Inunicosides A?K, rare polyacylated ent-kaurane diterpenoid glycosides from the flowers of Inula japonica

Article abstract of DOI:10.1016/j.tet.2019.130732

Inunicosides A?K (1–11), eleven unusual polyacylated ent-kaurane diterpenoid glycosides, were isolated from the flowers of a traditional Chinese herbal plant Inula japonica. Their structures with absolute configurations were determined on the basis of comprehensive spectroscopic analyses, chemical degradation, enzymatic hydrolysis and ECD experiments. Among these isolates, inunicoside K (11) showed mild antibacterial activity against Staphylococcus aureus ATCC 25923. Diterpene constituents have been rarely reported in the Inula species, and thus the discovery of this panel of compounds greatly enriches the chemical diversity of metabolites from the genus Inula.

Full text of DOI:10.1016/j.tet.2019.130732

Journal Pre-proof
Inunicosides A−K, rare polyacylated ent-kaurane diterpenoid glycosides from the
flowers of Inula japonica
Zhi-Pu Yu, Jin-Hai Yu, Jun-Sheng Zhang, Shu-Juan Yu, Hua Zhang
PII:
S0040-4020(19)31118-4
https://doi.org/10.1016/j.tet.2019.130732
TET 130732
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 11 August 2019
Revised Date: 28 September 2019
Accepted Date: 22 October 2019
Please cite this article as: Yu Z-P, Yu J-H, Zhang J-S, Yu S-J, Zhang H, Inunicosides A−K, rare
polyacylated ent-kaurane diterpenoid glycosides from the flowers of Inula japonica, Tetrahedron (2019),
doi: https://doi.org/10.1016/j.tet.2019.130732.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier Ltd.

Graphical abstract:
Inunicosides A−K, rare polyacylated ent-kaurane
diterpenoid glycosides from the flowers of Inula japonica
Zhi-Pu Yu, Jin-Hai Yu, Jun-Sheng Zhang, Shu-Juan Yu, Hua Zhang

Inunicosides A−K, rare polyacylated ent-kaurane
diterpenoid glycosides from the flowers of Inula japonica
Zhi-Pu Yu,a,b Jin-Hai Yu,b Jun-Sheng Zhangb, Shu-Juan Yua,b, Hua Zhangb,*
aSchool of Chemistry and Chemical Engineering, University of Jinan, 336 West Road
of Nan Xinzhuang, Jinan 250022, China
bSchool of Biological Science and Technology, University of Jinan, 336 West Road of
Nan Xinzhuang, Jinan 250022, China
*Corresponding author
E-mail address: bio_zhangh@ujn.edu.cn (H. Zhang).
ORCID
Hua Zhang: 0000-0002-5959-6473
1

ABSTRACT
Inunicosides A−K (1−11), eleven unusual polyacylated ent-kaurane diterpenoid
glycosides, were isolated from the flowers of a traditional Chinese herbal plant Inula
japonica. Their structures with absolute configurations were determined on the basis
of comprehensive spectroscopic analyses, chemical degradation, enzymatic hydrolysis
and ECD experiments. Among these isolates, inunicoside K (11) showed mild
antibacterial activity against Staphylococcus aureus ATCC 25923. Diterpene
constituents have been rarely reported in the Inula species, and thus the discovery of
this panel of compounds greatly enriches the chemical diversity of metabolites from
the genus Inula.
Keywords: Compositae; Inula japonica; ent-kaurane; diterpenoid glycoside;
polyacylation; antibacterial
2

1. Introduction
The genus Inula (family Compositae) comprises about 100 species widely distributed
on the continents of Europe, Africa and Asia, particularly around the Mediterranean
area [1]. Over 20 species and a number of variants grow in China, with several ones
such as I. britanica and I. japonica being widespread species [1]. Inula is an
important tribe of medicinal plants, among which I. japonica is well known in
traditional Chinese medicine and its dried capitula have long been used under the
name ‘Xuanfuhua’ for the treatment of cough, phlegm and related symptoms [2]. The
plant I. japonica is very famous in the natural products field, especially for its
abundant and diverse sesquiterpenoid constituents, e.g. eudesmane, germacrane and
guaiane types, with monomers, dimers and even trimers [3-8]. In addition, four rare
diterpenoid glycosides were also reported from this herb [9].
During the course of our research to explore new bioactive molecules from Chinese
traditional herbs, the flower buds of I. japonica were phytochemically studied in the
current work. Apart from the well-investigated sesquiterpenoid compounds [10], new
interesting chemical constituents were also detected in the most polar fractions
generated from the EtOAc partition of the ethanol extract. Subsequently, an intensive
fractionation of these polar fractions led to the separation of a panel of diterpenoid
glycosides. Through a combination of various means including MS, NMR, ECD,
chemical hydrolysis and enzymatic degradation, these compounds were identified to
be ten new 18-nor-ent-kaurane diterpenoid glycosides (1−10) with a 4-carboxyl group
and a disaccharide moiety of glucose and arabinose at C-2, a new intact ent-kaurane
diterpenoid glycoside (11) with 4-geminal carboxyl groups and the same sugar
connection as 1−10, and a known diterpenoid glycoside analogue (12) [11]. The in
vitro cytotoxicity (MDA-MB-231 & HeLa), NO release inhibition (in RAW264.7
macrophages) and antibacterial (Staphylococcus aureus & bBacillus subtilis)
activities of all the isolates were evaluated in the current work. Moreover, their
acetylcholinesterase (AChE), α-glucosidase and pancreatic lipase inhibitory activities
were also tested. Details of the isolation, structure characterization and biological
evaluations of these glycosides are described below.
3

2. Results and discussion
Compound 1, a white amorphous powder, was assigned a molecular formula of
C39H58O16 based on its (+)-HRESIMS (m/z 805.3607 [M + Na]+, calcd 805.3617) and
13C NMR data, corresponding to 11 indices of hydrogen deficiency. The 1H NMR data
(Table 1) showed two doublet signals at δH 4.60 (d, J = 8.0 Hz) and 4.46 (d, J = 6.6
Hz) diagnostic of two sugar anomeric protons, which along with the corresponding
carbon resonances at δC 99.0 and 100.7 suggested the presence of two
1
monosaccharide units. Further analyses of the H−1H COSY, HSQC and HMBC data
(Fig. 2A) unambiguously assigned all their NMR signals as shown in Tables 1 and 2,
based on which the two sugars were determined be one glucopyranose and one
arabinopyranose, with β- and α-configurations for the two anomeric carbons being
determined via the coupling constants of anomeric protons (8.0 Hz for glucopyranose
and 6.6 Hz for arabinopyranose), respectively [12,13]. The assignments of
D-glucose
and -arabinose were established via HPLC comparison of their
L
(S)-(−)-1-phenylethylamine derivatives with those of authentic standard sugar samples
(Fig. S95, supporting information (SI)). In addition, the key HMBC correlation (Fig.
2A) from H-1′′ (δH 4.46) to C-3′ (δC 83.4) established the connection between the two
sugar moieties as shown. Characteristic signals for one isovaleryloxy [δH 2.24, 2.21,
2.11, 0.99, 0.98; δC 171.4, 43.7, 25.8, 22.8 (2C)] and two acetoxy [δH 2.11 (s) and δC
21.0, 171.2; δH 2.15 (s) and δC 21.2, 170.8] groups were also evident from the NMR
data, and they were assigned to locate at C-2′, C-2′′ and C-4′′ by the pivotal HMBC
correlations from H-2′ (δH 4.94), H-2′′ (δH 4.95) and H-4′′ (δH 5.06) to the
corresponding ester carbonyl carbons, respectively.
Besides those for the disaccharide moiety, the NMR data for 1 also displayed
typical signals for a carboxyl group (δC 179.3), an exocyclic terminal double bond (δH
5.21, 5.08; δC 160.0, 108.9), two sp3 oxymethines (δH 3.80, 4.23; δC 82.8, 71.3) and a
singlet methyl (δH 0.97; δC 16.9). The aforementioned data accounted for seven
indices of hydrogen deficiency, and the remaining four supported a tetracyclic
scaffold for the aglycone part of 1. Comprehensive analyses of 2D NMR data (Fig.
4

2A) confirmed the planar structure for the aglycone of 1 as drawn, incorporating a
kaurane or ent-kaurane type diterpenoid skeleton. In detail, two spin-spin coupling
systems [a (H2-1 to H2-7) and b (H-9 via H2-11 to H2-14), see Fig. 2A] as shown in
1
bold bonds were readily built up via examination of the H–1H COSY data. The
HMBC correlations from H3-20 to C-1 (δC 47.3), C-5 (δC 49.3), C-9 (δC 52.7) and
C-10 (δC 40.8), H2-7 to C-8 (δC 47.8) and C-9, and H-9 to C-8 (Fig. 2A), along with
fragment a, established rings A and B as depicted. Subsequently, the HMBC
correlations from H-9 to C-8 and C-14 (δC 36.4), in combination with fragment b,
assembled ring C, and ring D was then constructed via the HMBC interactions from
H2-17 to C-13 (δC 42.4), C-15 (δC 82.8) and C-16 (δC 160.0), and H-15 to C-7, C-8
and C-14. Furthermore, the presence of 15-OH was confirmed by the chemical shifts
of C-15 (δC 82.8) and H-15 (δH 3.80). Finally, the HMBC correlations from H-1′ to
C-2 (δC 71.3), and H2-3, H-4 and H-5 to C-19 (179.3), attached the disaccharide
moiety and the carboxyl group at C-2 and C-4, respectively.
The relative configuration for the aglycone unit of 1 was established by
interpretation of the ROESY data (Fig. 2B). The strong interactions of H3-20 with H-2,
H-6
same side of the molecule and were assigned to be
disposed on top of ring C as shown in Fig. 2B. Accordingly, the correlations of H-5
with H-3 δH 1.18), H-4, H-7 δH 1.43) and H-9, H-3 with H-1 δH 0.76), and
H-9 with H-15, revealed that they were on the other side and -directed, thus leaving
4-CO2H and 15-OH -bonded. Subsequent hydrolysis of 1 with snailase afforded the
α
(
δH 1.91), H-12
α
(
δH 1.57) and H-14
α
(
δH 1.86) indicated that they were on the
α
-oriented, with ring D being
β
(
β
(
β
β (
β
α
aglycone (1a), and it was identified to be atractyligenin (Table S4 and Figs. S96–S98,
SI) whose absolute configuration had been previously established via single-crystal
X-ray diffraction analysis [14]. Furthermore, the theoretical ECD spectrum of the
aglycone (atractyligenin) was also acquired by TD-DFT (time-dependent density
functional theory) method and matched the experimental one perfectly (Fig. 3). It was
interesting to note that the ECD curve of 1 was also highly consistent with that of the
aglycone (1a, Fig. 3), indicating that the sugar part contributed little or even no effect
5

to the ECD data. Therefore, compound 1 was unambiguously elucidated as an
18-nor-ent-kaurane type diterpenoid glycoside and it was named inunicoside A.
Compound 2 displayed an ammonium adduct ion peak at m/z 842.4188 (calcd
842.4169) in the (+)-HRESIMS analysis, which together with the 13C NMR data
suggested a molecular formula of C41H60O17, with 42 mass units more than that of 1
and suggestive of an extra acetyl group. Comparison of the NMR data (Tables 1 and 2)
for 2 with those of 1 revealed that they were structural analogues, with the only
difference occurring to the glucopyranosyl moiety, whose H2-6′ signals (δH 4.44, 4.24)
were remarkably downfield shifted as compared with those (δH 3.93, 3.76) for 1. In
addition, the appearance of NMR resonances for an additional acetoxyl group (δH
2.05; δC 21.1, 171.2) implied that the 6′-OH in 2 was acetylated, and this was further
confirmed by the HMBC correlations from H2-6′ to the newly introduced acetyl
carbonyl carbon. Detailed examination of the full 1D and 2D (Figs. S10−S15, SI)
NMR data for 2, especially of key proton-proton couplings, HMBC and ROESY
correlations, established the gross structure of 2 as drawn (Fig. 1) and also
corroborated the relative configuration of 2 to be identical with that of 1. Compound 2
was thus structurally characterized as the 6′-O-acetyl derivative of 1 and was named
inunicoside B.
Compounds 3 (inunicoside C) and 4 (inunicoside D) were assigned the same
molecular formula of C41H60O17 as 2, which was evidenced by their (+)-HRSEIMS
ion peaks at m/z 842.4193 and 842.4183 (both [M + NH4]+, calcd 842.4169),
respectively, as well as the 13C NMR data. Analysis of the NMR data (Tables 1 and 2)
for 3 confirmed that it was isomeric with 2, and the only difference between the two
cometabolites was attributed to the acetylation position within the arabinopyranosyl
fragment. Compared with those of 2, H-3′′ (δH 4.89) of 3 was significantly downfield
shifted while H-4′′ (δH 4.07) of 3 was observed in the normal range, suggesting that
the 3′′-OH in 3 instead of the 4′′-OH in 2 was acetylated. This conclusion was further
supported by the HMBC correlation from H-3′′ to an acetyl carbonyl carbon
(
δC170.5). Similarly, the difference between 4 and 2 was also determined to be the
acetylation location. As evidenced by the chemical shifts for H-4′ (δH 4.91) and H-6′
6

(
δH 3.69, 3.61) of 4 in comparison with other analogues, the 6′-O-acetyl group in 2
moved to C-4′ position in 4, and the 2′′,4′′-O-diacetyl units retained. The
aforementioned structural assignments for 3 and 4 were subsequently verified by
careful check of 2D NNR spectra (Figs. S21-S32, SI). The structures of compounds 3
and 4 were hence elucidated.
Analyses of the (+)-HRESIMS data for 5 (inunicoside E) and 6 (inunicoside F)
revealed ammonium adduct ion peaks at m/z 842.4177 and 842.4161 (both calcd
13
842.4169), respectively, which in combination with their C NMR data assigned the
same molecular formula of C41H60O17 as 2. As with the cases of 3–4, compounds 5
and 6 were also regio-isomers of 2 with different acetyl substitution patterns.
Inspection of the NMR data (Tables 1 and 2) uncovered that the major difference
between 5 and 4 occurred to the disaccharide moiety, where H-3′′ and H-4′′ of 5 were
observed at δH 4.87 and 4.01, respectively, while their counterparts of 4 appeared at
δH 3.80 and 5.01, respectively. These observations suggested that the 4′′-O-acetyl
substitution in 4 was replaced by the 3′′-O-acetyl substitution in 5. Similarly, the
difference between 6 and 4 was established that the 4′′-O-acetyl group in 4 moved to
C-6′ position in 6, as supported by the chemical shift variations between the
corresponding proton pairs of H-4′′ (δH 5.01 for 4 Vs. 3.85 for 6) and H2-6′ (δH 3.69
& 3.61 for 4 Vs. 4.19 & 4.15 for 6). Moreover, the structural assignments for 5 and 6
were also confirmed by detailed examination of 2D NMR data (Figs. S38−S49, SI).
(+)-HRESIMS analyses of compounds 7 (inunicoside G) and 8 (inunicoside H)
exhibited ammonium and sodium adduct ion peaks at m/z 884.4279 (calcd 884.4274)
889.3827 and (calcd 889.3828), respectively, which were consistent with the same
molecular formula of C43H62O18 and suggested a pairs of isomers with an additional
acetyl unit more than their cometabolites 2–6. Analysis of the NMR data (Tables 2
and 3) for 7 confirmed this hypothesis, with diagnostic signals for four acetoxy groups
at δH 2.13, 2.05, 2.04 and 2.00 (all s, 3H), as well as δC 171.1, 170.5, 170.4 and 169.8
(only carbonyl noted). Careful comparison between the NMR data for 7 and 2
indicated that the 3′′-OH of 7 was further acetylated as evidenced by the chemical
shifts for H-3′′ (δH 4.97 for 7 Vs. 3.77 for 2) of the two analogues, which was also
7

corroborated by the HMBC correlation from H-3′′ to the newly introduced acetyl
carbonyl carbon in 7. Similar to the case of 2 and 4, the difference between 8 and 7
was also attributed to that the acetylation pattern at 4′-OH and 6′-OH in the two
analogues was reversed, which was clearly revealed by the chemical shift changes
between the two proton pairs of H-4′ (δH 4.90 for 8 Vs. 3.54 for 7) and H2-6′ (δH 3.69
& 3.62 for 8 Vs. 4.45 & 4.23 for 7). Compounds 7 and 8 were thus characterized as
two penta-acylated diterpenoid glycosides with four acetyl and one isovaleryl groups
all located on the sugar part, which was further confirmed by cautious inspections of
2D NMR spectroscopic data (Figs. S55−S66, SI).
Compounds 9 and 10 (inunicosides I and J) possessed the same molecular formula of
C43H62O18 as determined by their respective (+)-HRESIMS ammonium adduct ion
13
peaks at m/z 884.4271 and 884.4280 (calcd 884.4274), as well as the C NMR data.
Detailed analyses of their NMR data (Tables 2 and 3) established that they were
isomeric with 7 and 8, and the structural differences were also assignable to the
acetylation pattern on the disaccharide moeity. As with the case of 2 and 3,
compounds 9 and 10 only differed from each other in that the acetylation mode at
3′′-OH and 4′′-OH was inverted in their structures, and this was unambiguously
demonstrated by the chemical shift variations between the two proton pairs of H-3′′
(
δH 3.78 for 9 Vs. 4.85 for 10) and H-4′′ (δH 5.00 for 9 Vs. 4.00 for 10). Therefore, the
two cometabolites had the same acylation pattern at all other positions except that 9
was 4′′-O-acetylated and 10 was 3′′-O-acetylated. Furhter examination of the full 2D
NMR spectra (Figs. S72−S84, SI) of 9 and 10 corroborated their strucural
assignments as shown.
Compound 11 (inunicoside K) was assigned a molecular formula of C46H64O21, as
deduced from the ammonium adduct ion peak at m/z 970.4284 [M + NH4]+ (calcd
13
970.4278) in the (+)-HRESIMS spectrum and the C NMR data. Analysis of the
NMR data (Tables 2 and 3) for 11 revealed that it was also a structural analogue of
compounds 1–10 with the differences being depicted below. Twenty carbon
resonances were observed for the aglycone moeity of 11, which indicated an intact
diterpenoid skeleton. Compared with 10, the absence of the CH-4 methine signals (δH
8

2.69 and δC 43.5 for 10), along with the appearance of a sp3 quaternary carbon (δC
59.4, C-4) and a carboxyl carbonyl (δC 176.0, C-18) resonances, suggested that Me-18
in 11 was also oxidized to a carboxyl functionality [11], which was further confirmed
1
by examination of 2D H–1H COSY and HMBC correlations (Fig. 4). As for the
disaccharide fragment, all hydroxyls were acylated with an isovaleryloxy group at
C-2′ and five acetoxy groups at C-4′, C-6′, C-2′′ C-3′′ and C-4′′, respectively, as also
supporpted by the HMBC correlations from these oxymethine protons to the
corresponding ester carbonyl carbons. Careful comparisons of the remaining NMR
data for 10 and 11, including key proton-proton couplings, supported common
structural features and identical relative configurations at all respective chiral centers
between the two cometabolites, which was then corroborated by interpretation of 2D
NMR data (Fig. 4). The structure of 11 was thereby characterized as shown, bearing
an interesting 4-geminal carboxyl motiff and full acylation of the sugar part. Based on
a biogenetic consideration and the highly consistent ECD curves of 2–11 (Fig. S99, SI)
with that of 1, the absolute configurations of 2–11 were also determined to be the
same as that of 1.
Compounds 1–12 were subjected to a series of bioassays (see experimental section).
The assay results showed that only compound 11 exhibited moderate antibacterial
activity against S. aureus ATCC 25923, with an IC50 value of 29.6 ± 3.5 µM, while
others were inactive in all tested models.
3. Conclusion
Plants of the genus Inula grow throughout Europe, Africa and Asia, and many
species are used as folk medicines by local residents [6]. This genus is very famous in
the literature for its rich sesquiterpenoid constituents and their diverse biological
properties especially anti-tumor and anti-inflammatory activities [6], whereas studies
on the diterpenoid ingredients remain very limited. On reviewing the literature, in
addition to the several diterpene quinones isolated from I. royleana in 1962 [15] and
1975 [16], the ent-kaurane diterpenoids from Chinese Inula species in later work
[9,17] all bear at least one carboxyl group and/or sugar moiety. Therefore, the
9

previously focused attention on sesquiterpenes and the separation difficulty due to
increased polarity of diterpenes could account for the limited diterpenoid reports on
Inula plants. As we know, triterpenoid saponins are widely distributed in many plant
species, while the diterpenoid glycosides present a much less occurrence in nature, so
the isolation of this array of ent-kaurane diterpenoid glycosides from I. japonica is
very interesting and increases the structural diversity of this class of natural products.
Moreover, the diverse acylation patterns of these glycosides indicate the presence of
some specific acylation enzymes in this species, which definitely deserves further
investigations from researchers of related areas.
4. Experimental
4.1. General experimental procedures
Optical rotations were measured on a Rudolph VI polarimeter (Rudolph Research
Analytical, Hackettstown, USA) with a 10 cm length cell. UV and ECD spectra were
obtained on a Chirascan Spectrometer (Applied Photophysics Ltd., Leatherhead, UK)
with a 1 mm pathway cell. NMR experiments were performed on a Bruker Avance
DRX600 spectrometer (Bruker BioSpin AG, Fällanden, Switzerland) and referenced
to residual solvent peaks (δC 49.00 and δH 3.31 ppm for CD3OD; δC 77.16 and δH 7.26
ppm for CDCl3). ESIMS analyses were carried out on an Agilent 1260-6460 Triple
Quad LC-MS instrument (Agilent Technologies Inc., Waldbronn, Germany).
HR-ESIMS data were acquired on an Agilent 6545 QTOF mass spectrometer (Agilent
Technologies Inc., Waldbronn, Germany). All HPLC analyses and separations were
performed on an Agilent 1260 series LC instrument (Agilent Technologies Inc.,
Waldbronn, Germany). D101-macroporous absorption resin (Sinopharm Chemical
Reagent Co. Ltd., Shanghai, China), CHP20/P120 MCI gel (75–150 µm, Mitsubishi
Chemical Co., Tokyo, Japan) and silica gel (300–400 mesh, Qingdao Marine
Chemical Co. Ltd., Qingdao, China) were used for column chromatography (CC).
Agilent ZORBAX SB-C18 columns (5 µm, 4.6 × 250 and 9.4 × 250 mm; Agilent
Technologies Inc., Santa Clara, USA) were used for HPLC analyses and separations.
All solvents used for CC were of analytical grade (Tianjin Fuyu Fine Chemical Co.
10

Ltd., Tianjin, China) and solvents used for HPLC were of HPLC grade (Oceanpak
Alexative Chemical Ltd., Goteborg, Sweden). All solvent mixtures used for analyses
and separations (HPLC & CC) were presented in the ratio of volume to volume,
unless otherwise specified.
4.2. Plant material
The flowers of Inula japonica Thunb. were collected in September 2016 in Mount
Kunyu of Shandong Province, China, and were authenticated by Prof. Guo-hua Ye
from Shandong College of Traditional Chinese Medicine. A voucher specimen has
been deposited at School of Biological Science and Technology, University of Jinan
(Accession number: npmc-003).
4.3. Extraction and isolation
The air-dried flowers of I. japonica (15 kg) were extracted with 95% EtOH at room
temperature for four times (one week per time) to afford a crude extract (1.5 kg). The
extract was suspended in 3.0 L water and then partitioned successively with EtOAc (3
× 2.0 L) and n-BuOH (3 × 2.0 L) to afford the corresponding layers. After evaporation
of solvents under reduced pressure, the EtOAc (750 g) and n-BuOH (120 g) soluble
parts were finally obtained.
The EtOAc partition was subjected to CC over D101 macroporous absorption resin,
eluted with EtOH-H2O (50%, 80% and 95%) to get three fractions. The 80% ethanol
elution was separated by MCI gel CC (MeOH-H2O, 50%, 80% and 100%) to afford
three fractions (A, B and C). Fraction A (50%, 60 g) was chromatographed on a silica
gel column (petroleum ether (PE)-acetone, 20:1 to 1:2) to give five subfractions (A1–
A5), and fraction A5 was separated by silica gel CC (PE-EtOAc, 5:1 to 1:2) to give
four subfractions (A5-1–A5-4). Fraction A5-1 (0.3 g) was subjected to passage over a
silica gel column, eluted with CHCl3-MeOH (100:1 to 30:1), to give three
subfractions, and the third one was then purified by semi-preparative HPLC (60%
MeCN-H2O) to afford compound 7 (6.8 mg, tR = 14.1 min). Using similar procedures,
fraction A5-3 (3.1 g) was separated into three subfractions (A5-3a–A5-3c), and each
of them was successively purified by semi-preparative HPLC (48% MeCN-H2O) to
11

yield compounds 2/9/10 (17.8/26.7/18.4 mg; tR = 16.1/22.3/27.1 min), 3 (17.2 mg, tR
= 18.2 min) and 8 (11.7 mg, tR = 20.0 min), respectively. Fraction A5-4 (3.6 g) was
fractionated by silica gel CC first with CHCl3-MeOH (100:1 to 30:1) and then with
PE-EtOAc (3:1 to 1:2), and the major elution containing the target compounds were
finally purified by semi-preparative HPLC (40% MeCN-H2O) to afford 1 (13.8 mg, tR
= 12.7 min), 4 (6.3 mg, tR = 19.5 min), 6 (20.2 mg, tR = 20.8 min), 5 (24.1 mg, tR =
24.7 min) and 11 (9.4 mg, tR = 36.4 min).
The n-BuOH partition was chromatographed on a silica gel column
(CH2Cl2-MeOH-H2O, 130:13:10) to furnish four subfractions (D1–D4). Fraction D3
was separated by silica gel CC (EtOAc-MeOH-H2O, 5:1:1) to give one major portion
which was further purified by semi-preparative HPLC (16% MeCN-H2O) to furnish
compound 12 (74.3 mg, tR = 30.1 min).
Inunicoside A (1): white amorphous powder; [α]
25 −39.1 (c 1.36, MeOH); ECD (c
D
1
13
0.73, MeOH)
λ (∆ε) 206 (5.63); H and C NMR data (CDCl3), see Tables 1 and 2;
(+)-ESIMS: m/z 805.3 [M + Na]+; (−)-ESIMS: m/z 817.3 [M + Cl]−; (+)-HRESIMS:
m/z 805.3607 [M + Na]+ (calcd for C39H58O16Na, 805.3617).
Inunicoside B (2): white amorphous powder; [α]
25 −27.6 (c 1.77, MeOH); ECD (c
D
1
13
0.73, MeOH)
λ (∆ε) 206 (6.37); H and C NMR data (CDCl3), see Tables 1 and 2;
(+)-ESIMS: m/z 842.3 [M + NH4]+; (−)-ESIMS: m/z 823.2 [M − H]−, 859.3 [M + Cl]−;
(+)-HRESIMS: m/z 842.4188 [M + NH4]+ (calcd for C41H64NO17, 842.4169).
Inunicoside C (3): white amorphous powder; [α]
25 −36.7 (c 1.70, MeOH); ECD (c
D
1
13
0.73, MeOH)
λ (∆ε) 206 (6.64); H and C NMR data (CDCl3), see Tables 1 and 2;
(−)-ESIMS: m/z 859.1 [M + Cl]−; (+)-HRESIMS: m/z 842.4193 [M + NH4]+ (calcd for
C41H64NO17, 842.4169).
Inunicoside D (4): white amorphous powder; [α]
25 −30.4 (c 0.62, MeOH); ECD (c
D
1
13
0.73, MeOH)
λ (∆ε) 206 (7.11); H and C NMR data (CDCl3), see Tables 1 and 2;
(+)-ESIMS: m/z 842.3 [M + NH4]+; (−)-ESIMS: m/z 823.2 [M − H]−, 859.3 [M + Cl]−;
(+)-HRESIMS: m/z 842.4183 [M + NH4]+ (calcd for C41H64NO17, 842.4169).
Inunicoside E (5): white amorphous powder; [α]
25 −37.9 (c 2.41, MeOH); ECD (c
D
12

1
13
0.73, MeOH)
λ (∆ε) 206 (6.49); H and C NMR data (CDCl3), see Tables 1 and 2;
(−)-ESIMS: m/z 823.3 [M − H]−, 859.3 [M + Cl]−; (+)-HRESIMS: m/z 842.4177 [M +
NH4]+ (calcd for C41H64NO17, 842.4169).
Inunicoside F (6): white amorphous powder; [α]
25 −73.6 (c 2.00, MeOH); ECD (c
D
1
13
0.73, MeOH)
λ (∆ε) 206 (6.50); H and C NMR data (CDCl3), see Tables 1 and 2;
(+)-ESIMS: m/z 842.3 [M + NH4]+; (−)-ESIMS: m/z 823.3 [M − H]−; (+)-HRESIMS:
m/z 842.4161 [M + NH4]+ (calcd for C41H64NO17, 842.4169).
Inunicoside G (7): white amorphous powder; [α]
25 −41.2 (c 0.65 MeOH); ECD (c
D
1
13
0.80, MeOH)
λ (∆ε) 206 (6.23); H and C NMR data (CDCl3), see Tables 2 and 3;
(−)-ESIMS: m/z 865.5 [M − H]−, 901.5 [M + Cl]−; (+)-HRESIMS: m/z 884.4279 [M +
NH4]+ (calcd for C43H66NO18, 884.4274).
Inunicoside H (8): white amorphous powder; [α]
25 −45.6 (c 1.17, MeOH); ECD (c
D
1
13
0.80, MeOH)
λ (∆ε) 206 (6.43); H and C NMR data (CDCl3), see Tables 2 and 3;
(+)-ESIMS: m/z 889.1 [M + Na]+; (−)-ESIMS: m/z 865.4 [M − H]−, 901.3 [M + Cl]−;
(+)-HRESIMS: m/z 889.3827 [M + Na]+ (calcd for C43H62O18Na, 889.3828).
Inunicoside I (9): white amorphous powder; [α]
25 −31.7 (c 2.66, MeOH); ECD (c
D
1
13
0.80, MeOH)
λ (∆ε) 206 (6.65); H and C NMR data (CDCl3), see Tables 2 and 3;
(+)-ESIMS: m/z 884.3 [M + NH4]+; (−)-ESIMS: m/z 865.3 [M − H]−, 901.2 [M + Cl]−;
(+)-HRESIMS: m/z 884.4271 [M + NH4]+ (calcd for C43H66NO18, 884.4274).
25
Inunicoside J (10): white amorphous powder; [
(c 0.80, MeOH)
α
]
−24.3 (c 1.83, MeOH); ECD
D
λ
(∆ε) 206 (6.88); 1H and 13C NMR data (CDCl3), see Tables 2 and 3;
(+)-ESIMS: m/z 884.3 [M + NH4]+; (+)-HRESIMS: m/z 884.4280 [M + NH4]+ (calcd
for C43H66NO18, 884.4274).
25
Inunicoside K (11): white amorphous powder; [
(c 0.85, MeOH)
α
]
−38.4 (c 1.24, MeOH); ECD
D
λ
(∆ε) 205 (6.67); 1H and 13C NMR data (CD3OD), see Tables 2 and
3; (−)-ESIMS: m/z 951.3 [M − H]−; (+)-HRESIMS: m/z 970.4284 [M + NH4]+ (calcd
for C46H68NO21, 970.4278).
4.4. Determination of
D-glucose and
L
-arabinose
13

A solution of ca. 1.0 mg 1 in 2.0 N HCl (3.0 mL) was stirred at 90 °C for 4 h. After
removal of excess HCl under reduced pressure, the residual aqueous mixture was
filtered to remove the aglycone part and then evaporated under reduced pressure to
afford the free monosaccharide. The monosaccharide was dissolved in ethanol and
then excess (S)-(−)-1-phenylethylamine and NaBH3CN were added. With a catalytic
amount of acetic acid added, the mixture was stirred at 40 °C for 3 h. After
evaporation of the solvent, the residue was stirred with acetic anhydride in pyridine at
room temperature for 5 h to obtain the amino derivative. Using the same method, the
amino derivatives of authentic D- and L-glucoses, D- and L-arabinoses were obtained.
These derivatives were subjected to HPLC analysis on an Agilent ZORBAX SB-C18
column (5 µm, 4.6 × 250 mm) with 52% MeCN-H2O as mobile phase at 1.00 mL/min.
As shown in Fig. S95 (SI), the retention times of standard
-arabinoses were observed at tR 7.158, 6.819, 6.053 and 5.953 min, respectively. The
retention times of the hydrolytic samples of 1 were observed at tR 7.174 and 5.967
D- and L-glucoses, D- and
L
min, respectively, and thus the D-glucose and L-arabinose were determined for 1.
4.5. Enzymatic hydrolysis of 1
Compound 1 (6.0 mg) and snailase (1.2 mg) were mixed in 2.0 mL sodium acetate
buffer solution (pH 5.0) and then stirred at 37 °C overnight. After the hydrolysis, 2.0
mL CHCl3 was added to the aforementioned buffer solution, mixed completely and
then layered. When solvents of this CHCl3 extract were evaporated under reduced
pressure, the aglycone (1a) was then obtained by purification on semi-preparative
HPLC (44% MeCN-H2O, 3.00 mL/min), with tR at 6.8 min. Atractyligenin (1a): white
amorphous powder; [α] λ (∆ε) 206
25 −131.2 (c 0.22, MeOH); ECD (c 0.31, MeOH)
D
(4.26); 1H and 13C NMR data (CD3OD), see Table S4 in SI; (−)-ESIMS: m/z 319.1 [M
− H]−, 355.1 [M + Cl]−.
4.6. ECD calculations
The ChemDraw_Pro_14.1 software with MM2 force field was used to establish the
initial conformations of target molecules. Conformational searches using mixed
torsional/Low-mode sampling method with MMFFs force field in an energy window
14

of 3.01 kcal/mol were carried out by means of the conformational search module in
the Maestro 10.2 software [18]. The re-optimization and the following TD-DFT
calculations of the re-optimized conformations were all performed with Gaussian 09
software at the B3LYP/6-311G(d,p) level in vacuo for 1 and 1a. Frequency analysis
was performed as well to confirm that the re-optimized conformers were at the energy
minima. Finally, the SpecDis 1.64 software [19] was used to obtain the
Boltzmann-averaged ECD spectra.
4.7 Bioassays
4.7.1 Cytotoxic assay
The cytotoxicity toward two human female cancer cell lines, MDA-MB-231 (breast)
and Hela (cervical), was screened by the MTT method as reported formerly [20]
. A
single concentration of 30 M was chosen for the primary screening. Adriamycin was
µ
used as the positive control with IC50 values of 0.67 ± 0.04 and 0.91 ± 0.11 µM for
MDA-MB-231 and HeLa cells, respectively.
4.7.2 NO production inhibitory assay
The inhibitory activity against LPS-induced NO production in murine RAW264.7
macrophages was performed by colorimetric assay based on the Griess reaction, as
previously described [21]. The initial screening concentration was set as 50 µM
(Table S2, SI). Dexamethasone was used as a positive control (IC50 = 15.3 ± 1.7 µM).
4.7.3 Anti-acetylcholinesterase assay
The AChE inhibitory activity was tested using modified Ellman′s method as we
described previously [22]. The preliminary screening concentration was set as 50 µM
(Table S3, SI). Galantamin was used as a positive control (IC50 = 0.85 ± 0.12 µM).
4.7.4
α
-Glucosidase inhibitory assay
-glucosidase inhibitory activity was examined by a previously described
The
α
method in our recent article [23]. All the isolates were first tested at a primary
concentration of 100 µM. Acarbose was as used as a control drug (IC50 = 883 ± 189
µM).
15

4.7.5 Antibacterial assay
The antibacterial activity against two Gram-positive trains, Staphylococcus aureus
ATCC 25923 and Bacillus subtilis ATCC 6633, was tested using liquid growth
inhibition method as we described previously [24]. The initial testing concentration
was set at 50 µM, and only compounds with >50% growth inhibition rate were further
subjected for IC50 measurements. Penicillin was used as the positive control (IC50 =
0.12 ± 0.03 and 0.23 ± 0.02 µM for S. aureus and B. subtilis, respectively).
4.7.6 Pancreatic lipase inhibitory assay
Pancreatic lipase inhibitory activity was evaluated referring to the procedures as
described in the literature [25]. All the compounds was screened at the initial
concentration of 50 µM. Orlistat was used as a positive control (IC50 = 0.84 ± 0.32
µM).
Conflict of interest
The authors declare no conflicts of interest.
Acknowledgements
This project was financially supported by Natural Science Foundation of Shandong
Province [No. JQ201721], the Young Taishan Scholars Program [No. tsqn20161037],
and Shandong Talents Team Cultivation Plan of University Preponderant Discipline
[No. 10027]. We also thank Prof. Guo-hua Ye from Shandong College of Traditional
Chinese Medicine for the identification of the plant material.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://xxx
References
[1]
Flora of China Editorial Committee of Chinese Academy of Sciences, The
Flora of China, Vol. 75, Science Press, Beijing, 1999, p. 263.
National Pharmacopoeia Committee, in: Pharmacopoeia of the People's
Republic of China, Part 1, China Medical Science and Technology Press,
16
[2]

Beijing, 2015, p. 325.
[3]
Q.H. Jin, J.W. Lee, H. Jang, J.E. Choi, D.H. Lee, J.T. Hong, Y.S. Kim, M.K.
Lee, B.Y. Hwang, J. Nat. Prod. 79 (2016) 1548−1553.
[4] Y. Li, Z.Y. Ni, M.C. Zhu, M. Dong, S.M. Wang, Q.W. Shi, M.L. Zhang, Y.F.
Wang, C.H. Huo, H. Kiyota, B. Cong, Z NATURFORSCH C. 67 (2012)
375−380.
[5] J.J. Qin, L.Y. Wang, J.X. Zhu, H.Z. Jin, J.J. Fu, X.F. Liu, H.L. Li, W.D. Zhang,
Chem. Commun. 47 (2011) 1222−1224.
[6]
[7]
[8]
[9]
G.W. Wang, J.J. Qin, X.R. Cheng, Y.H. Shen, L. Shan, H.Z. Jin, W.D. Zhang,
Expert Opin. Invest. Drugs 23 (2014) 317−345.
X.D. Wu, L.F. Ding, W.C. Tu, H. Yang, J. Su, L.Y. Peng, Y. Li, Q.S. Zhao,
Phytochemistry 129 (2016) 68−76.
J.J. Qin, H.Z. Jin, J.X. Zhu, J.J. Fu, Q. Zeng, X.R. Cheng, Y. Zhu, L. Shan,
S.D. Zhang, Y.X. Pan, W.D. Zhang, Tetrahedron 66 (2010) 9379−9388.
J.J. Qin, J.X. Zhu, W.D. Zhang, Y. Zhu, J.J. Fu, X.H. Liu, H.Z. Jin, Arch.
Pharm. Res. 32 (2009) 1369−72.
[10] Z.P. Yu, J.S. Zhang, Q.Q. Zhang, S.J. Yu, Y.Y. Zhang, J.H. Yu, H. Zhang,
Fitoterapia (2019). https://doi.org/10.1016/j.fitote.2019.104292
[11] R. Lang, T. Fromme, A. Beusch, A. Wahl, M. Klingenspor, T. Hofmann,
Phytochemistry (Elsevier) 93 (2013) 124−135.
[12] S.Y. Lee, W.S. Suh, H.K. Kim, I.K. Lee, S.U. Choi, K.H. Kim, K.R. Lee,
Heterocycles 91 (2015) 1187-1197.
[13] Z.P. Yu, Y.Y. Wang, S.J. Yu, J. Bao, J.H. Yu, H. Zhang, Fitoterapia 135 (2019)
99-106.
[14] F. Brucoli, M.T. Borrello, P. Stapleton, G.N. Parkinson, S. Gibbons, J. Nat.
Prod. 75 (2012) 1070−1075.
[15] O.E. Edwards, G. Feniak, M. Los, Can. J. Chem. 40 (1962) 1540−1546.
[16] S.V. Bhat, P.S. Kalyanaraman, H. Kohl, N.J. De Souza, Tetrahedron 31 (1975)
1001−1004.
[17] Y. Shao, N.S. Bai, B.N. Zhou, Phytochemistry 42 (1996) 783−786.
17

[18] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R.
Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H.
Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G.
Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.
Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J.A. Montgomery, Jr., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E.
Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K.
Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N.
Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A.
Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.B.
Foresman, J.V. Ortiz, J. Cioslowski, and D.J. Fox, Gaussian 09 Revision B.01,
Gaussian Inc, Wallingford CT, 2009.
[19] T. Bruhn, A. Schaumloeffel, Y. Hemberger, G. Bringmann, Chirality 25 (2013)
243−249.
[20] M. He, L. Jiang, B. Li, G. Wang, J. Wang, Y. Fu, Oncotarget 8 (2017)
65100−65110.
[21] J.J. Zhao, Y.Q. Guo, D.P. Yang, X. Xue, Q. Liu, L.P. Zhu, S. Yin, Z.M. Zhao, J.
Nat. Prod. 79 (2016) 2257−2263.
[22] Z.Q. Cheng, K.K. Zhu, J. Zhang, J.L. Song, L.A. Muehlmann, C.S. Jiang, C.L.
Liu, H. Zhang, Bioorg. Chem. 83 (2019) 277−288.
[23] J. Sun, J.H. Yu, J.S. Zhang, X.Q. Song, J. Bao, H. Zhang, Chem. Biodiversity
16 (2019). https://doi.org/10.1002/cbdv.201800581.
[24] J. Bao, H.J. Zhai, K.K. Zhu, J.H. Yu, Y.Y. Zhang, Y.Y. Wang, C.S. Jiang, X.
Zhang,
Y.
Zhang,
H.
Zhang,
Mar.
Drugs
16
(2018).
https://doi.org/10.3390/md16050174.
[25] X.C. Jia, D. Yang, H.H. Xie, Y.M. Jiang, X.Y. Wei, J. Funct. Foods 32 (2017)
419−425.
18

Table 1. 1H NMR spectroscopic data of compounds 1−6 in CDCl3.
No
1
2
3
2.27, brdd (12.5, 4.9)
0.79, dd (12.5, 11.5)
4.12, m
4
5
6
1
1
2
3
3
4
5
6
6
7
7
9
α
2.19, brdd (12.2, 5.0)
0.76, dd (12.2, 12.0)
4.23, m
2.29, brdd (13.0, 5.2)
0.79, dd (13.0, 10.8)
4.12, m
2.22, m
2.21, m
2.27, m
β
0.79, t (12.1)
4.24, m
0.77, t (12.1)
4.24, m
0.81, t (12.1)
4.13, m
α
β
2.42, dt (12.0, 2.7)
1.18, td (12.0, 5.3)
2.68, m
2.33, dt (12.4, 2.6)
1.22, td (12.4, 4.9)
2.68, m
2.33, m
2.41, m
2.41, dt (12.0, 2.7)
1.17, td (12.0, 5.2)
2.69, m
2.35, dt (12.1, 2.5)
1.23, td (12.1, 5.4)
2.70, m
1.21, td (12.0, 5.4)
2.68, m
1.20, m
2.70, m
1.44, m
1.45, m
1.43, m
1.45, m
1.43, m
1.45, m
α
β
α
β
1.91, m
1.88, m
1.88, m
1.91, m
1.90, m
1.88, m
1.64, m
1.64, m
1.63, m
1.65, m
1.64, m
1.65, m
1.73, dt (13.4, 3.2)
1.43, m
1.73, dt (13.2, 3.1)
1.43, m
1.72, dt (13.2, 2.9)
1.42, m
1.74, dt (13.5, 3.1)
1.44, m
1.73, dt (13.4, 3.2)
1.43, m
1.74, dt (13.4, 3.3)
1.44, m
1.03, brd (8.1)
1.54, m
1.03, brd (8.0)
1.56, m
1.03, brd (8.0)
1.58, m
1.04, brd (8.0)
1.54, m
1.02, brd (8.1)
1.54, m
1.04, brd (8.1)
1.56, m
11
α
β
11
1.41, m
1.41, m
1.40, m
1.43, m
1.42, m
1.43, m
12
12
13
14
14
15
α
1.57, m
1.57, m
1.55, m
1.56, m
1.56, m
1.56, m
β
1.48, m
1.47, m
1.48, m
1.48, m
1.47, m
1.47, m
2.74, m
2.73, m
2.73, m
2.75, m
2.74, m
2.74, m
α
β
1.86, brd (11.8)
1.39, dd (11.8, 5.0)
3.80, s
1.83, brd (11.7)
1.38, dd (11.7, 5.0)
3.80, s
1.83, brd (11.5)
1.38, dd (11.5, 4.9)
3.80, s
1.85, brd (11.8)
1.39, dd (11.8, 5.2)
3.81, s
1.85, brd (11.7)
1.39, dd (11.7, 5.3)
3.81, s
1.84, brd (11.5)
1.39, dd (11.5, 4.9)
3.81, s
17a
17b
20
5.21, brs
5.21, brs
5.21, brs
5.21, brs
5.21, brs
5.21, brs
5.08, brs
5.08, brs
5.07, brs
5.08, brs
5.08, brs
5.08, brs
0.97, s
0.96, s
0.96, s
0.97, s
0.96, s
0.96, s
1′
4.60, d (8.0)
4.94, dd (9.1, 8.0)
3.64, dd (9.2, 9.1)
3.56, dd (9.2, 9.2)
3.42, m
4.51, d (8.0)
4.90, dd (9.4, 8.0)
3.59, dd (9.4, 9.2)
3.52, dd (9.6, 9.2)
3.51, m
4.50, d (8.1)
4.89, dd (9.3, 8.1)
3.62, dd (9.3, 8.8)
3.51, m
4.60, d (8.0)
4.99, dd (9.2, 8.0)
3.87, dd (9.3, 9.2)
4.91, dd (9.7, 9.3)
3.51, m
4.59, d (8.0)
4.95, dd (9.4, 8.0)
3.84, dd (9.4, 9.1)
4.91, dd (9.6, 9.1)
3.51, m
4.55, d (8.0)
4.97, dd (9.3, 8.0)
3.83, dd (9.3, 8.8)
4.98, dd (8.8, 8.4)
3.67, m
2′
3′
4′
5′
3.51, m
6′a
6′b
1′′
3.93, dd (11.9, 3.1)
3.76, dd (11.9, 5.7)
4.46, d (6.6)
4.95, dd (7.6, 6.6)
3.80, dd (7.6, 3.9)
5.06, m
4.44, dd (11.9, 1.8)
4.24, dd (11.9, 5.0)
4.41, d (7.1)
4.96, dd (9.3, 7.1)
3.77, dd (9.3, 3.7)
5.06, m
4.45, dd (12.0, 2.3)
4.22, dd (12.0, 6.2)
4.43, d (7.2)
5.19, dd (9.6, 7.2)
4.89, dd (9.6, 3.3)
4.07, m
3.69, dd (12.5, 2.3)
3.61, dd (12.5, 5.8)
4.53, d (5.2)
4.79, dd (7.6, 5.2)
3.80, dd (7.6, 3.3)
5.01, m
3.69, dd (12.4, 2.1)
3.61, dd (12.4, 5.1)
4.47, d (5.3)
4.99, dd (7.6, 5.3)
4.87, dd (7.6, 3.5)
4.01, m
4.19, dd (12.2, 5.2)
4.15, dd (12.2, 2.5)
4.51, d (5.2)
4.76, dd (6.4, 5.2)
3.69, dd (6.4, 3.8)
3.85, m
2′′
3′′
4′′
5′′a
5′′b
2′′′a
2′′′b
3′′′
4′′′
5′′′
4′-OAc
6′-OAc
4.14, dd (13.0, 3.8)
3.60, dd (13.0, 2.1)
2.24, dd (14.9, 7.4)
2.21, dd (14.9, 7.1)
2.11, m
4.11, dd (13.3, 2.9)
3.60, dd (13.3, 1.8)
2.24, dd (14.9, 7.1)
2.21, dd (14.9, 7.1)
2.11, m
4.07, dd (13.7, 3.2)
3.62, dd (13.7, 2.2)
2.22, dd (15.1, 7.2)
2.20, dd (15.1, 7.0)
2.10, m
3.98, dd (12.5, 5.6)
3.51, dd (12.5, 2.9)
2.26, dd (15.1, 7.1)
2.22, dd (15.1, 7.0)
2.11, m
3.89, dd (12.0, 5.5)
3.51, dd (12.0, 2.7)
2.25, dd (15.1, 7.0)
2.21, dd (15.1, 7.0)
2.10, m
3.79, dd (12.0,6.2)
3.50, dd (12.0, 3.2)
2.27, dd (15.1, 7.0)
2.22, dd (15.1, 7.1)
2.12, m
0.99, d (6.7)
0.98, d (6.6)
0.99, d (6.6)
0.98, d (6.6)
0.99, d (6.6)
0.98, d (6.6)
0.99, d (6.7)
0.98, d (6.6)
2.05, S
0.98, d (6.6)
0.97, d (6.6)
2.09, S
0.98, d (6.7)
0.98, d (6.6)
2.07, s
2.05, s
2.10, s
2.05, s
2.03, s
2.09, s
2.05, s
2′′-OAc 2.11, s
3′′-OAc
2.10, s
2.13, s
2.03, s
2.09, s
2.08, s
4′′-OAc 2.15, s
2.15, s
19

Table 2. 13C NMR spectroscopic data of compounds 1−11.
No.
1
1a
2a
3a
4a
5a
6a
7a
8a
9a
10a
11b
47.3
71.3
34.2
43.5
49.3
25.4
34.9
47.8
52.7
40.8
18.3
32.6
42.4
36.4
82.8
160.0
108.9
47.5
73.3
34.2
43.6
49.1
25.4
34.9
47.7
52.6
40.7
18.3
32.6
42.3
36.4
82.7
159.9
108.9
47.4
73.2
34.2
43.6
49.1
25.4
34.9
47.7
52.7
40.7
18.3
32.6
42.34
36.4
82.7
159.9
108.9
47.2
71.7
33.9
43.4
49.3
25.4
34.9
47.8
52.7
40.8
18.3
32.5
42.4
36.4
82.8
160.0
108.9
47.2
71.4
33.9
43.5
49.3
25.4
34.9
47.7
52.7
40.8
18.3
32.5
42.4
36.4
82.8
160.0
108.9
47.4
71.8
34.2
43.5
49.1
25.4
34.9
47.7
52.6
40.7
18.3
32.6
42.3
36.4
82.7
159.9
109.0
47.5
73.3
34.2
43.6
49.1
25.4
34.9
47.7
52.6
40.7
18.3
32.6
42.3
36.4
82.7
159.9
108.9
47.2
71.6
33.8
43.5
49.3
25.4
34.9
47.7
52.7
40.8
18.3
32.5
42.3
36.4
82.8
160.0
108.9
47.4
73.3
34.3
43.6
49.1
25.4
34.9
47.7
52.7
40.7
18.3
32.6
42.4
36.4
82.7
160.0
108.9
47.4
73.3
34.1
43.5
49.0
25.4
34.8
47.7
52.6
40.7
18.3
32.5
42.3
36.4
82.7
159.8
108.9
48.5
74.0
41.1
59.4
52.2
24.1
36.2
48.9
54.9
41.4
19.3
33.6
43.6
37.3
83.4
160.2
109.1
176.0
175.9
17.7
100.8
74.2
80.6
70.0
72.9
63.4
102.2
70.6
71.7
69.3
64.0
173.0
44.4
26.7
23.0
23.0
20.6
171.6
21.1
172.4
20.7
171.3
20.7
171.7
20.7
171.8
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
1′
179.3
16.9
99.0
72.2
83.4
69.4
75.8
62.8
100.7
72.0
70.2
70.2
62.8
171.4
43.7
25.8
22.8
22.8
180.2
16.8
100.3
72.3
83.9
68.9
73.6
63.6
101.1
72.1
70.5
70.5
63.5
171.2
43.7
25.7
22.8
22.8
180.1
16.8
100.3
72.3
83.6
69.1
73.6
63.7
101.4
67.0
72.7
66.8
65.9
171.2
43.6
25.7
22.8
22.8
179.2
16.9
98.8
72.8
80.2
69.6
74.4
62.0
100.4
72.1
69.6
69.5
60.9
171.4
43.6
25.7
22.8
22.8
21.0
170.4
179.3
16.8
98.7
72.7
80.7
69.7
74.4
61.9
100.7
69.3
71.9
65.7
63.9
171.3
43.5
25.7
22.8
22.8
21.2
170.9
180.0
16.8
100.0
72.7
80.5
69.3
73.4
62.7
100.2
72.1
70.6
66.4
63.0
171.6
43.5
25.6
22.8
22.8
21.2
170.3
21.0
171.0
21.0
171.2
180.4
16.8
100.3
72.3
83.9
68.8
73.6
63.6
101.6
68.7
70.3
67.7
64.3
171.2
43.6
25.7
22.8
22.8
179.5
16.8
98.9
72.8
80.3
69.7
74.5
62.0
100.7
69.2
69.4
66.9
61.7
171.2
43.5
25.7
22.8
22.7
21.1
170.4
179.5
16.8
100.0
72.7
80.3
69.3
71.8
62.7
100.5
72.2
69.7
69.6
61.1
171.4
43.6
25.7
22.8
22.8
21.0
169.8
21.0
171.0
21.0
171.2
180.0
16.7
100.1
72.7
80.6
69.2
71.8
62.7
100.8
69.4
72.0
65.7
64.2
171.2
43.4
25.6
22.8
22.8
21.2
170.3
21.0
171.0
20.8
169.8
21.1
170.8
2′
3′
4′
5′
6′
1′′
2′′
3′′
4′′
5′′
1′′′
2′′′
3′′′
4′′′
5′′′
4′-OAc
6′-OAc
21.1
21.0
20.8
171.2
20.9
171.2
20.8
171.1
21.1
2′′-OAc 21.0
171.2
21.1
20.9
20.9
171.4
170.1
21.1
170.1
169.8
21.1
169.8
20.7
169.6
20.8
3′′-OAc
170.5
170.8
170.4
21.1
170.6
21.1
4′′-OAc 21.2
21.2
21.2
21.2
170.8
170.9
170.8
170.5
170.4
170.7
a Data were measured in CDCl3. b Data were measured in methanol-d4.
20

Table 3. 1H NMR spectroscopic data of compounds 7−11.
No.
7a
8a
9a
10a
11b
1
1
2
3
3
4
5
6
6
7
7
9
α
2.29, m
2.21, m
2.28, m
2.28, m
2.30, m
β
0.79, t (12.1)
4.12, m
0.78, t (12.1)
4.23, m
0.81, t (12.1)
4.15, m
0.81, t (12.0)
4.13, m
0.80, t (12.1)
4.15, m
α
β
2.32, dt (11.9, 2.6)
1.24, td (11.9, 5.4)
2.69, m
2.40, dt (11.9, 2.7)
1.18, td (11.9, 5.4)
2.69, m
2.35, m
2.33, dt (12.5, 2.6)
1.22, td (12.5, 4.9)
2.69, m
2.60, dd (12.5, 4.6)
1.32, dd (12.5, 11.2)
1.23, m
2.69, m
1.44, m
1.44, m
1.44, m
1.44, m
1.79, m
α
β
α
β
1.88, m
1.90, m
1.89, m
1.88, m
1.85, m
1.64, m
1.64, m
1.65, m
1.64, m
1.67, m
1.73, dt (13.3, 3.0)
1.42, m
1.74, dt (13.3, 3.1)
1.43, m
1.74, dt (13.3, 3.2)
1.44, m
1.73, dt (13.3, 2.9)
1.43, m
1.65, m
1.44, m
1.03, brd (8.0)
1.56, m
1.03, brd (8.1)
1.54, m
1.04, brd (8.0)
1.56, m
1.03, brd (8.0)
1.55, m
1.12, brd (7.7)
1.63, m
11
α
β
11
1.41, m
1.42, m
1.41, m
1.40, m
1.49, m
12
12
13
14
14
15
16
α
1.56, m
1.56, m
1.56, m
1.55, m
1.63, m
β
1.48, m
1.47, m
1.48, m
1.48, m
1.49, m
2.73, m
2.74, m
2.74, m
2.73, m
2.71, m
α
β
1.83, brd (11.5)
1.38, dd (11.5, 5.7)
3.80, s
1.85, brd (11.5)
1.39, dd (11.5, 5.7)
3.80, s
1.84, brd (11.6)
1.39, dd (11.6, 4.8)
3.81, s
1.82, brd (11.5)
1.38, dd (11.5, 5.2)
3.80, s
1.87, m
1.39, dd (11.6, 4.9)
3.78, s
17a
17b
20
5.21, brs
5.21, brs
5.22, brs
5.21, brs
5.18, brs
5.07, brs
5.08, brs
5.08, brs
5.08, brs
5.08, brs
0.96, s
0.96, s
0.97, s
0.95, s
1.01, s
1′
4.50, d (8.1)
4.58, d (8.0)
4.51, d (8.0)
4.99, dd (9.1, 8.0)
3.83, dd (9.3, 9.1)
4.97, dd (9.7, 9.3)
3.67, m
4.52, d (8.0)
4.93, dd (9.4, 8.0)
3.79, dd (9.4, 9.1)
4.95, dd (9.8, 9.1)
4.12, m
4.68, d (8.1)
2′
4.90, dd (9.3, 8.1)
3.60, dd (9.3, 8.5)
3.54, dd (9.4, 8.5)
3.51, m
4.96, dd (9.1, 8.0)
3.86, dd (9.3, 9.1)
4.90, dd (9.7, 9.3)
3.51, m
4.84, dd (9.6, 8.1)
4.00, dd (9.6, 9.1)
4.90, dd (10.2, 9.1)
3.80, ddd (10.2, 5.3, 2.2)
4.23, dd (12.2, 5.3)
4.12, dd (12.2, 2.2)
4.58, d (6.9)
3′
4′
5′
6′a
6′b
1′′
4.45, dd (11.8, 1.9)
4.23, dd (11.8, 5.8)
4.42, d (7.5)
3.69, dd (12.4, 2.5)
3.62, dd (12.4, 5.9)
4.49, d (5.2)
4.18, dd (12.3, 5.0)
4.15, dd (12.3, 2.8)
4.50, d (5.4)
4.79, dd (7.7, 5.4)
3.78, dd (7.7, 3.5)
5.00, m
4.17, dd (11.2, 4.9)
4.15, dd (11.2, 2.0)
4.44, d (5.5)
4.98, dd (7.7, 5.5)
4.85, dd (7.7, 3.5)
4.00, m
2′′
5.19, dd (9.9, 7.5)
4.97, dd (9.9, 3.5)
5.25, m
4.95, dd (7.7, 5.2)
4.97, dd (7.7, 3.3)
5.18, m
4.94, dd (9.3, 6.9)
5.02, dd (9.3, 3.5)
5.20, m
3′′
4′′
5′′a
5′′b
2′′′a
2′′′b
3′′′
4′′′
5′′′
4′-OAc
6′-OAc
4.04, dd (13.2, 2.7)
3.67, dd (13.2, 1.4)
2.22, dd (14.9, 7.2)
2.19, dd (14.9, 7.0)
2.10, m
3.95, dd (12.5, 5.3)
3.54, dd (12.5, 2.5)
2.24, dd (14.9, 7.1)
2.20, dd (14.9, 7.2)
2.10, m
3.97, dd (12.5, 5.5)
3.50, dd (12.5, 2.8)
2.27, dd (15.1, 7.1)
2.22, dd (15.1, 7.1)
2.13, m
3.88, dd (12.1, 5.3)
3.50, dd (12.1, 2.5)
2.25, dd (15.2, 6.9)
2.19, dd (15.2, 7.2)
2.10, m
3.93, dd (13.1, 3.4)
3.72, dd (13.1, 1.9)
2.31, dd (15.1, 7.4)
2.28, dd (15.1, 6.8)
2.10, m
0.99, d (6.7)
0.98, d (6.7)
0.99, d (6.6)
0.98, d (6.6)
2.02, s
0.97, d (6.7)
0.97, d (6.6)
2.02, s
0.99, d (6.7)
0.98, d (6.6)
0.97, d (6.6)
0.99, d (6.6)
2.06, s
1.96, s
2.05, s
2.06, s
2.06, s
2.05, s
2′′-OAc 2.04, s
3′′-OAc 2.00, s
4′′-OAc 2.13, s
2.04, s
2.03, s
2.09, s
2.10, s
2.04, s
2.02, s
2.06, s
2.10, s
2.13, s
2.05, s
a Data were measured in CDCl3. b Data were measured in methanol-d4.
21

R3
13
17
11
9
R6
20
10
5'
O
R2
1
16
5''
O
O
8
O
R5
O 1'
3'
1''
15
H
3''
5
R4
3
7
OH
2'''
5'''
O
R1
O
H
4'''
3'''
19
OH
R5
R1
H
H
H
H
H
H
H
H
H
H
R2
R3
R4
R6
1
2
3
4
5
6
7
8
9
OH OH OAc OH OAc
OH OAc OAc OH OAc
OH OAc OAc OAc OH
OAc OH OAc OH OAc
OAc OH OAc OAc OH
OAc OAc OAc OH OH
OH OAc OAc OAc OAc
OAc OH OAc OAc OAc
OAc OAc OAc OH OAc
OAc OAc OAc OAc OH
10
11 CO2H OAc OAc OAc OAc OAc
OH
O
O
HO
HO
H
OH
OH
H
O
OH
12
Fig. 1. Structures of compounds 1−12.
22

O
OH
b
13
6'
11
20
O
C
5'
17
14
O
1
5''
D
16
HO
O
O
8
O
3'
1'
HO
10
5
O
1'''
1''
O
3''
A
B
15
3
OH
a
2'''
3'''
6
O
O
19
HO
O
4'''
5'''
A
1
2
11
15
3
17
10
9
16
5
7
12
8
6
13
14
20
19
: Me
B
Fig. 2. Key 1H−1H COSY (
), HMBC (
) and ROESY (
) correlations for 1
(for clarity, H-11β and H-12β were not shown).
23

Fig. 3. Experimental and calculated ECD spectra for 1a, and experimental ECD
spectrum for 1.
24

1
Fig. 4. Key H−1H COSY (
), HMBC (
) and ROESY (
) correlations for
11.
25

Highlights
1. Eleven unusual 18-norditerpenoid glycosides were isolated from the flowers of
Inula japonica.
2. The absolute configurations were determined via chemical derivatization,
enzymatic hydrolysis and ECD analysis.
3. An intact diterpenoid glycoside with 4-geminal carboxyl groups showed moderate
antibacterial activity.
4. The discovery of these compounds enriches the chemical diversity of metabolites
from Inula species.