.92, t (10.8)
0
0
0
0
5
6
7
8
1
68.1 s
67.7 s
142.6 s
126.3 d
63.4 d
41.8 t
142.4 s
128.9 d
63.4 d
41.1 t
5.63, s
4.76, s
3.30, overlap
5.69, s
4.97, s
3.25, m
0
0
2
.75, m
2.86, overlap
.52, m
2.92, overlap
3.05, m
2.50, dt (16.2, 3.7)
0
1
1
24.1 t
23.9 t
2
0
0
0
0
0
0
1
1
1
1
1
1
3
1
1
2
3
4
5
127.5 s
132.1 s
110.8 d
148.1 s
148.3 s
113.2 d
56.4 q
56.3 q
56.1 q
127.6 s
132.8 s
110.9 d
148.2 s
149.2 s
113.2 d
56.3 q
56.1 q
56.0 q
6.88, s
6.96, s
6
7
6.70, s
6.75, s
0
-OCH
3
3.31, s (3H)
3.71, s (3H)
3.69, s (3H)
3.35, s (3H)
3.72, s (3H)
3.83, s (3H)
0
5 -OCH
3
0
6 -OCH
3
Fig. 2. The key HMBC correlations of alkaloids 1 and 2.

B.-J. Zhang et al. / Tetrahedron 75 (2019) 130515
3
þ.
S in all reported Erythrina alkaloids. Besides that, the absolute
The HR-ESI-MS of 3 at m/z ¼ 498.1733 [MþH] ) in combination
1 13
0
0
0
configuration of 1 was determined to be 3(3 )R,5R,5 S,8 S. All signals
with its H and C NMR spectra gave a molecular formula
1
13
1
of H and C NMR were assigned by HSQC, HMBC spectra.
Alkaloid 2 was obtained as a yellow powder, which was
consistent with the UV absorptions at 312 and 438 nm in its UV
spectrum. The similar UV spectra of 2 to the previous reported
erythrivarine B3 indicated a similar conjugated system for both
C
24
H
29NO
9
. Besides a sugar unit, its H NMR spectrum (Table 2)
indicated 3 singlets at 7.21, 6.61 and 5.75; two sp [2] doublets
6.55 (dd, J ¼ 10.2, 1.2 Hz), 5.99 (d, J ¼ 10.2 Hz), one methoxy 3.22
(3H, s), one methylenedioxy at 5.97 and 5.95 (each 1H, s), similar
to those of erythrinine [see supporting information]. The C NMR
spectrum of 3 (Table 3) displayed six quaternary carbons, seven
methines, three methylenes, very same to erythrinine. In the HMBC
d
H
d
H
d
H
d
H
13
compounds. The HRESIMS (m/z 621.2955) of 2 suggested the mo-
þ
lecular formula as C38
40
H N
2
O
6
[MþH] , an additional degree of
13
unsaturation than 1. Furthermore, compared to the C NMR data of
erythrivarine B, two methylenedioxy groups in erythrivarine B
spectrum, correlations from H-11 (
d
H
4.59, overlap) to
d
C
104.3 (d,
3
0
C-1 ) and 108.1 (d, C-17) confirmed glucosyl moiety connected to C-
were substituted by four additional methoxyl groups in 2, which
were confirmed by the HMBC crosspeaks. Additionally, the CH-11/
11. The ROESY correlations of
4.59 (H-11)/ 2.35 (H-4) placed 3-OMe and H-11 at
Molecular formula of alkaloid 4 was determined to C25H33NO
9
d
H
3.78 (H-3)/
d
H
6.61 (H-14) and of
d
H
d
H
a-orientation.
1
1' (ab.
d
C
70) in erythrivarine B were substituted by two methylene
0
þ.
ꢃ
groups in 2, suggesting the absence of hydroxyl groups at C-11/11
through its HR-ESI-MS (m/z 492.2226 [MþH] ) with 10 of unsa-
1 13
in 2. This presumption was supported by the HMBC correlations
turation. The H and C NMR (Tables 2 and 3) were very close to
those of 3, except the two additional methoxyls 3.80 (3H, s) and
3.70 (3H, s). The HMBC correlations of 3.80 (3H, s)/ 149.6 (s, C-
15) and of 3.70 (3H, s)/ 149.0 (s, C-16) supported additional
methoxyls at C-15/16 in 4 rather than the methylenedioxy in 3.
0
from
2
d
H
6.86 (H-17) to
d
C
36.0 (C-11) and from
d
H
6.75 (H-17 ) to
d
C
d
H
0
3.9 (C-11 ). Its stereo-configuration of 2 was identical to that of
d
H
d
C
erythrivarine B based on their biosynthesis and close optical rota-
tions. The co-occurrence of 1 and 2 further the supported pre-
sumption of structure and biosynthetic relationship from
erythrivarine A to erythrivarine B as stated in the previous study
d
H
d
C
Molecular formula of alkaloid 5 was determined to C24
H
29NO10
þ
by HR-ESI-MS (m/z ¼ 514.1685 [MþNa] ), showing one additional
1
13
[
3]. Thus 1 and 2 were subsequently named as erythrivarines H and
oxygen atom than 3. The H and C NMR (Tables 2 and 3) shift
values of 5 were very similar to those of 3, with exception for three
I because of the previously reports of the dimeric and trimeric
erythrivarines A-G [3,4].
13
downfield signals (
C
d 72.9, 63.3, 81.0) in the C NMR spectrum. The
Compounds 3e7 were obtained as white powders. Their UV and
IR spectra showed characteristic of Erythrina alkaloids. Their H and
differences between 3 and 5 suggested 5 was N-oxide derivative of
1
3. The molecular formula C24
H
27NO10 of 6 was elucidated on the
1
3
þ
C NMR displayed a
at
4.5e4.9 (d, J ¼ ab. 8.0 Hz) and an anomeric carbon [ab.
d)], a methylene at ab. 62, and the four methine signals between
69 and 80. The identification of the sugar residues were
continued by hydrolysis with 10% HCl to afford -glucose which
were confirmed by comparison with determination of their optical
b-glucose unit on the basis of each proton signal
basis of the HRESIMS m/z 512.1531 ([M þ Na] ). Compared to that of
13
d
H
d
C
104
3, the C NMR spectrum of 6 indicated an additional carbonyl
(
d
d
C
signal (
correlations from
d
C
172.7), instead of one of the methylenes in 6. The HMBC
0
C
d
C
d
H
5.75 (H-11) to
d
C
133.0 (C-13), 103.2 (C-1 ) and
D
172.7 placed the carbonyl signal at C-10. Alkaloid 7 possessed the
molecular formula of C25
528.1844 (M þ Na] ). Comparison of the NMR spectra obtained
H
31NO10 based on the HRESIMS m/z
2
1
ꢃ
ꢃ
ꢃ
ꢃ
ꢃ
þ.
rotation values ([
a
]
D
¼ þ46.5 , þ48.0 , þ47.4 , þ54.9 , þ46.2 ) [7].
Table 2
1
H (600 MHz) NMR data of alkaloids 3e7 (
d
in ppm, J in Hz).
(4)
Entry
d
H
(3)
d
H
d
H
(5)
d
H
(6)
H
d (7)
1
2
3
4
6.55, dd (10.2, 1.2)
5.99, d (10.2)
3.78, m
6.58, dd (10.2, 2.4)
6.03, d (10.2)
4.04, m
2.42, dd (11.4, 5.4)
1.69, t (11.4)
5.70, s
3.87, d (15.6)
3.68, overlap
3.92, dd (12.0, 3.0)
3.72, overlap
4.67, overlap
6.85, s
6.64, dd, 10.2, 1.8
6.12, d, 10.2
4.07, m
2.92, overlap
1.92, overlap
5.82, s
4.87, overlap
3.95, dd (15.6, 3.0)
4.20, dd (13.0, 2.0)
3.75, overlap
4.90, overlap
6.54, s
6.74, (d 10.2)
6.04, d (10.2)
3.70, m
2.67, dd (11.4, 5.4)
1.87, t (11.4)
5.92, s
6.76, dd (10.2, 1.8)
6.06, d (10.2)
3.72, m
2.69, dd (11.4, 4.8)
1.88, t (11.4)
5.90, s
2.35, dd (12.0, 5.4)
1.53, t (12.0)
7
8
5.75, s
3.67, d (15.6)
4.33, d (10.2, 2H)
4.33, d (10.2, 2H)
3
.58, d (15.6)
3.44, overlap
.95, dd (13.2, 4.8)
10
2
1
1
1
3
1
1
1
4
7
4.59, overlap
6.61, s
7.21, s
5.75, s
6.91, s
7.35, s
3.18, s (3H)
5.74, s
7.03, s
7.56, s
3.24, s (3H)
3.74, s (3H)
3.84, s (3H)
7.20, s
7.03, s
3.29, s (3H)
-OCH
5-OCH
6-OCH
3
3.22, s (3H)
3.29, s (3H)
3.70, s (3H)
3.80, s (3H)
3
3
OCH
2
O
5.97, s
6.06, s
6.02, s
5
.95, s
6.02, s
6.00, s
4.85, d (8.4)
3.46, m
3.48, m
3.44, m
3.43, m
3.85, d (10.8)
3.70, m
0
0
0
0
0
0
1
2
4.47, d (8.4)
3.18, overlap
2.99, m
3.05, m
3.16, overlap
3.71, m
4.69, overlap
3.40, overlap
3.27, overlap
3.27, overlap
3.28, overlap
3.48, m
3.39, overlap
3.54, d (4.8)
3.52, d (4.2)
3.30, d (3.0)
3.33, d (3.6)
4.57, d (7.8)
3.32, overlap
2.95, overlap
3.04, m
3.23, overlap
3.74, overlap
3.45, overlap
3.73, overlap
3.31, overlap
3.21, overlap
3.30, overlap
4.85, d (7.2)
3.48, m
3
3.50, m
4
3.51, m
5
3.49, m
6
3.85, overlap
3.70, overlap
3.47, overlap
0
2
3
4
6
-OH
-OH
-OH
-OH
5.25, d (5.4)
5.02, d (4.8)
4.56, overlap
4.98, d (5.4)
0
0
0
6 6
Alkaloids 3 and 5 were recorded in DMSO‑d , 4, 6 and 7 in acetone-d .

4
B.-J. Zhang et al. / Tetrahedron 75 (2019) 130515
Table 3
inhibitory activities, with allopurinol as the positive control. The
obtained IC50 values inhibition activity of 1e2, 21, 26, 37e38 were
given in Table 4.
1
3C (150 MHz) NMR data of alkaloids 3e7 (
d
in ppm, J in Hz).
(5)
Entry
d
C
(3)
d
C
(4)
d
C
d
C
(6)
d
C
(7)
1
2
3
4
5
6
7
8
1
1
1
1
1
1
1
1
3
1
1
124.9 d
131.3 d
75.6 d
41.3 t
126.0 d
132.4 d
77.0 d
41.8 t
125.5 d
132.5 d
75.0 d
30.0 t
125.0 d
132.3 d
76.9 d
41.1 t
72.0 s
139.6 s
121.2 d
55.0 t
172.7 s
71.0 d
129.6 s
133.0 s
104.4 d
147.7 s
148.2 s
107.6 d
56.3 q
125.0 d
132.5 d
77.1 d
41.1 t
3. Experimental section
3.1. General information
66.4 s
66.9 s
81.0 s
72.0 s
141.6 s
123.7 d
59.1 t
143.2 s
124.3 d
59.1 t
137.9 s
119.9 d
72.9 t
139.7 s
121.0 d
54.9 t
Optical rotations were measured with either a Horiba SEPA-300
polarimeter (Horiba Scientific, Kyoto, Japan) or JASCO DIP-370
digital polarimeter (Jasco International Co., Tokyo, Japan). UV
spectra were obtained using a Shimadzu UV-2401A spectropho-
tometer (Shimadzu Corp., Kyoto, Japan). Scanning IR spectroscopy
was performed on a Tenor 27 spectrophotometer using KBr pellets.
MS data were recorded on an Agilent G6230 TOF MS (Applied
Biosystems, Ltd., Warrington, UK). 1D- and 2D- NMR spectra were
obtained on Bruker Avance III-600, DRX-500, and AM-400 spec-
trometers (Bruker BioSpin GmBH, Rheinstetten, Germany) using
TMS as an internal standard. Column chromatography (CC) was
performed on silica gel (200e300 mesh, Qing-dao Haiyang
0
1
2
3
4
5
6
7
50.4 t
50.2 t
63.3 t
172.6 s
71.8 d
127.6 s
131.3 s
108.3 d
148.9 s
149.9 s
110.7 d
56.3 q
56.3 q
56.2 q
72.7 d
129.0 s
131.5 s
104.6 d
146.5 s
145.9 s
108.1 d
55.7 q
74.0 d
127.7 s
131.5 s
109.7 d
149.6 s
149.0 s
113.2 d
56.2 q
56.0 q
56.0 q
73.8 d
124.6 s
129.4 s
104.3 d
147.2 s
148.1 s
109.0 d
55.9 q
-OCH
5-OCH
6-OCH
3
3
3
OCH
1
2
3
4
5
6
2
O
100.8 t
104.3 d
77.0 d
73.9 d
70.2 d
76.9 d
61.3 t
101.7 t
104.3 d
76.9 d
73.8 d
70.1 d
77.3 d
61.3 t
102.6 t
103.2 d
78.6 d
77.3 d
72.9 d
70.7 d
62.6 t
0
0
0
0
0
0
105.6 d
78.0 d
75.2 d
71.6 d
77.6 d
63.0 t
103.6 d
78.4 d
77.5 d
73.3 d
71.0 d
62.5 t
Chemical Co., Ltd, Qingdao, China) and C18-silica gel (20e45 mm,
Fuji Silysia Chemical Ltd.). Fractions were analyzed by TLC on silica
gel plates (GF254, Qingdao Haiyang Chemical Co., Ltd.) and spots
visualized with Dragendorff's reagent. Medium pressure liquid
chromatography (MPLC) was employed using a Buchi pump system
coupled with C18-silica gel-packed glass column (15 ꢁ 230 and
6 6
Alkaloids 3 and 5 were recorded in DMSO‑d , 4, 6 and 7 in acetone-d .
2
6 ꢁ 460 mm). High performance liquid chromatography (HPLC)
was performed using a Waters 600 pump (Waters Corp., Milford,
MA, USA) coupled with Sunfire analytical, or preparative Sunfire,
Xbridge, and Cosmosil C18 columns (150 ꢁ 4.6, and 250 ꢁ 19(20)
mm, respectively). The HPLC analyses were performed on a e
from 7 and 6 showed that the differences between them were
caused by a methoxyl group in 7, instead the methylenedioxy in 6.
Due to the structure similarities, 3e7 were named as erythra-
1525 EF Waters instrument coupled with 2998 photodiode array
line-11-O-
erythrartine N-oxide-11-O-
O- -glucose (6) and 10-oxo-erythrartine-11-O-
respectively.
b
-D
-glucose (3), erythrartine-11-O-
-glucose (5), 10-oxo-erythraline-11-
-glucose (7),
b-D-glucose (4),
detector and a Waters fraction collector II (Waters Corp.).
b-D
b
-
D
b-D
3.2. Plant material
The other known compounds were determined as 10-oxo-
erythraline (8), erymelanthine (9), erytharbine (10), 8-oxo-eryth-
raline (11), 10, 11-dioxoerysotrine (12), 8-oxo-erythrinine (13), 10-
hydroxy-11-oxoerysotrine (14), erysotramidine (15), crystamidine
Leaves of Erythrina arborescens Roxburgh were collected in
September 2014 in Yunnan Province, P. R. China, and identified by
Dr. Chun-Xia Zeng. A voucher specimen (No. Cai20140907) was
deposited in the State Key Laboratory of Phytochemistry and Plant
Resources in West China, Kunming Institute of Botany, Chinese
Academy of Sciences.
(
16), 10,11-dioxoerythraline (17), norreticuline (18), erybidine (19),
erysothrine (20), turcomanidine (21), erythraline (22), cristanine A
23), erythrinine (24), 11-hydroxyerysotrine (25), isoboldine (26),
erythrivarine B (27), erythriarborine B (28), -erythroidine (29),
-hydroxyerythratidine (30), erythrinine N-oxide (31), 8-oxo-
-methoxyerythraline (32), erythrartine N-oxide (33), 11
-methoxyerysodine (35),
-methoxyerysotrine (36), erytharborine A (37), erytharborine B
38) based on their NMR spectra and MS data.
All alkaloids were evaluated for their cytotoxicity against breast
cancer (MCF-7), colon cancer (SW480), and HeLa cells using the
MTT method. However, they did not show activities (IC50 > 20 g/
(
b
3
.3. Extraction and isolation
1
1
1
1
b
b
b-
Dried leaves of E. arborescens (19 kg) were powdered and
hydroxyerysotramidine (34), 8-oxo-11
11b
(
b
extracted three times with MeOH at room temperature. After
removing the solvent, the residue was dissolved in 0.5% HCl soln.
and filtered. The acidic soln. was washed with EtOAc three times.
The aqueous layer was then adjusted to pH 7e8 with NH
subsequently extracted with EtOAc to obtain crude alkaloid extract
85 g). The extract was subjected to column chromatography (CC)
over silica gel and eluted with gradient CHCl /MeOH (1:0e5:1) to
afford nine fractions (IꢀIX). Base water layer was loaded on D101
resin column eluded with H O, then MeOH. Concentrated MeOH
3 2
$H O and
m
(
mL). Alkaloids 1e38 were screened for the xanthine oxidase (XO)
3
2
Table 4
washer (200 g) was subjected to Silica gel column and CH
acetone (from 1:0 to 0:1), to give six Fractions (Fr. X-XV).
3
Cl-
Xanthine oxidase inhibition activities of 1e2, 21, 26, 37
and 38.
Fraction I (4.5 g) was further chromatographed on a C18 MPLC
column eluted with a gradient of MeOH-H O(30%e80%)to give 3
subfraction I-1~I-3. Alkaloid 12 (30 mg) was crystalized from Sub-
Compound
IC50 (mg/mL)
2
1
2
5.3
4.6
2.7
2.4
4.1
4.2
0.60
fraction I-1. Mother liquid of I-1 (0.5 g) was separated by C18 MPLC
2
2
3
3
1
6
7
8
column, eluting with MeOH-H
1a~I-1c. I-1a was purified by C18 Xbridge HPLC with MeOH-H
45%e60%) to obtain 14 (1 mg). I-1b was purified on Sunfire C18
HPLC (5 O (25%e40%) to give 17
m, 20 ꢁ 250 mm) with MeOH-H
2
O (30%e80%) to yield 3 mixtures I-
2
O
(
Allopurinol
m
2

B.-J. Zhang et al. / Tetrahedron 75 (2019) 130515
5
(
3 mg). I-1c was purified by the same column with acetonitrile
32 (35 mg). Fr. XIII (12 g) was separated by C18 MPLC column with
MeOH-H O (50e80%) to give 4 subfractions XIII-1~XIII-4. XIII-1 was
purified by Sunfire HPLC column with 50% MeOH-H O to give 13
(20 mg) and 32 (7 mg). XIII-3(74 mg) was separated by HPLC col-
umn MeOH-H O (57e67%) to obtain 12 (7 mg) and 20 (6 mg). XIII-4
(74 mg) was separated by Sunfire HPLC column with MeOH-H
(55e65%) to yield 6 (8 mg) and 25 (21 mg). Fr. XIIII (11 g) was
separated by C18 MPLC column with MeOH-H O (50e80%). XIIII-1
(474 mg) was separated by HPLC column MeOH-H O (50e60%) to
give 27 (5 mg). Alkaloid 24 (21 mg) was crystalized from XIIII-2.
XIIII-2 (121 mg) was separated by HPLC column MeOH-H
(50e60%) to give 34 (13 mg) and 10 (16 mg). Fr. XV (9 g) was
separated by C18 MPLC column with MeOH-H O (50e70%) to pro-
duce 3 subfractions XV-1e3. SubFr. XV-1 (160 mg) was purified by
Sunfire HPLC (MeOH-H O:40e55%) to obtain 35 (19 mg) and 36
(26 mg). SubFr. XV-2 (130 mg) was separated by HPLC (MeOH-
O:50e60%) to give 7 (11 mg) and 17 (7 mg). SubFr. XV-3 (135 mg)
was purified by Sunfire HPLC (MeOH-H O: 50e60%) to give 29
(9 mg) and 30 (7 mg). Fr. IXX (7 g) was subjected to silica gel CC
with CH Cl-MeOH (9:1e4:1) to obtain two subfractions. IXX-1
(190 mg) was purified by Sunfire HPLC with MeOH-H O (45e60%)
to yield 22 (9 mg). Finally, IXX-2 was purified on same column with
MeOH-H O (50e55%) to give 32 (11 mg).
eH O (15%e40%) to give 15 (116 mg). I-2 (0.8 g) was further puri-
fied on the C18 MPLC column with a gradient flow from 20% to 80%
aqueous methanol to give I-2a and I-2b. I-2a was separated on a
preparative Sunfire column with a gradient of MeOH-H
2
2
2
2
O (45%e
2
6
0%) to afford 10 (46 mg) and 11 (45 mg). Same method was used to
2
O
2
purify I-2b with acetonitrileeH O (30%e45%) to obtain 15 (152 mg).
I-3 (0.5 g) was loaded on the C18 MPLC column with a gradient flow
from 20% to 80% aqueous methanol to give (20%e80%) to give I-3a
and I-3b. I-3a was separated by preparative Xbridge C18 column
eluted with aqueous methanol (45%e60%) to yield 8 (28 mg). I-3b
was separated on a preparative Sunfire C18 column with aqueous
methanol (45%e60%) to get 16 (15 mg). Fraction II (2.2 g) was pu-
rified on a C18 silica gel CC with aqueous methanol (30%e80%, v/v)
to afford three subfractions (II-1~II-3). Subfraction II-1 (0.5 g) was
2
2
2
O
2
2
separated by C18 MPLC column with a gradient of MeOH-H
2
O (15%e
0%) to II-1a and II-1b. II-1a was purified by Cosmosil C18 (5 m,
0 ꢁ 250 mm) HPLC column with a gradient of MeOH-H O (55%e
0%) to afford 37 (3 mg). II-1b was separated using C18 MPLC col-
O (10%e50%) to give 38 (2 mg). II-2
0.2 g) was separated by Sephadex LH-20, eluting with MeOH-
8
2
7
m
H
2
2
2
umn using acetonitrileeH
(
2
3
2
H
2
O (10%e40%) to yield II-2a and II-2b. II-2a was purified on Cos-
O (50%e75%) to give 9 (1 mg). II-3
0.8 mg) was separated by RP-18 with acetonitrile eH O (10%e
0%), and then purified on Xbridge C18 with acetonitrileeH
mosil C18 column, MeOH-H
(
6
2
2
2
2
O
3.4. Spectroscopic data
(
50%e65%) to obtain 1 (3 mg). Fraction III (5.0 g) was subject to C18
silica gel CC with MeOH-H O (20%e80%) to give III-1~III-2. Alkaloid
3 (93 mg) was crystalized from III-1. Fraction IV (5.0 g) was further
separated by C18 MPLC column with a gradient of MeOH-H O (20%e
0%) to give four subfractions IV-1~IV-4. IV-1 (0.5 g) separated by
18 MPLC column with a gradient of MeOH-H O (25%e40%) to give
2
2
Erythrivarine H (1): C38
H
42
N
2
O
7
; white oil; ½aꢂ 5-160.9 (c 0.10,
D
1
MeOH); UV(MeOH)
nm; IR(KBr):
l
max (logε) ¼ 202 (4.03), 232 (3.45), 280 (3.09)
ꢀ
1
ꢀ1
; H
1
2
n
max cm : 2928, 1730, 1615, 1489, 1452 cm
13
8
C
(600 MHz) and C (150 MHz) NMR spectroscopic data, see Table 1;
þ
2
positive HRESIMS m/z
Na, 661.2884).
Erythrivarine I (2): yellow powder; [
(MeOH) max (log ε) 203 (4.05), 233 (3.83), 297 (3.28), 312 (3.30)
and 438 (2.73) nm; IR (KBr) max 2932, 1629, 1503, 1482, 1235, 1098,
¼
661.2881 [M
þ
Na] (calc. for
1
8 (6 mg). Subfraction IV-2 (0.2 g) was loaded on the Sephadex LH-
0 column with MeOH-H O (30%e50%) to give 19 (20 mg). Sub-
fraction IV-3 (0.5 g) was separated by C18 MPLC column with a
gradient of MeOH-H O (20%e80%) to give IV-3a and IV-3b. IV-3a
was purified on Sunfire C18 with a gradient of MeOH-H O (40%e
0%) to yield 4 (38 mg). Subfraction IV-3b was purified with same
method to give 5 (2 mg). IV-4 was purified by same column with
MeOH-H O (45%e55%) to give 3 (1.9 mg). Fraction V (7.1 g) was
subjected to C18 MPLC column with a gradient of MeOH-H O (20%e
0%) to give 3 Subfraction V-1~V-3. Subfraction V-1 (0.2 g) was
38 42 2 7
C H N O
2
0
2
2
D
a] -138 (c 0.2, MeOH); UV
l
2
n
ꢀ
1
1
13
2
1038 cm
; H (600 MHz) and C (150 MHz) NMR spectroscopic
5
data (acetone-d
6
), see Table 1; positive HRESIMS m/z ¼ 621.2955
, 621.2959).
-glucose (3): C24
½aꢂ þ78.2 (c 0.10, MeOH); UV(MeOH) max (logε) 202 (4.06), 232
þ
[M þ H] (calc. for C38
H40NO
6
2
Erythraline 11-O-
b
-
D
9
H29NO ; white powder;
2
5
2
D
l
ꢀ1
8
(3.55), 286 (3.12) nm; IR(KBr) nmax cm : 3382, 3285, 3050, 1605,
ꢀ1
1
13
subjected to C18 MPLC column again with MeOH-H
and then purified by Sunfire C18 column with MeOH-H
6
2
O (10%e80%)
O (45%e
0%) to get 23 (10 mg). Subfraction V-2 (2.5 g) subjected to C18
1479, 1262 cm
; H (600 MHz) and C (150 MHz) NMR spectro-
2
scopic data (DMSO‑d
6
), see Tables 2 and 3, respectively; positive
þ
HR-ESI-MS m/z ¼ 498.1733 (calc. for
C
24
H
29NO
9
[MþNa] ,
MPLC column again with acetonitrileeH
2
O (20%e80%) give 20
498.1735).
(
700 mg). Subfraction V-3 (0.8 g) was separated on C18 MPLC col-
Erythrartine 11-O-
b
-
D
-glucose (4): C25
9
H33NO ; white powder;
2
5
umn again with MeOH-H
2
O (30%e80%) to give V-3a and V-3b. V-3a
½aꢂ þ89.6 (c 0.10, MeOH); UV(MeOH) max (logε) ¼ 204 (4.13), 235
l
D
ꢀ1
(
(
0.3 g) was separated on the C18 MPLC column with MeOH-H
10%e60%) to afford 22 (300 mg). Fraction VI (3.0 g) was separated
O (20%e80%) to give 21
42 mg). Fraction VII (8.0 g) was separated on C18 MPLC column
2
O
(3.68), 285 (3.12) nm; IR(KBr):
n
max cm : 3251, 2930, 1472,
ꢀ1
1
13
1424 cm
data (acetone-d
; H (600 MHz) and C (150 MHz) NMR spectroscopic
on C18 MPLC column with MeOH-H
(
2
6
), see Tables 2 and 3, respectively; positive HR-ESI-
þ
MS m/z ¼ 492.2226 (calc. for C25
H
33NO
9
[MþH] , 492.2228).
with MeOH-H
2
O (20%e80%) to give subfraction VII-1 and VII-2.
Erythrartine N-oxide-11-O-
b-
D-glucose (5): C24
H29NO10; col-
2
5
Subfraction VII-1 was purified on Xbridge C18 column with
MeOH-H O (45%e60%) to give 25 (50 mg). Alkaloid 24 (600 mg)
was crystalized from VII-2. Fraction VIII (0.4 g) was separated on
Sephadex LH-20 column with MeOH-H O (30%e50%) to give 19
1 mg). Fraction IX (4.8 g) was separated on C18 MPLC column with
ourless oil; ½aꢂ þ120.6 (c 0.10, MeOH); UV(MeOH)
l
max
D
ꢀ1
2
(logε) ¼ 206 (4.17), 241 (3.53), 288 (3.09) nm; IR(KBr):
n
max cm
:
ꢀ1
1
13
3312, 2952, 1492, 1263 cm
;
H (600 MHz) and C (150 MHz)
), see Tables 2 and 3, respec-
2
NMR spectroscopic data (DMSO‑d
6
(
tively; positive HR-ESI-MS m/z ¼ 514.1685 (calc. for C24
H29NO10Na
þ
MeOH-H
MeOH-H
2
O (10%e80%), then purified on Xbridge C18 column with
O (45%e60%) to give 26 (8 mg).
[MþNa] , 514.1684).
2
0
2
10-Oxo-erythraline-11-O-
b
-D
-glucose (6): white powder; [
a]
D
Fr. X (10.0 g) separated on C18 MPLC column with MeOH-H
70%~85%) get 31 (17 mg), 9 (10 mg) and 2 (9 mg). Fr. XI (9 g) was
separated by C18 MPLC column with 70%, 75% and 80% aqueous
2
O
167 (c 0.1, MeOH); UV (MeOH)
l
max (log ε) 204 (4.05), 247 (3.89) and
ꢀ
1 1
(
289 (3.18) nm; IR (KBr)
n
max 1648, 1503, 1448 cm ; H (600 MHz)
13
and C NMR (150 MHz) spectroscopic data (acetone-d ), see
6
MeOH to yield 23 (10 mg). Fr. XII (5.2 g) was separated by C18 MPLC
Tables 2 and 3, respectively; positive HRESIMS m/z ¼ 512.1531 [M þ
þ
column with MeOH-H
2
O (60e80%) then purified by Sunfire HPLC
O (65e73%) to obtain 13 (420 mg) and
Na] (calc. for C22
H28NO
5
Na, 512.1532).
C18 Column with MeOH-H
2
10-Oxo-erythrartine 11-O-b-D-glucose (7): white powder;

6
B.-J. Zhang et al. / Tetrahedron 75 (2019) 130515
2
0
[
(
1
a
]
D
141 (c 0.1, MeOH); UV (MeOH)
l
max (log ε) 204 (4.04), 249
max 3324, 2928, 1648, 1502,
H (600 MHz) and C NMR (150 MHz) spectroscopic
solution (pH ¼ 7.50, 0.2 mM). Alkaloids were dissolved in DMSO
3.88) and 289 (3.18) nm; IR (KBr)
443 cm
n
and immediately diluted with phosphate buffer solution to 0.5 mg/
ꢀ1
1
13
ꢃ
;
mL. The mixture (total 100
m
L) was incubated at 37 C. The uric acid
data (acetone-d
6
), see Tables 2 and 3, respectively; positive HRE-
production was calculated from the differential absorbance with a
blank solution in which the xanthine oxidase was replaced by
buffer solution. A test mixture containing without any alkaloid was
prepared to measure the total uric acid production. Different con-
centrations of alkaloids were analyzed, and then the half-maximal
inhibitory concentration (IC50) was calculated by linear regression
analysis. Different concentrations of allopurinol were measured in
triplicate.
þ
SIMS m/z ¼ 528.1844 [M þ Na] (calc. for C18
3
H16NO Na, 528.1843).
Acid hydrolysis of 3e7. Compounds 3e7 (6 mg each) were
ꢃ
refluxed with 10% HCl-MeOH (20 mL) at 80 C for 6 h. After cooling,
the reaction mixture was evaporated to dryness and partitioned
with EtOAc. The sugars were identified as glucose by TLC compar-
ison using MeCOEt-isoPrOH-Me
of the H O layer was performed by preparative TLC eluted four
times with CHCl -MeOH-H O (70:30:1) to afford -glucose (R 0.50)
with positive values of specific rotation, respectively.
2 2
CO-H O (20:10:7:6). Purification
2
3
2
D
f
Acknowledgments
3
.5. Cytotoxicity assay
This project is supported in part by the National Natural Science
Foundation of China (31872677). The authors are grateful to Dr.
Chun-Xia Zeng for identification of plant samples.
Three human cancer cell lines, HeLa, SGC-7901 gastric cancer,
and A-549 lung cancer, were used for cytotoxic assays. Cells were
cultured in RPMI-1640 (Sigma-Aldrich, St. Louis, MO, USA) or
DMEM medium (Hyclone, Logan, UT, USA), supplemented with 10%
Appendix A. Supplementary data
ꢃ
fetal bovine serum (Hyclone) in 5% CO
2
at 37 C. Cytotoxicity assays
were performed according to the MTT (3-(4, 5-dimethylthiazol-2-
yl)-2, 5-diphenyl tetrazolium bromide) method in 96-well micro-
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tet.2019.130515.
plates. Briefly, 100 mL of adherent cell types were seeded into each
well of 96-well cell culture plates and allowed to adhere for 12 h
before the addition of test compounds. Suspended cell types were
References
5
seeded at an initial density of 1 ꢁ 10 cells/mL just before drug
[
[
1] R. Sa, M.G. Gilbert, Flora of China, vol. 10, Science Press and Missouri Botanical
Garden Press, Beijing and St. Louis, 2008, p. 238.
2] a D. Kalaitzakis, T. Montagnon, E. Antonatou, G. Vassilikogiannakis, One-pot
synthesis of the tetracyclic framework of the aromatic Erythrina alkaloids from
simple furans, Org. Lett. 15 (2013) 3714e3717;
addition. Each tumor cell line was exposed to a test compound at
concentrations of 0.04, 0.20, 1.00, 5.0, and 25.0 mM in DMSO in
triplicate for 48 h, with cisplatin (Sigma-Aldrich) as the positive
control. After treatment, cell viability was assessed, cell growth
graphed, and IC50 values calculated by Reed and Muench's method.
b L.F. Tietze, D. Tolle, D. Kratzert, D. Stalke, Efficient formal total synthesis of the
Erythrina alkaloid (þ)-erysotramidine, using a domino process, Org. Lett. 11
(2009) 5230e5233.
[
3] B.J. Zhang, M.F. Bao, C.X. Zeng, X.H. Zhong, L. Ni, Y. Zeng, X.H. Cai, Dimeric
3.6. Xanthine oxidase inhibition activity
Erythrina alkaloids from flower of Erythrina variegata, Org. Lett. 16 (2014)
6400e6403.
[
[
[
[
4] B.J. Zhang, B. Wu, M.F. Bao, L. Ni, X.H. Cai, New dimeric and trimeric Erythrina
alkaloids from Erythrina variegata, RSC Adv. 6 (2016) 87863e87868.
5] J. Wu, B.J. Zhang, W.N. Xiao, M.F. Bao, X.H. Cai, Alkaloids from the flower of
Erythrina arborescens, RSC Adv. 7 (2017) 51245e51251.
6] M. Ozawa, S. Kawamata, T. Etoh, M. Hayashi, K. Komiyama, A. Kishida,
C. Kuroda, A. Ohsaki, Chem. Pharm. Bull. 58 (2010) 1119e1122.
Alkaloids 1e38 were bio-assayed for inhibitory activity of
xanthine oxidase. The uric acid production was calculated accord-
ing to the increasing absorbance at 290 nm. Test solutions (final
concentration of 50
m
g/mL) were prepared by adding xanthine
final concentration 29.2 g/mL). The reaction was started by add-
L of xanthine oxidase (0.1 U/mL) in a phosphate buffer
(
m
7] X.H. Cai, Z.Z. Du, X.D. Luo, Tirucallane triterpenoid saponins from Munronia
delavayi, Helv. Chim. Acta 90 (2007) 1980e1986.
ing 40
m