D. Zhou, et al.
Bioorganic Chemistry 97 (2020) 103715
Table 2
3.3. Neuroinflammatory activities of compounds 1–30 in LPS-induced BV2
Effects of extracts and identified compounds from B. striata on NO production
cells
by LPS-activated BV2 microglia cells (Mean ± SEM).d
IC50a
IC50a
To investigate the anti-inflammatory effects of the extracts and
isolated components (1–30), the inhibitory activities were evaluated by
NO assay in LPS-induced BV2 cells. In order to avoid that the inhibitory
activities exhibited by tested samples were due to their cytotoxicities,
the viabilities of BV2 cells were measured using MTT methods before
NO assays. Concentration higher than 10 µM used in our present study
is mainly based on other reports in which the highest concentration of
compounds was set to 100 µM or higher to investigate their anti-in-
flammation effect in BV2 cells [35,36]. Moreover, we also want to in-
vestigate compounds cytotoxic activities comprehensively in a wider
range, thus we set the highest dose to 100 µM. As you mentioned the
compounds are toxic at 100 µM which should not be used for anti-
inflammation evaluation.
Sample name
Ext-1b
Sample name
4.3 ± 1.6
75.3 ± 2.1
72.7 ± 1.1
> 100
5
31.8 ± 2.1
66.1 ± 2.4
61.1 ± 2.6
58.8 ± 2.6
18.1 ± 2.8
b
Ext-2
6
b
Ext-3
16
17
b
Ext-4
c
2
96.0 ± 2.6
Minocycline
a
b
IC50 (µg/mL for extracts and µM for compounds).
Ext-1: total ethanol extract of B. striata; Ext-2: PE extract of ethanol extract
of B. striata; Ext-3: EtOAc extract of ethanol extract of B. striata; Ext-4: n-BuOH
extract of ethanol extract of B. striata.
c
d
Minocycline was used as a positive control.
Compounds 1, 3, 4, 7, 8, 9, 10, 11, 13, 15, 18, 19, 20, 22, 24, 26, 27, 28,
2
9 and 30 showed no inhibitory activity at tested concentrations (1, 10, 30,
As shown in Fig. 4A, B and Table 2, the ethanol extract displayed
remarkable inhibitory effects against NO production, with IC50 value of
1
00 µM), compounds 12, 14, 21, 23, 25 showed toxicities at 30 or 100 µM.
4
.3 µg/mL and cytotoxicity was not observed at the all tested con-
s), δ
C
55.9] was deduced to be linked at C-3 from the HMBC correlation
3.75 (3-OCH ) and 161.4 (C-3). A pyrrolidone moiety was es-
centrations (1, 3, 10, 30, and 100 μg/mL). Of note, stilbenes-rich
fraction, PE extract could decrease the production of NO with IC50
value at 75.3 µg/mL without cytotoxicity observed in the experiment
of δ
H
3
tablished by 5.03 (1H, dd, J = 9.0, 5.0 Hz, H-5″), 2.39 (1H, m, H-3″a),
2
.50 (1H, m, H-3″b), 2.06 (1H, m, H-4″a) and 2.32 (1H, m, H-4″b), as
well as the HMBC correlations of H-3″/C-5″, H-4″/C-2″ (C]O), H-5″/C-
″, and C-2″ (see Fig. 2). The HMBC correlations from δ 5.03 (H-5″) to
[
the tested results of other portions, EtOAc (IC50, 72.7 µg/mL) and n-
BuOH extract (IC50 > 100 µg/mL)]. Therefore, PE and ethanol extracts
may have potential anti-inflammation activity, however PE is less ac-
tive compared to ethanol extract (Fig. 4).
3
H
δ 143.3 (C-1) and 161.4 (C-3), and from H-4″ to C-2 allowed the at-
C
tachment of the 2-oxopyrrolidin-5-yl group at C-2 (See Fig. 1 and
As shown in Fig. 4C, compounds 12, 14, 21, 23, 25 were toxic to
cells at a concentration of 30 µM or 100 µM, which may affect their
inhibitory effects on LPS-induced NO release. Taking the anti-in-
flammatory and cytotoxic activities into consideration, we found that
compounds 2, 5, 6, 16, 17 can greatly inhibit the production of NO
without showing cytotoxicity, with IC50 values at 96.0, 31.8, 66.1, 61.1,
Table 1). Therefore, structure 23 was determined as dusuanlansin E
(
Fig. 2).
Due to negligible or weak optical rotations and Cotton effects in
circular dichroism (CD) spectrum of compound 23, chiral analysis and
optical resolution of 23 were achieved by HPLC with a Chiral pack IF
chiral column (n-hexane/EtOH, 85:15) at 1.0 mL/min, which afforded
compounds 23a and 23b. The relative peak area ratio of 23a to 23b
was approximately 1:1, the absolute configuration of 23a was con-
firmed to be 5′R, and the absolute configuration of 23b was confirmed
as 5′S by combining the CD spectrum of 23a (Fig. 3) [31]. Thus, the
structures of 23a and 23b were assigned and named dusuanlansin E1
and dusuanlansin E2 (Fig. 3).
5
8.8 μM, respectively. Previous studies also reported that the effect of
compound on NO release can reflect its anti-inflammation activity.
However, this single assay may not be sufficient to indicate anti-in-
flammation activity of a compound. In future, we will further in-
vestigate the level of other pro-inflammatory mediators such as TNF-α,
IL-6 and IL-1β, and explore the effect of the compound on signaling
pathways related with inflammatory response to clarify its anti-in-
flammatory activity comprehensively.
Compound 27 was obtained as a brownish powder. High resolution
ESI-MS analysis of 27 yielded a quasi-molecular ion peak at m/z
−
On the corresponding references data (only some compounds):
compounds 5 (50.2 µM), 14 (80.2 µM), and 25 (> 100 µM) [31,37].
And these results are also very close to our measured values. But there
are few reports in the literature about SAR. Combining the activities
with structures of the isolated components, brief structure-activity re-
lationships could be suggested as follows. Firstly, the glycosidations of
stilbenes were considered to be the negative factor for their anti-in-
flammatory activities, such as compounds 5 vs 27, 29, 30 (31.8 μM
vs > 100 μM). Secondly, the presence of pyrrolidone moiety was det-
rimental to the inhibitory effects and increased the cytotoxicities, such
as 5 vs 23 (31.8 μM vs 67.7 μM). In addition, as to compounds 2 and 6,
the presence of 3-methoxy was beneficial to the activity of dihy-
dropinosylvi (2) with IC50 values of 96.0 and 66.1 μM, respectively
4
05.1554 [M−H] (calcd. 405.1555 for C21
H
25
O
8
), in accordance with
1
13
the molecular formula, C21
H
26
O
8
. Its H NMR and C NMR spectral
characteristics were very similar to those of analogous compound ba-
1
tatasin III. However, signals of 4.80 (H-1″) in H NMR spectrum in-
dicated the presence of a glucose substituent. This conclusion was also
supported by signals of 100.4 (C-1″), 73.3 (C-2″), 76.7 (C-3″), 69.7 (C-
4
″), 77.0 (C-5″), 60.7 (C-6″). And glycosyl unit was shown to be located
at C-3′ according to the long range correlations between 4.80 (1H, d,
J = 7.2 Hz, H-1″) and 157.5 (C-3′). According to the large coupling
constant, the configuration was confirmed to be β. Acid hydrolysis of 27
produced glucose as the sole sugar identified on the basis of derivati-
zation by comparing with an authentic sugar sample [14]. The structure
of 27 was elucidated as 3-hydroxy-5-methoxybibenzyl-3′-O-β-D-gluco-
pyranoside.
(
Fig. 5).
Fig. 5. The relationships of.
5