Glycosides from Buddleja officinalis with their protective effects on photoreceptor cells in light-damaged mouse retinas

Article abstract of DOI:10.1080/14786419.2021.1944138

A new phenylethanoid, hebitol IV (1), along with fifteen known glycosides (2-16), were isolated from water extract of the flower buds of Buddleja officinalis. Their structures were elucidated on the basis of 1 D-NMR, 2 D-NMR and MS data. Molecular docking showed the potential activities of the natural products against VEGFR-2. Bioassay results revealed that the compounds 10 and 14 exhibited strong inhibitory activity against VEGFR-2 with IC₅₀ values of 0.51 and 0.32 μM, respectively. Moreover, the potential retinal protective effects of 10 and 14 were then investigated in the mouse model featuring bright light-induced retinal degeneration. The results demonstrated remarkable photoreceptor protective activities of 10 and 14in?vivo.

Full text of DOI:10.1080/14786419.2021.1944138

Natural Product Research
Formerly Natural Product Letters
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/gnpl20
Glycosides from Buddleja officinalis with their
protective effects on photoreceptor cells in light-
damaged mouse retinas
Zehai Long, Xiaoye Du, Qiyao Wang, Zhijian Xu, Bo Li, Yong Zhang, Yu Chen,
Qi Jia, Yiming Li & Weiliang Zhu
To cite this article: Zehai Long, Xiaoye Du, Qiyao Wang, Zhijian Xu, Bo Li, Yong Zhang, Yu Chen,
Qi Jia, Yiming Li & Weiliang Zhu (2021): Glycosides from Buddlejaꢀofficinalis with their protective
effects on photoreceptor cells in light-damaged mouse retinas, Natural Product Research, DOI:
10.1080/14786419.2021.1944138
To link to this article: https://doi.org/10.1080/14786419.2021.1944138
View supplementary material
Published online: 26 Jun 2021.
Submit your article to this journal
Article views: 37
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=gnpl20

NATURAL PRODUCT RESEARCH
https://doi.org/10.1080/14786419.2021.1944138
Glycosides from Buddleja officinalis with their protective
effects on photoreceptor cells in light-damaged
mouse retinas
a
b,c
a
d
d,e
d,e
Zehai Long , Xiaoye Du , Qiyao Wang , Zhijian Xu
, Bo Li , Yong Zhang
,
b,c,f
a
a
d,e
Yu Chen , Qi Jia , Yiming Li and Weiliang Zhu
a
School of Pharmacy, Shanghai University of Traditional Chinese Medicine, Shanghai, China;
b
c
Yueyang Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai, China; Clinical
Research Institute of Integrative Medicine, Shanghai Academy of Traditional Chinese Medicine,
d
Shanghai, China; CAS Key Laboratory of Receptor Research; Drug Discovery and Design Centre,
Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai, China; University of
Chinese Academy of Sciences, Beijing, China; Laboratory of Clinical and Molecular Pharmacology,
e
f
Yueyang Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai, China
ABSTRACT
ARTICLE HISTORY
Received 19 April 2021
Accepted 8 June 2021
A new phenylethanoid, hebitol IV (1), along with fifteen known
glycosides (2-16), were isolated from water extract of the flower
buds of Buddleja officinalis. Their structures were elucidated on
the basis of 1 D-NMR, 2 D-NMR and MS data. Molecular docking
showed the potential activities of the natural products against
VEGFR-2. Bioassay results revealed that the compounds 10 and
KEYWORDS
Loganiaceae; Buddleja
officinalis; glycosides;
VEGFR-2 inhibitory activity;
photoreceptor
1
4 exhibited strong inhibitory activity against VEGFR-2 with IC50
values of 0.51 and 0.32 lM, respectively. Moreover, the potential
retinal protective effects of 10 and 14 were then investigated in
the mouse model featuring bright light-induced retinal degener-
ation. The results demonstrated remarkable photoreceptor pro-
tective activities of 10 and 14 in vivo.
protective activity
CONTACT Yong Zhang
zhangyong109@simm.ac.cn; Qi Jia
q_jia@126.com
Supplemental data for this article can be accessed online at https://doi.org/10.1080/14786419.2021.1944138.
ß 2021 Informa UK Limited, trading as Taylor & Francis Group

2
Z. LONG ET AL.
1. Introduction
Retinal degenerative disease is a common and serious disease that causes blindness
with high morbidity year by year, which is classified as "Shizhan Hunmiao" in Chinese
and sudden blind by traditional Chinese medicine (TCM). Recent fact sheets (World
Health Organization 2021) by the WHO (World Health Organization) indicated that at
least 1 billion people have a preventable vision impairment globally. The research
(American Macular Degeneration Foundation [date unknown]) on the pathological
mechanism of retinal degenerative disease revealed that VEGF (vascular endothelial
growth factor) signal pathway is one of the essential mechanisms on retinal neovascu-
larization, closely related to the onset of retinal degenerative disease. And the VEGF
receptor (VEGFR) inhibitor drugs have been used for treating retinal degenerative dis-
ease in clinical (Dugel et al. 2020).
Buddleja officinalis Maxim., belonging to the family Loganiaceae, is widely distrib-
uted in China (Editorial Committee of flora of China 1992), and its flower buds have
been used as a traditional Chinese medicine for treating opacity of the cornea, photo-
phobia, delacrimation, conjunctival congestion, and other eye diseases (Chinese
Pharmacopoeia Commission 2020). Comprehensive phytochemical studies on B. offici-
nalis revealed that more than 100 compounds have been isolated from this plant,
including flavonoids (Matsuda et al. 1995), phenylethanoids (Zhang and Pan 1996), tri-
terpene saponins (Guo et al. 2004), iridoids (Tai et al. 2011), monoterpene (Lee et al.
2
013), lignans (Park et al. 2016) and alkaloids (Xie et al. 2019), which have shown sev-
eral biological activities, such as anti-inflammatory (Tai et al. 2009), antiproliferative
Zhang and Pan 1996) and neuroprotective effect (Sheng et al. 2002). However, previ-
(
ous research on treatment of eye diseases of this plant mainly focused on the effect
of extract (Jie and Gao 2004; Li et al. 2009; Shen et al. 2016), and no studies have
been reported on compounds in this plant for the treatment of eye diseases.
Furthermore, decoction boiled with water was the traditional usage of this herb to
treat eye diseases. Herein, we report the isolation and structural elucidation of a new
phenylethanoid, named hebitol IV (1), along with fifteen known glycosides (2-16)
(Figure 1; Figure S1 in Supporting Information) from water extract of the flower buds
of B. officinalis, as well as the evaluation of in vitro inhibitory activity of the sixteen
Figure 1. Structures of compounds 1, 10 and 14.

NATURAL PRODUCT RESEARCH
3
compounds on VEGFR-2. Moreover, the potential retinal protective effects of 10 and
4 were then investigated in the mouse model featuring bright light-induced retinal
degeneration.
1
2. Results and discussion
Compound 1 was obtained as a yellowish amorphous power. Its molecular formula
þ
was established as C H O by HR-ESI-MS (m/z: 513.1584 [M þ Na] ). The IR spectrum
21 30 13
ꢀ
1
ꢀ1
ꢀ1
ꢀ1
showed absorption bands at 3355 cm , 1695 cm , 1630 cm ,1603 and 1514 cm
073 cm , which suggested the existence of hydroxy, a, b-unsaturated ester carbonyl,
,
ꢀ
1
1
olefinic, aromatic and ether functional groups, respectively. And the UV absorption
1
bands at k max 225 and 310 nm also revealed the presence of aromatic rings. The H
0
0
NMR spectrum of 1 exhibited signals for AA XX system at d 7.47 (2H, d, J ¼ 8.6 Hz, H-
H
2
, 6) and 6.80 (2H, d, J ¼ 8.6 Hz, H-3, 5), trans-olefinic protons at d 7.64 (1H, d,
H
J ¼ 15.9 Hz, H-7) and 6.36 (1H, d, J ¼ 15.9 Hz, H-8), and sugar moieties at d 3.27-4.52.
H
1
3
The C NMR spectrum revealed the presence of 21 carbon atoms, including a a,
b-unsaturated ester carbonyl group at d 169.1 (C-9), 146.9 (C-7), and 114.9 (C-8), six
C
aromatic carbons at d 161.4 (C-4), 131.3 (C-2, 6), 127.1 (C-1), and 116.8 (C-3, 5), and
C
twelve carbons attributable to a sugar moiety (d 64.6-105.0) and a hexitol moiety (d
C
C
6
5.2-73.7). The carbon signals of a, b-unsaturated ester carbonyl group, along with the
0
0
signal of trans-double bond and the AA XX system suggested the presence of p-
hydroxycinnamoyl group. The sugar moiety was deduced as glucopyranoside unit
based on the six carbon signals (d 105.0, 77.7, 75.6, 75.2, 71.6, and 64.6) and the cor-
C
0
responding proton signals (d 3.27-4.52). In addition, the moiety acylated in the C-6
H
0
position of glucopyranoside unit was deduced due to the downfield shift of C-6 (dC
6
4.6). The left six carbon signals (d 65.2-73.7) with the lack of an anomeric carbon sig-
C
nal (around d 100) were assigned to a hexytol moiety, which was determined as man-
C
nitol based on the coincidence of the corresponding carbon atoms of hebitol II
(Taskova et al. 2011). Furthermore, the NMR data of 1 were highly similar to those of
hebitol II, except for those of the p-hydroxyphenyl group in 1, suggesting that the p-
0
00
hydroxycinnamoyl group was attached to the C-6 of glucopyranosyl and the C-6 of
0
mannitol moiety was attached to the C-1 . Subsequently, the linkage was confirmed
by the HMBC correlations (Figure S2) of d 4.52 and 4.29 (H-6 ) with d 169.1 (C-9),
and of d 4.36 (H-1 ) with d 73.7 (C-6 ). The coupling constant of the anomeric pro-
0
H
C
0
00
H
C
0
0
ton (d 4.36, d, J ¼ 7.8 Hz, H-1 ) and the signal at d 105.0 (C-1 ) indicated the b-config-
H
C
uration of the glucose, and the absolute configuration of the sugar moieties were
determined as D-glucose and D-mannitol by acid hydrolysis and GC analysis.
Therefore, the new compound 1 was elucidated as 6-O-p-hydroxycinnamoyl-b-D-gluco-
pyranosyl-(1!6)-D-mannitol, and named hebitol IV.
The other known compounds were identified as hebitol II (2) (Taskova et al. 2011),
globularitol (3) (Cali s¸ et al. 2002), syringin (4) (Venditti et al. 2017), 2-phenylethyl-b-D-
glucopyranoside (5) (Liu et al. 2019), salidroside (6) (Yang et al. 2017), phenethyl alco-
hol-8-O-b-D-glucopyranosyl-(1!2)-b-D-glucopyranoside (7) (Ono et al. 1999), cistano-
side F (8) (Kim et al. 2007), isocistanoside F (9) (Liu et al. 2016), verbascoside (10) (Li
et al. 2013), isoverbascoside (11) (Li et al. 2013), campneoside II (12) (Kim et al. 2007),

4
Z. LONG ET AL.
isocampneoside II (13) (Kim et al. 2007), echinacoside (14) (Porter et al. 2015), angoro-
side A (15) (Li et al. 2011), forsythoside B (16) (Yamasaki et al. 2007), with their NMR
spectroscopic data identical to those reported in the literature.
The compounds 1–16 were evaluated as potential VEGFR inhibitors by molecular
docking and in vitro inhibitory activity assay. As shown in Table S2, the docking results
showed that nearly half of the compounds were rated above 10.00 kcal/mol in abso-
lute value, indicating their potential VEGFR-2 inhibitory activities. And a preliminary
bioassay revealed that nine compounds (7, 9-16) exhibited inhibitory activity on
VEGFR-2 at the concentration of 10 lM (the inhibition rate of each compound is over
50%). Further assay demonstrated that the compounds 10 and 14, are potent natural
inhibitors against VEGFR-2 with the IC50 values of 0.51 and 0.32 lM, respectively
(Figure S3).
To investigate the potential retinal protective activity of 10 and 14, BALB/c mice
were administered with 10 or 14 at 100 mg/kg body weight (bw) 30 min before
bright light exposure (10,000 lux for 30 min). Seven days after the bright light expos-
ure, non-invasive optical coherence tomography (OCT) was employed to assess the
changes of the retinal structure in a non-biased manner. As shown in Figure S6A
and B, compared to the intact photoreceptor structure found in vehicle-treated mice
without bright light exposure, bright light-exposed vehicle-treated mice exhibited
severe impairment of photoreceptor structure, which was characterized by dimin-
ished outer nuclear layer (ONL). In contrast, well-preserved ONL and significantly
increased thickness of ONL was observed in light-exposed mice treated with 10 or
1
4. The results demonstrated remarkable photoreceptor protective activities of 10
and 14 in vivo.
3. Experimental
3.1. General experimental procedures
HR-ESI-MS were measured by an Agilent G6520 Q-TOF LC-MS spectrometer (Agilent
Technologies Inc., Santa Clara, USA). Optical rotations were obtained on an Autopol VI
9
0079 polarimeter (Rudolph Research Analytical, Hackettstown, NJ, USA). IR spectra
were recorded on a Thermo nicolet 6700 spectrophotometer (Thermo Fisher Scientific,
MA, USA). UV spectra were measured on an Agilent 1260 infinity HPLC equipped with
a DAD detector (Agilent Technologies Inc., Palo Alto, CA, USA). NMR spectra including
1
13
H NMR, C NMR and 2 D NMR were recorded on a Bruker Avance III 600 spectrom-
eter (Bruker Corporation, Rheinstetten, Switzerland), with chemical shifts referenced to
tetramethylsilane (TMS) as an internal standard. GC analysis was carried out on a
Shimadzu GCMS-QP2010E instrument (Shimadzu Corporation, Kyoto, Japan) equipped
with a DB-5MS column (30 m ꢁ 0.25 mm ꢁ 0.25 mm). D-101 macroporous adsorption
resin (Sinopharm Chemical Reagent Co., Ltd, Shanghai, China), Reversed-phase C18 sil-
ica gel (50 lm, YMC Co., Ltd, Kyoto, Japan) and Sephadex LH-20 (Pharmazia, Uppsala,
Sweden) were used for column chromatography. TLC was performed on precoated sil-
ica gel plates (Yan Tai Zi Fu Chemical Group Co.; G60, F-254).

NATURAL PRODUCT RESEARCH
5
3.2. Plant material
The dried flowers and inflorescences of B. officinalis were purchased in 2018 from the
market of Chinese herbal medicine in Bozhou, Anhui Province, China. The plants were
identified by Dr. Yan-Hong Shi, Shanghai University of Traditional Chinese Medicine. A
voucher specimen (No. MMH-20170427) was deposited in the Drug Discovery and
Design Centre, Shanghai Institute of Materia Medica, Chinese Academy of Sciences.
3.3. Extraction and isolation
The air-dried flower buds of B. officinalis (10.2 kg) were reflux-extracted two times with
ꢂ
distilled water (80 L) at 100 C for 1 hour each time. The solution was combined and
concentrated to 12 L, which was then partitioned with EtOAc and n-BuOH, succes-
sively. The n-BuOH fraction was subjected to column chromatography (CC) on a mac-
roporous adsorption resin column, eluted with a gradient system of EtOH-H O (0:100
2
to 95:5) to yield four fractions (Fr.1-Fr.4). Fr.2 was chromatographed on a reversed-
phase C18 silica gel (ODS) column and eluted with MeOH-H O (0:100 to 100:0) to give
2
five fractions (Fr.2.1-Fr.2.5). Fr.2.1 was subjected to repeated CC on ODS columns
eluted with MeOH-H O (0:10 to 3:7) and Sephadex LH-20 columns eluted with 10%
2
MeOH to yield compounds 6 (53.7 mg). Fr.2.2 was separated on a Sephadex LH-20 col-
umn eluted with 5% MeOH to afford four subfractions (Fr.2.2.1-Fr.2.2.4). Fr.2.2.1 was
purified subsequently by repeated CC on ODS columns eluted with MeOH-H O (0:10
2
to 3:7) and Sephadex LH-20 columns eluted with 10% MeOH, resulting in the isolation
of compounds 1 (3.4 mg), 3 (9.5 mg), 4 (75.9 mg). Fr.2.2.2 was subjected to repeated
CC on ODS columns eluted with MeOH-H O (0:10 to 3:7) and Sephadex LH-20 columns
2
eluted with 10% MeOH to yield compounds 2 (4.3 mg), 8 (85.2 mg), 14 (719.0 mg), and
9
(12.7 mg). Compounds 10 (890.7 mg) were obtained from Fr.2.2.3 by CC on ODS
eluted with MeOH-H O (0:10 to 3:7), and compound 12 (31.1 mg) was obtained from
2
the further Sephadex LH-20 columns eluted with 10% MeOH. Fr.2.2.4 were purified by
CC on ODS with a gradient of MeOH-H O (0:10 to 3:7) to afford compounds 13
2
(22.2 mg) and 11 (138.8 mg). Fr.2.3 was separated into six fractions (Fr.2.3.1-Fr.2.3.6)
with the elution of MeOH-H O (0:10 to 6:4) on an ODS column. Fr.2.2.2 and Fr.2.2.5
2
was both subjected to repeated CC on Sephadex LH-20 columns eluted with 10%
MeOH to give compounds 5 (30.1 mg), 7 (38.5 mg) from the former, together with 15
(6.1 mg), 16 (19.4 mg) from the latter.
3
.3.1. Hebitol IV (1)
ꢀ
1
Yellowish amorphous power; [a]20D-10.6 (c 0.14, MeOH); IR (KBr) mmax (cm ): 3355,
2
923, 1695, 1630, 1603, 1586, 1514, 1437, 1261, 1169, 1073, 834, 525; UV (MeOH) k
max
þ
(
nm): 225, 310; HR-ESI-MS: m/z 513.1584 [M þ Na] (calcd for C H O Na, 513.1579).
21 30 13
1
H NMR (600 MHz, CD OD) d : 7.64 (1H, d, J ¼ 15.9 Hz, H-7), 7.47 (2H, d, J ¼ 8.6 Hz, H-2,
3
H
6
), 6.80 (2H, d, J ¼ 8.6 Hz, H-3, 5), 6.36 (1H, d, J ¼ 15.9 Hz, H-8), 4.52 (1H, dd, J ¼ 11.9,
0
2
.0 Hz, H-6’a), 4.36 (1H, d, J ¼ 7.8 Hz, H-1 ), 4.29 (1H, dd, J ¼ 11.9, 5.8 Hz, H-6’b), 4.16
0
0
(
1
6
1H, dd, J ¼ 10.5, 2.4 Hz, H-6 ’a), 3.84 (1H, overlapped, H-4 ’), 3.81 (1H, overlapped, H-
0
0
0
’b), 3.79 (1H, overlapped, H-3 ’), 3.78 (1H, overlapped, H-5 ’), 3.72 (1H, dd, J ¼ 10.5,
0
0
0
.2 Hz, H-6 ’b), 3.68 (1H, m, H-2 ’), 3.61 (1H, dd, J ¼ 11.2, 6.0 Hz, H-1 ’a), 3.55 (1H, m, H-

6
Z. LONG ET AL.
0
0
0
5
8
(
(
1
), 3.41 (1H, t, J ¼ 8.9 Hz, H-3 ), 3.37 (1H, overlapped, H-4 ) , 3.27 (1H, dd, J ¼ 8.9,
0
13
.0 Hz, H-2 ); C NMR (150 MHz, CD OD) d : 169.2 (C-9), 161.3 (C-4), 146.9 (C-7), 131.2
C-2, 6), 127.2 (C-1), 116.8 (C-3, 5), 114.9 (C-8), 105.1 (C-1 ), 77.7 (C-3 ), 75.6 (C-5 ), 75.2
C-2 ), 73.7 (C-6 ’), 73.0 (C-2 ’), 71.7 (C-5 ’), 71.7 (C-4 ), 71.2 (C-4 ’), 71.0 (C-3 ’), 65.2 (C-
0
3
C
0 0 0
0
0
0
0
0
0
0
0
’), 64.6 (C-6 ).
3.4. Determination of sugar components
The absolute configuration of sugars was determined by acid hydrolysis and GC ana-
lysis with the method described in the literature (Gao et al. 2017). Compound 1 (1 mg)
ꢂ
was added in 2 mL 10% HCl/dioxane (1:1, v/v) and then heated at 80 C for 2 h. After
the reaction mixture was evaporated under N , the residue was dissolved in anh. pyri-
2
dine (100 lL). Subsequently, 0.1 M L-cystein methyl ester hydrochloride (200 lL) was
ꢂ
added, and the mixture was warmed for 1 h at 60 C. Then, the 2.6 mL trimethylsilyla-
tion reagent HMDS-TMCS (hexamethyldisilazane/Me SiCl/pyridine, 2:1:10, v/v/v) was
3
ꢂ
added, and warmed for 30 min at 60 C. The mixture was partitioned with H O and
2
cyclohexane (each for 2 mL), and the cyclohexane layer was analyzed by GC to identify
the sugars. GC was conducted on the following conditions: cap. column, DB-5MS,
(
30 m ꢁ 0.25 mm ꢁ 0.25 mm); detection, FID; carrier, N gas; detector and injection
2
ꢂ
ꢂ
temp., 280 C; initial temp. was maintained at 100 C for 5 min and then raised to
2
ꢂ ꢂ
80 C at the rate of 35 C/min, and final temp. was maintained for 9 min. As a result,
the sugars were determined as D-mannitol (t ¼ 11.45) and D-glucose (t ¼ 14.48),
R
R
with the retention times (t ) of derivatives of the hydrolyzed 1 identical to those of
R
authentic D-mannitol (t ¼ 11.45) and D-glucose (t ¼ 14.51) prepared in the same
R
R
way as above.
3.5. Molecular docking study
Ensemble docking was performed for the identified compounds 1-16 against VEGFR-2.
Firstly, all compounds were pre-processed by the Ligprep, with Epik to generate the
proper protonation states at pH 7.0. Secondly, each ligand in ATP binding site of 26
VEGFR-2 structures (PDB IDs: 1y6b, 2oh4, 2p2h, 2p2i, 2qu5, 2qu6, 2rl5, 3b8q, 3b8r,
3
4
be2, 3c7q, 3cp9, 3cpb, 3cpc, 3ewh, 3u6j, 3vhe, 3vhk, 3hng, 3vnt, 3vo3, 4ag8, 4agc,
agd, 4asd, 4ase) was selected to define and generate the receptor grid. Thirdly, in sil-
ico docking of each compound against the 26 VEGFR-2 structures was performed by
Glide in standard precision (SP) with default parameters. Fourthly, for each compound,
the VEGFR-2 structure with the best docking score was chosen for further analysis. The
docking modes of compounds 10 and 14 against VEGFR-2 are shown in Figure S4 and
S5. All these computational calculations were performed in Schr o€ dinger 2020
(
Schr o€ dinger LLC, New York, NY, USA).
3.6. Vegfr-2 inhibition assay
VEGFR-2(Cat#08-191, Carna) was diluted with enzyme buffer. The VEGFR-2 Kinase
Assay Kit measures VEGFR-2 kinase activity for screening applications using HTRF

NATURAL PRODUCT RESEARCH
7
kinEASETM TK (Cat#62TKOPEC, Cisbio) as a detection reagent (Xu et al. 2020). The
assay was carried out according to the supplied manual with modifications. The per-
centage of inhibition was calculated with the following equation: (ERpositiveꢀER sample)/
(
ERpositiveꢀER
)ꢁ100%, where ER (emission ratio) is the ratio of 665 nm emission
negative
signal to 615 nm emission signal, and staurosporine was used as control. The 50%
inhibitory concentration (IC ) was obtained from the concentration-inhibition
50
response curve (n ¼ 2) comparing to staurosporine. Data were expressed
as mean ± SD.
3.7. The potential retinal protective activity in vivo
3.7.1. Animals
BALB/c mice (4-5 weeks old) were obtained from Shanghai Laboratory Animal Research
Center, China. Mice were maintained under a 12/12 h light/dark cycle with access to
food and water at libitum. After dark adaption for 24 h, bright light exposure (compact
fluorescence lamp, 45 W, Chaoya Lighting, Shanghai, China) was delivered at
1
0,000 lux for 30 min. 10 or 14 dissolved in H O was intraperitoneally administered to
2
the mice 30 min prior to bright light exposure at 100 mg/kg bw (n ¼ 4 per treatment).
Mice unexposed to bright light (n ¼ 4) and light-exposed mice (n ¼ 4) without 10 and
1
4 treatment received H O vehicle treatment administered in the same manner. The
2
animal care, handling and experimental protocols were reviewed and approved by the
Institutional Animal Care and Use Committee of Yueyang Hospital, Shanghai University
of TCM (Approval No. YYLAC-2019-021), which were carried out in adherence to the
National Institutes of Health guide for the care and use of Laboratory animals (NIH
Publications No. 8023, revised 1978).
3.7.2. Optical coherence tomography (OCT)
OCT (Micron IV, Phoenix Research Labs, USA) was carried out for in vivo imaging of
mouse retina (Drexler and Fujimoto 2008). Ketamine hydrochloride (82.5 mg/kg bw)
and xylazine (8.25 mg/kg bw) cocktail was intraperitoneally administered to the mice
as anesthetics and the pupils were dilated by 1% tropicamide before OCT imaging. To
quantify morphological changes, the OCT scans were subject to the measurement of
the thickness of the outer nuclear layer (ONL) using Phoenix Reveal OCT software
(Phoenix Research labs, USA). Data were expressed as mean ± standard deviation (SD).
The statistical analyses were performed by one-way ANOVA. Statistical significance was
defined by P < 0.05.
4. Conclusion
A new phenylethanoid, hebitol IV (1), along with fifteen known glycosides (2-16), were
isolated from water extract of the flower buds of Buddleja officinalis. Their chemical
structures were elucidated by MS, NMR spectrum data and comparisons with data
reported in the literature. Among them, 9 compounds (7, 9-16) showed VEGFR-2
inhibitory activities in vitro, and compounds 10 and 14 are potent natural inhibitors
against VEGFR-2. Further assay in vivo demonstrated remarkable photoreceptor

8
Z. LONG ET AL.
protective activities of the compounds 10 and 14 in the mouse model featuring bright
light-induced retinal degeneration.
Disclosure statement
No potential conflict of interest was reported by the authors.
Funding
This work was supported by the Natural Science Foundation of China (81603270, YZ; 81673790,
YC), the Shanghai Sailing Program (16YF1414100), which are gratefully acknowledged.
ORCID
Zhijian Xu
http://orcid.org/0000-0002-3063-8473
References
American Macular Degeneration Foundation. [date unknown]. Macular degeneration treatments.
https://www.macular.org/treatments.
Cali s¸ I, Kirmizibekmez H, Tasdemir D, Sticher O, Ireland CM. 2002. Sugar esters from Globularia
orientalis. Z Naturforsch C J Biosci. 57(7–8):591–596. c:
Chinese Pharmacopoeia Commission. 2020. Pharmacopoeia of the People’s Republic of China,
vol. 1, Beijing: China Medical Science Press; p. 343.
Drexler W, Fujimoto JG. 2008. State-of-the-art retinal optical coherence tomography. Prog Retin
Eye Res. 27(1):45–88.
Dugel PU, Koh A, Ogura Y, Jaffe GJ, Schmidt-Erfurth U, Brown DM, Gomes AV, Warburton J,
Weichselberger A, Holz FG. 2020. HAWK and HARRIER: Phase 3, multicenter, randomized, dou-
ble-Masked trials of brolucizumab for neovascular age-related macular degeneration.
Ophthalmology. 127(1):72–84.
Editorial Committee of flora of China. 1992. Chinese Academy of Sciences, Flora of China, vol.
61. Beijing: Science Press; p. 277.
Gao YY, Zeng P, Jia CL, Zhang Y, Zhang T, Jia Q, Li YM. 2017. Two new phenols from Lysimachia
patungensis. J Asian Nat Prod Res. 19(1):28–34.
Guo H, Koike K, Li W, Satou T, Guo D, Nikaido T. 2004. Saponins from the flower buds of
Buddleja officinalis. J Nat Prod. 67(1):10–13.
Jie CH, Gao JS. 2004. Study of the effect of traditional Chinese herb butterflybush flower on the
proliferation of vascular endothelial cells. Yanke. 13(6):348–350.
Kim JK, Si CL, Bae YS. 2007. Epimeric phenylpropanoid glycosides from inner bark of Paulownia
coreana Uyeki. Holzforschung. 61(2):161–164.
Lee C, Lee S, Park SY. 2013. A New Monoterpene from the Flower Buds of Buddleja officinalis.
Nat Prod Sci. 19(4):355–359.
Li HF, Peng QH, Yao XL, Wang F, Chen JW, Wu QL. 2009. Effects of Buddleja offcinalis flavone on
dry eye syndrome and the expression of IL-1b, TNF-a in castrated rats. Guoji Yanke Zazhi.
9(7):1248–1251.
Li HM, Kim JK, Jang JM, Cui CB, Lim SS. 2013. Analysis of the inhibitory activity of Abeliophyllum
distichum leaf constituents against aldose reductase by using high-speed counter current
chromatography. Arch Pharm Res. 36(9):1104–1112.
Li L, Peng Y, Liu Y, Xu LJ, Guo N, Shi RB, Xiao PG. 2011. Two new phenethanol glycosides from
Ligustrum robustum. Chin Chem Lett. 22(3):326–329.

NATURAL PRODUCT RESEARCH
9
Liu X, Li J, Li J, Liu Q, Xun M. 2019. A new flavonoid glycoside from Ligularia fischeri. Chem Nat
Compd. 55(4):638–641.
Liu YF, Shi GR, Wang X, Zhang CL, Wang Y, Chen RY, Yu DQ. 2016. Chemical constituents from
whole plants of Rehmannia chingii. Zhongcaoyao. 47(11):1830–1833.
Matsuda H, Cai H, Kubo M, Tosa H, Iinuma M. 1995. Study on anti-cataract drugs from natural
sources. II. Effects of buddlejae flos on in vitro aldose reductase activity. Biol Pharm Bull.
18(3):463–466.
Ono M, Yoshida A, Ito Y, Nohara T. 1999. Phenethyl alcohol glycosides and isopentenol glyco-
side from fruit of Bupleurum falcatum. Phytochemistry. 51(6):819–823.
Park TW, Lee C, Lee JW, Jang H, Jin Q, Lee MK, Hwang BY. 2016. Chemical constituents from
Buddleja officinalis and their inhibitory effects on nitric oxide production. Nat Prod Sci. 22(2):
129–133.
Porter EA, Kite GC, Veitch NC, Geoghegan IA, Larsson S, Simmonds MSJ. 2015. Phenylethanoid
glycosides in tepals of Magnolia salicifolia and their occurrence in flowers of Magnoliaceae.
Phytochemistry. 117:185–193.
Shen ZH, Zuo ZQ, Liu J. 2016. Effect of Chinese medicine Flos Buddlejae on vascular endothelial
growth factor of corneal neovascularization in alkali-burned mice. China J Chin Ophthalmol.
26(3):161–165.
Sheng GQ, Zhang JR, Pu XP, Ma J, Li CL. 2002. Protective effect of verbascoside on 1-methyl-4-
phenylpyridinium ion-induced neurotoxicity in PC12 cells. Eur J Pharmacol. 451(2):119–124.
Tai BH, Jung BY, Cuong NM, Linh PT, Tung NH, Nhiem NX, Huong TT, Anh NT, Kim JA, Kim SK,
et al. 2009. Total peroxynitrite scavenging capacity of phenylethanoid and flavonoid glyco-
sides from the flowers of Buddleja officinalis. Biol Pharm Bull. 32(12):1952–1956.
Tai BH, Nhiem NX, Quang TH, Ngan NT, Tung NH, Kim Y, Lee JJ, Myung CS, Cuong NM, Kim YH.
2011. A new iridoid and effect on the rat aortic vascular smooth muscle cell proliferation of
isolated compounds from Buddleja officinalis. Bioorg Med Chem Lett. 21(11):3462–3466.
Taskova RM, Kokubun T, Ryan KG, Garnock-Jones PJ, Jensen SR. 2011. Iridoid and phenyletha-
noid glucosides from Veronica lavaudiana. J Nat Prod. 74(6):1477–1483.
Venditti A, Sanna C, Lorenzetti LM, Ballero M, Bianco A. 2017. New coumarinyl ethers in Daphne
oleoides Schreb. collected from Sardinia Island. Chem Biodiversity. 14(6):e1700072.
World Health Organization. 2021. Blindness and vision impairment. https://www.who.int/en/
news-room/fact-sheets/detail/blindness-and-visual-impairment.
Xie G, Xu Q, Li R, Shi L, Han Y, Zhu Y, Wu G, Qin M. 2019. Chemical profiles and quality evalu-
ation of Buddleja officinalis flowers by HPLC-DAD and HPLC-Q-TOF-MS/MS. J Pharm Biomed
Anal. 164:283–295.
Xu HC, Wang MS, Wu FX, Zhuo LS, Huang W, She NF. 2020. Discovery of N-substituted-3-phenyl-
1,6-naphthyridinone derivatives bearing quinoline moiety as selective type II c-Met kinase
inhibitors against VEGFR-2. Bioorg Med Chem. 28(12):115555.
Yamasaki T, Masuoka C, Nohara T, Ono M. 2007. A new phenylethanoid glycoside from the fruits
of Callicarpa japonica Thunb. var. luxurians Rehd. J Nat Med. 61(3):318–322.
Yang Y, Yang YB, Lu WQ, Wu ZJ, Chen WS. 2017. A new hydroxyjasmonic acid derivative from
Polygonum capitatum. Chem Nat Compd. 53(3):417–421.
Zhang HY, Pan JX. 1996. Phenylpropanoid glycosides and flavonoid glycosides isolated from
buds of Buddleja officinalis Maxim. J Chin Pharm Sci. 5(2):105–110.