Introduction
Alisma orientale (Sam.) Juz. (Alismataceae) is an aquatic medicinal herb mainly harvested in Fujian,
Jiangxi, and Sichuan, China. This plant’s dried tubers (Rhizoma alismatis), known
as “Zexie” in China [1], are often used in clinics for eliminating dampness, reducing edema, and promoting
diuresis [2]
[3]. Triterpenes [4]
[5] and sesquiterpenes [6] are the major components in A. orientale tubes. Pharmacological studies showed that this plant’s ethanol or water extracts
have diuretic [7], lipid-lowering [8], liver-protecting [9], hypoglycemic [10], anti-inflammatory [11], and renal protective activities [12]. Besides, A. orientale’s ethanol extract attenuates lung inflammation in LPS-induced acute lung injury mice
by suppressing NF-κB and nuclear factor erythroid-2 related factor 2 activities [13]
[14]. However, the A. orientale’s triterpenoids’ inhibitory activities on NF-кB are still to be studied.
In this study, 15 protostane-type triterpenoids (1–15), including a new triterpenoid, alismaketone B (1), and a new natural nortriterpene, noralisolic acid A (2), as well as a known fatty acid compound (16), were identified from A. orientale’s rhizomes ([Fig. 1]). The new compounds’ absolute structures were elucidated by various spectroscopic
or spectrometric methods, including 1D and 2D NMR, HRESIMS, and ECD. HEK293/NF-κB
cells were used to investigate the 15 triterpenoids’ NF-κB inhibition potential, and
the isolates’ IC50 values were evaluated for their activities. These compounds exhibited NF-κB inhibition
properties and could explain the plant’s traditional use to treat renal diseases.
Fig. 1 Compounds isolated from A. orientale.
Results and Discussion
The rhizomes of A. orientale were extracted with 60% EtOH to produce a brown crude extract. The extract was separated
using silica gel and ODS-A column chromatography as well as semi-preparative HPLC
into a new triterpenoid (1), alismaketone B; a new natural nortriterpene, noralisolic acid A (2); and 14 known compounds (3–16). Compound identification was performed using mass spectrometry and NMR spectroscopy,
and spectroscopic data were compared with reference compounds reported in the literature.
The known compounds were identified as follows: alisol C 23-acetate (3) [15], alisol Q 23-acetate (4) [15], 16,23-oxido-alisol B (5) [16], alismalactone 23-acetate (6) [17], alisol A (7) [15], 25-O-methylalisol A (8) [18], alisol B (9) [15], 25-O-ethylalisol A (10) [18], alisol C (11) [16], alisol F (12) [19], 16-oxo-11-deoxy-alisol A (13) [16], alisol A 24-acetate (14) [15], 16-oxoalisol A (15) [16], and (9Z,12Z)-2,3-dihydroxypropyl octadecadienoate (16) [20].
Compound 1 was obtained as a yellow amorphous powder. This compound’s molecular formula was
established as C30H48O5 with a positive HRESIMS (m/z 533.3499, calcd. 533.3478, [M+HCOO]−). The 1H NMR spectrum ([Table 1]) of 1 showed the presence of 8 methyl groups at δ
H 1.32 (s), 1.27 (s), 1.19 (s), 1.17 (d, J=7.1 Hz), 1.07 (s), 1.06 (s), 1.05 (s), and 0.89 (s) and 4 oxygenated methines at
δ
H 5.13 (t, J=6.5 Hz), 4.10 (m), 3.28 (m), and 3.83 (m). The 13C NMR spectrum displayed 30 signals, including a carbonyl carbon at δ
C 220.0; 2 olefinic carbons at δ
C 142.6 and 135.5; 4 oxygenated methine carbons at δ
C 69.9, 71.7, 76.6, and 83.6; and an oxygenated quaternary carbon at δ
C 73.7. Furthermore, 4 quaternary carbons, 3 methines, 7 methylenes, and 8 methyls
were determined using 13C NMR and DEPT experiments. These data closely resembled those of alismaketones B
23-acetate [21], except for the absence of an acetyl group at C-23. Accordingly, compound 1 should have a hydroxyl group at C-23 instead of the 23-acetate of alismaketones B
23-acetate. The structure of compound 1 was confirmed by the correlations of H-26 (δ
H 1.32)/C-27 (δ
C 28.6), H-24 (δ
H 3.28)/C-27 (δ
C 28.6), H-23 (δ
H 4.10)/C-20(δ
C 28.7), H-21(δ
H 1.17)/C-22(δ
C 41.2), and H-16 (δ
H 5.13)/C-24 (δ
C 76.6) observed in the HMBC spectrum, which we have shown in [Fig. 2].
Fig. 2 Key 1H-1H COSY, HMBC (H→C) and NOESY (↔) correlations of compound 1.
Table 1
1H (400 MHz) and 13C (100 MHz) NMR for compounds 1 and 2 in CDCl3.
Pos.
|
1
|
2
|
δ
H, J (Hz)
|
δ
C
|
δ
H, J (Hz)
|
δ
C
|
1
|
2.20 m
|
30.7 (t)
|
2.23 m
|
31.0 (t)
|
2.06 m
|
2.10 m
|
2
|
2.67 m
|
33.5 (t)
|
2.66 m
|
33.7 (t)
|
2.34 m
|
2.33 m
|
3
|
–
|
220.0 (s)
|
–
|
220.4 (s)
|
4
|
–
|
46.9 (s)
|
–
|
47.0 (s)
|
5
|
2.11 m
|
48.2 (d)
|
2.08 m
|
48.5 (d)
|
6
|
1.47 m
|
19.9 (t)
|
1.45 m
|
20.0 (t)
|
1.28 m
|
1.29 m
|
7
|
2.02 m
|
34.4 (t)
|
2.70 m
|
34.3 (t)
|
1.28 m
|
1.23 m
|
8
|
–
|
40.6 (s)
|
–
|
40.5 (s)
|
9
|
1.75 d (10.7)
|
49.3 (d)
|
1.72 d (10.6)
|
49.6 (d)
|
10
|
–
|
36.9 (s)
|
–
|
36.9 (s)
|
11
|
3.83 m
|
69.9 (d)
|
3.85 m
|
70.0 (d)
|
12
|
2.64 m
|
34.1 (t)
|
2.76 m
|
34.4 (t)
|
2.08 m
|
2.02 m
|
13
|
–
|
142.6 (s)
|
–
|
137.3 (s)
|
14
|
–
|
54.8 (s)
|
–
|
57.0 (s)
|
15
|
2.32 m
|
40.0 (t)
|
1.87 m
|
30.5 (t)
|
1.33 m
|
1.34 m
|
16
|
5.13 t (6.5)
|
83.8 (d)
|
2.29 m
|
29.3 (t)
|
2.17 m
|
17
|
–
|
135.6 (s)
|
–
|
134.5 (s)
|
18
|
1.19 s
|
23.6 (q)
|
0.97 s
|
24.1 (q)
|
19
|
1.06 s
|
25.4 (q)
|
1.05 s
|
25.6 (q)
|
20
|
2.69 m
|
28.7 (d)
|
3.07 m
|
29.4 (d)
|
21
|
1.17 d (7.1)
|
20.8 (q)
|
1.06 d (7.1)
|
19.5 (q)
|
22
|
1.88 m
|
41.2 (t)
|
2.31 m
|
40.0 (t)
|
1.70 m
|
23
|
4.10 m
|
71.7 (d)
|
–
|
177.0 (s)
|
24
|
3.28 m
|
76.6 (d)
|
–
|
–
|
25
|
–
|
73.7 (s)
|
–
|
–
|
26
|
1.32 s
|
26.1 (q)
|
–
|
–
|
27
|
1.27 s
|
28.6 (q)
|
–
|
–
|
28
|
1.07 s
|
29.5 (q)
|
1.07 s
|
29.6 (q)
|
29
|
1.05 s
|
20.0 (q)
|
1.05 s
|
20.1 (q)
|
30
|
0.89 s
|
23.5 (q)
|
1.10 s
|
22.8 (q)
|
The relative configuration of compound 1 was established from the 3
J
H,H coupling value as well as NOESY data. A considerable 3
J
9,11 value of 10.7 Hz indicated the axial-axial relationship of H-9 and H-11. As shown
in [Fig. 2], the NOESY correlations of Me-18 (δ
H 0.89) with H-5 (δ
H 2.11), H-11 (δ
H 3.83), H-16 (δ
H 5.13), and H-15α (δ
H 2.32); the correlations of Me-30 (δ
H 1.19) with H-9 (δ
H 1.75) and H-24 (δ
H 3.28); and the correlation of H-24 (δ
H 3.28) with H-15β (δ
H 1.33) indicated that H-5, H-11, H-16, and Me-18 were α-oriented, whereas H-9, H-24, and Me-30 were β-oriented. Additionally, the correlations of Me-19 (δ
H 1.06)/H-9 (δ
H 1.75), H-20 (δ
H 2.70)/H-23 (δ
H 4.10), H-11(δ
H 3.83)/Me-21 (δ
H 1.17), and H-16 (δ
H 5.13)/Me-21 (δ
H 1.17) indicated the β-configurations of Me-19, H-20, and H-23. Compound 1’s experimental ECD spectrum showed a positive cotton effect at 290 nm and a negative
cotton effect at 205 nm. The comparison between the TDDFT calculated spectrum (B3LYP/6-311 G*
level) and the experimental data showed good agreement with 5 R, 8 R, 9 S, 10 S, 11 S, 14 S, 15 S, 20 R, 23 S, and 24 R. Still, the alternative form revealed a curve with the opposite cotton effect ([Fig. 3]). To the best of our knowledge, 1 was designated as the new protostane-type triterpenoid alismaketone B.
Fig. 3 Experimental and calculated ECD spectra of 1 in MeOH.
Compound 2 was obtained as a yellow amorphous powder. This compound’s molecular formula was
assigned as C26H40O4 by negative HRESIMS at m/z 461.2883 [M+HCOO]− (calcd. 461.2903). The 1H-NMR spectrum ([Table 1]) showed 6 tertiary methyl groups at δ
H 0.97 (s), 1.05 (s), 1.05 (s), 1.06 (d, J=7.1 Hz), 1.07 (s), and 1.10 (s) and an oxygenated proton signal at δ
H 3.85 (1H, m). The 13C-NMR data displayed signals for 2 carbonyls at δ
C 220.4 and 177.0. Compound 2’s 1H and 13C NMR spectroscopic data were highly similar to those reported for alisol A (7) [15], except for the absence of oxygenated carbon signals at C-23, C-24, and C-25, and
the presence of an additional carbonyl group (δ
C 177.0). Accordingly, 2 should have a carboxyl group at C-23 instead of the triol unit of alisol A (7). This feature is consistent with HMBC correlations observed ([Fig. 4]) between H-22 (δ
H 2.31) and C-23 (δ
C 177.0), C-21 (δ
C 40.0) and C-17 (δ
C 134.5), H-20 (δ
H 3.07) and C-23 (δ
C 177.0), as well as between H-20 (δ
H 3.07) and C-16 (δ
C 29.3).
Fig. 4 Key 1H-1H COSY, HMBC (H→C) and NOESY (↔) correlations of compound 2.
A considerable 3
J
9,11 value of 10.6 Hz indicated the axial-axial relationship of H-9 and H-11. The NOESY
correlations of Me-18 (δ
H 0.97)/H-11 (δ
H 3.85), H-11/H-5 (δ
H 2.08), Me-18/H-15α (δ
H 1.87), Me-30 (δ
H 1.10)/H-9 (δ
H 1.72), Me-30/H-15β (δ
H 1.34), and H-9/Me-19 (δ
H 1.05) supported the chair, boat, and chair conformations of rings A, B, and C, as
well as their ring junction’s trans-cis-trans relationships ([Fig. 4]). The experimental ECD spectrum of 2 showed a positive cotton effect at 293 nm and a negative cotton effect at 203 nm.
A comparison of the TDDFT-calculated spectrum (B3LYP/6–311 G* level) and the experimental
data showed good agreement with 5 R, 8 R, 9 S, 10 S, 11 S, and 14 S, while the alternative form revealed a curve with the opposite cotton effect ([Fig. 5]). Besides, the absolute configuration at C-20 was considered to be R based on the biogenesis of the protostane skeleton [22]. For the first time, the absolute configuration of 2 was discussed. Compound 2 was assigned as a new natural nortriterpenoid noralisolic acid A.
Fig. 5 Experimental and calculated ECD spectra of compound 2 in MeOH.
Lee reported the synthesis of noralisolic acid A, and its structure was assigned based
on IR analysis and limited NMR data [23]. The 13C-NMR data of synthetic noralisolic acid A closely resembled those of compound 2, except for the lower field shift of the signal of C-22 (δ
C 57.7). However, this assignment of C-22’s chemical shift was considered incorrect
based on the following evidence: (1) the methylene carbon attached to the carboxyl
group may give a signal at approximately δ
C 38.0 [24]; (2) an HMBC correlation from H-22 (δ
H 2.31) to C-23 (δ
C 177.0) permitted the carboxyl group of compound 2 to be attached to C-22.
Compounds 1–15 were evaluated for their NF-κB inhibition activity in TNF-α stimulated HEK 293 cell line. Triterpenoids 8–10 and 14 showed moderate inhibitory activity with IC50 values of 64.7, 32.3, 47.3, and 37.3 μM, respectively. However, the other compounds
displayed mild NF-κB inhibition at a maximum tested concentration of 50 μM. Additionally,
the IC50 of IMD-0354 was 13.3 μM, which was proved that the test system and calculation method
were feasible. The results were shown in [Table 2] and [Fig. 6]. The results indicated that the sp3 hybrid C-16 of Alisma triterpenoids (8–10
vs.
3, 11, 13, and 15) is essential for NF-κB inhibition. The side-chain cyclization at C-17 depleted the
NF-κB inhibition (9
vs.
5). The hydroxyl groups at C-23 and the less hydrophilic groups at C-24/25 (-OAc or
epoxy group) contributed to the NF-κB inhibitory activity (9 and 14
vs.
3, 4, 6, 8, 10, 13, and 15) ([Fig. 7]).
Fig. 6 Inhibitory activity of compounds 8, 9, 10, and 14 in TNF-α stimulated HEK293 cells. HEK293 cells were treated with indicated doses of constituents
for 24 h, and then chemiluminescence was determined. IMD-0354 was used as a positive
control and possessed an inhibitory rate of approximately 50% at 15 µM. Experiments
were performed in duplicate, and data were represented as means±SD.
Fig. 7 Protostane-type triterpenoids’ structure-activity relationships.
Table 2 NF-κB cell line HEK293 inhibitory activities of active compounds.
compound
|
inhibition rate (%)
|
IC50 (μM)
|
8
|
58.4±0.0
a
|
64.7 (C.I. 33.4–139.2)
|
9
|
94.8±0.4
a
|
47.3 (C.I. 29.2–78.3)
|
10
|
51.7±3.7
a
|
47.3 (C.I. 29.2–78.3)
|
14
|
64.8±3.8
a
|
37.3 (C.I. 24.3–59.5)
|
IMD0354
|
57.6±2.7
b
|
13.3 (C.I. 10.2–17.6)
|
a Measured in 50 μM;
b
Measured in 30 μM. IC50 was afforded with confidence interval (n=2); C.I.: 95% confidence interval. Positive
control: IMD-0354.
It has been reported that ethanol extract of A. orientale tuber could suppress LPS-induced NF-κB activity and NF-κB dependent gene expression
in RAW 264.7 cells. Alisol B (9) was detected in EEAO by HPLC analysis but was not tested for its NF-κB inhibitory
activity [13]. Our results confirmed that not only alisol B (9) but also 25-O-methylalisol A (8), 25-O-ethylalisol A (10), and alisol A 24-acetate (14) were involved in NF-κB activity. Compounds 9 and 14 were also reported to show anti-inflammatory and anti-allergic activities via inhibiting CD147 and MMP-9 secretion and leukotriene production [25]
[26]
[27].
In addition, compounds 1–15 were also tested for their antihuman dihydroorotate dehydrogenase activity by the
reported method [28]. Nonetheless, they did not show any benefit (Data not shown).
Materials and Methods
General experimental procedures
IR spectra were recorded on Chiral IR-2X vibrational circular dichroism spectrometer
in deuterated chloroform. The optical rotation was determined on JASCO P-2000 automatic
polarimeter in MeOH. NMR spectra were measured on Bruker Avance DRX-400. HRESIMS was
performed on Waters Xevo G2-XS Q-TOF mass spectrometer, and the semi-preparative HPLC
was performed on Chuangxin Tongheng 3050 N HPLC system with a C18 column (YMC-Pack ODS-A, 5 μm, 10.0×250 mm). The ECD spectra were acquired in MeOH
using JASCO J-815 circular dichroism spectrometer. Chemiluminescence measurements
for the NF-κB cell line HEK293 inhibition assay were recorded on Envision (PerkinElmer).
Solvents and chemicals
IMD-0354 (purity≥99%) was purchased from Aladdin. DMEM was acquired from BI and FBS
from Gibco. Penicillin and streptomycin (P/S) were obtained from Procell. Bright–Glo
was gifted by Promega. TNF-α was purchased from Peprotech. DMSO was acquired from Sigma. The solvents used for
extraction and chromatographic separation were of analytical purity.
Plant material
The rhizomes of A. orientale were sampled from Pengshan County, Sichuan Province in China, in November 2018 and
authenticated by Prof. Tong Wu from Pharmacognosy Department. A voucher specimen (20190409)
was deposited in the Shanghai Institute of Pharmaceutical Industry.
Extraction and isolation
The air-dried and powdered rhizomes (5.0 kg) of A. orientale were extracted with 60% EtOH (15 L×3) for 3 times (2 h each time) under reflux. The
combined solution was concentrated to crudeness in vacuo, and the crude extract was suspended in water (3 L), sequentially partitioning it
with petroleum ether (60–90°C) and dichloromethane (3×3 L each), respectively. The
dichloromethane extract (120.4 g) was chromatographed on a silica gel column (CC)
(6×72 cm; 200–300 mesh) and eluted by petroleum ether/ethyl acetate (50:1, 20:1, 10:1,
4:1, 2:1,1:1, v/v) to give 8 fractions (Fr.1–8) based on TLC analyses. Fr.4 (11.17 g)
was subjected to an ODS-A gel CC (6.6×20 cm; S-50 μm,12 nm; YMC) and eluted by MeOH/H2O (1:9, 3:7, 1:1, 7:3, 9:1, and 1:0 v/v) to yield 9 fractions (Fr.4a–4i). Fr.4b (1.73 g)
was purified using semi-preparative HPLC (60% CH3CN in water; 4.0 mL/min; UV detection at λ 210 nm) to isolate compound 3 (326.0 mg, tR=21 min). Fr.4d (0.79 g) was purified via semi-preparative HPLC (50% CH3CN in water; 4.0 mL/min; UV detection at λ 210 nm) to afford compounds 4 (16.5 mg, tR=37 min) and 5 (10.2 mg, tR=42 min). Compounds 6 (62.4 mg, tR=65.2 min) and 2 (22.9 mg, tR=27.9 min; 97% purity by HPLC) were obtained from Fr.4 f (0.44 g) through semi-preparative
HPLC eluted using CH3CN/H2O (78:22, v/v; 3.0 mL/min; UV detection at λ 210 nm). Fr.4 h (2.96 g) was subjected
to semi-preparative HPLC with CH3CN/H2O (60:40–80:20, v/v, 90 min; 4.0 mL/min; UV detection at λ 210 nm) to obtain compounds
7–10 (62.5, 85.8, 228.5, and 34.5 mg, respectively; tR=14.7, 28.2, 31.6, and 41.2 min). Compound 16 (17.3 mg, tR=21.3 min) was purified from Fr.4i (0.32 g) by semi-preparative HPLC (80% CH3CN in water; 4.0 mL/min; UV detection at λ 210 nm). Fr.6 (6.45 g) was subjected to
an ODS-A gel CC eluted with MeOH/H2O (1:9, 3:7, 1:1, 7:3, 9:1, and 1:0, v/v) to obtain 6 fractions (Fr.6a–6 f). Fr.6b
(1.17 g) was further purified via semi-preparative HPLC (40% CH3CN in water; 3.0 mL/min; UV detection at λ 210 nm) to obtain compounds 11, 12, 13 (150.0, 30.0, and 26.0 mg, tR=24, 56 and 31 min) and 1 (16.1 mg, tR=33 min; 93% purity by HPLC). Compound 14 (12.1 mg, tR=29 min) was obtained from Fr.6d (0.37 g) and purified using semi-preparative HPLC
eluted with CH3CN/H2O (57:43, v/v; 3.0 mL/min; UV detection at λ 210 nm). Finally, Fr.6e (0.25 g) was
subjected to semi-preparative HPLC with CH3CN/H2O (35:65, v/v; 3.0 mL/min; UV detection at λ 210 nm) to obtain compound 15 (40.1 mg, tR =21.1 min).
Alismaketone B (1): yellow amorphous powder;
+ 53.4 (c 0.02, MeOH); ECD (MeOH) λ max (∆ε) 290 (+2.98), 205 (−9.96); IR (CDCl3) Vmax 3390, 2941, 1695 cm−1; 1H and 13C NMR spectroscopic data (see [Tables 1]); HRESIMS at m/z 533.3499 [M+HCOO]− (calcd. for C31H49O7, 533.3478).
Noralisolic acid A (2): yellow amorphous powder;
+ 57.0 (c 0.04, MeOH); ECD (MeOH) λ max (∆ε) 293 (+2.44), 203 (−3.43); IR (CDCl3) Vmax 3444, 2941, 1747, 1699, 1654 cm−1; 1H and 13C NMR spectroscopic data (see [Tables 1]); HRESIMS at m/z 415.2817 [M−H]−, 831.5751 [2 M−H]−, 461.2883 [M+HCOO]− (calcd. for C27H41O6, 461.2903).
ECD calculation
Computational calculations were performed using the Gaussian 09 software. Conformers
were generated by the MMFF94 force field and were obtained at 6 kcal/mol. Conformational
analyses and geometry optimizations were performed at the B3LYP/6–311 G* level in
MeOH. ECD calculations were performed by TDDFT at the B3LYP/6–311 G* level in MeOH.
ECD spectra were obtained by weighing each geometric conformation’s Boltzmann distribution
rate with a bandwidth σ of 0.3 eV (1) or 0.4 eV (2).
Inhibitory effects on NF-кB cell line HEK293
The NF-кB activity in the TNF-α stimulated HEK293 cell line was used to evaluate the inhibitory effects of all these
isolates using the modified Bright-Glo method [29]. Compounds 1–15 were dissolved in DMSO to prepare a stock solution, and then each stock solution
was serially diluted 2-fold to different concentrations using a culture medium. HEK293
cells were cultured in a DMEM medium plus 10% FBS and 1% P/S were seeded in 96-well
plates at a density of 40,000 cells/well with 80 μL. After incubating at 37°C under
5% CO2 overnight, the tested compounds (10 μL) were added to the wells. They were incubated
at the same conditions for 2 h, and TNF-α (10 μL, 20 ng/mL) was then added to each well and incubated in the darkness for 24 h.
Chemiluminescence was measured on Envision after the addition of 50 µL Bright-Glo
Luciferase Assay Reagent to each well.
Inhibition (%)=100−(B/ A×100)
Where, B was the luminescence value of tested compounds, and A was luminescence of
the negative control group (with cells, medium, DMSO, and TNF-α).
Data were analyzed by GraphPad Prism v.7.0 and were presented as a geometric means
with 95% confidence intervals of 2 independent experiments. IMD-0354 was used as a
positive control and DMSO as a negative control.