Isolation and Structural Characterization of Echinocystic Acid Triterpenoid Saponins from the Australian Medicinal and Food Plant Acacia ligulata

Article abstract of DOI:10.1021/acs.jnatprod.7b00437

The Australian plant Acacia ligulata has a number of traditional food and medicinal uses by Australian Aboriginal people, although no bioactive compounds have previously been isolated from this species. Bioassay-guided fractionation of an ethanolic extract of the mature pods of A. ligulata led to the isolation of the two new echinocystic acid triterpenoid saponins, ligulatasides A (1) and B (2), which differ in the fine structure of their glycan substituents. Their structures were elucidated on the basis of 1D and 2D NMR, GC-MS, LC-MS/MS, and saccharide linkage analysis. These are the first isolated compounds from A. ligulata and the first fully elucidated structures of triterpenoid saponins from Acacia sensu stricto having echinocystic acid reported as the aglycone. Compounds 1 and 2 were evaluated for cytotoxic activity against a human melanoma cancer cell line (SK-MEL28) and a diploid fibroblast cell line (HFF), but showed only weak activity.

Full text of DOI:10.1021/acs.jnatprod.7b00437

Article
pubs.acs.org/jnp
Isolation and Structural Characterization of Echinocystic Acid
Triterpenoid Saponins from the Australian Medicinal and Food Plant
Acacia ligulata
†
,‡
†
§
⊥
∥
Diana Jæger, Chi P. Ndi, Christoph Crocoll, Bradley S. Simpson, Bekzod Khakimov,
†
,∇,□
†,∇,□
◆,#
◆,#
Ruth Marian Guzman-Genuino,
John D. Hayball,
,†Xiaohui Xing,
Vincent Bulone,
¶
‡
Philip Weinstein, Birger L. Møller, and Susan J. Semple*
†Sansom Institute for Health Research, School of Pharmacy and Medical Sciences, University of South Australia, Adelaide, South
Australia 5000, Australia
‡
Plant Biochemistry Laboratory, Department of Plant and Environmental Sciences, Faculty of Science, University of Copenhagen,
Thorvaldsensvej 40, DK-1871 Frederiksberg C, Denmark
§
DynaMo Center, Department of Plant and Environmental Sciences, Faculty of Science, University of Copenhagen, Thorvaldsensvej
4
0, DK-1871 Frederiksberg C, Denmark
⊥
Flinders Centre for Innovation in Cancer, Flinders University, Bedford Park, South Australia 5042, Australia
∥
Department of Food Science, Faculty of Science, University of Copenhagen, Rolighedsvej 26, DK-1958 Frederiksberg C, Denmark
∇
Experimental Therapeutics Laboratory, Hanson Institute and Sansom Institute, Adelaide, South Australia 5000, Australia
□
Robinson Research Institute and Adelaide Medical School, University of Adelaide, Adelaide, South Australia 5005, Australia
◆
ARC Centre of Excellence in Plant Cell Walls, School of Agriculture, Food and Wine, The University of Adelaide, Waite Campus,
Urrbrae, 5064, Australia
#
Division of Glycoscience, Royal Institute of Technology (KTH), School of Biotechnology, AlbaNova University Centre, Stockholm,
SE-10691, Sweden
¶
Department of Ecology and Environmental Sciences, School of Biological Sciences, The University of Adelaide, Adelaide, South
Australia 5005, Australia
*
S Supporting Information
ABSTRACT: The Australian plant Acacia ligulata has a
number of traditional food and medicinal uses by Australian
Aboriginal people, although no bioactive compounds have
previously been isolated from this species. Bioassay-guided
fractionation of an ethanolic extract of the mature pods of A.
ligulata led to the isolation of the two new echinocystic acid
triterpenoid saponins, ligulatasides A (1) and B (2), which
differ in the fine structure of their glycan substituents. Their
structures were elucidated on the basis of 1D and 2D NMR,
GC-MS, LC-MS/MS, and saccharide linkage analysis. These are the first isolated compounds from A. ligulata and the first fully
elucidated structures of triterpenoid saponins from Acacia sensu stricto having echinocystic acid reported as the aglycone.
Compounds 1 and 2 were evaluated for cytotoxic activity against a human melanoma cancer cell line (SK-MEL28) and a diploid
fibroblast cell line (HFF), but showed only weak activity.
he genus Acacia sensu stricto (Fabaceae: Mimosoideae) is a
chemical constituents of the pods of Acacia ligulata A. Cunn. ex
T
large genus consisting of 1073 species with 1063 naturally
Benth. were examined. This species is extensively distributed
across mainland Australia. Its leaves and bark have been used
6
occurring in Australia and others native to Southeast Asia, the
1
islands of the Pacific, and the Mascerene Islands. Previously,
in traditional medicine by Aboriginal Australians to treat
7
−9
pods of the Australian species Acacia victoriae Benth. were
found to contain triterpenoid saponins called avicins, which
were found to exhibit potent cytotoxic activity against
coughs, colds, and chest infections and to wash the skin,
whereas the seeds and possibly the pods have been used as a
10−12
food source.
However, A. ligulata has only been used
2
−4
during dry periods in some areas of Australia, and eating the
cancerous cells.
Therefore, it has been suggested that the
distribution and chemistry of saponins in the genus Acacia
should be more fully investigated in order to understand their
structure−activity relationships. In the present study the
Received: May 19, 2017
5
©
XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.7b00437
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Chart 1
1
a
Table 1. H NMR Spectroscopic Data (CD OD, 600 MHz) for Ligulatasides A (1) and B (2)
3
position
1 δH (J in Hz)
2 δH (J in Hz)
position
t-Glc
1 δH (J in Hz)
2 δH (J in Hz)
1
2
3
4
5
6
7
8
9
1.67, 1.02, m
1.96, 1.71, m
1.66, 1.01, m
1.96, 1.71 m
1
4.46, d (8.0)
3.37, m
4.47 d (7.9)
3.37, m
3.14, dd (11.5, 4.0)
3.13, dd (11.6, 4.1)
2
3
3.39, m
3.37, m
0.81, m
0.81, m
4
3.32, m
3.32, m
1.63, 1.43, m
1.61, 1.44, m
1.63, 1.43, m
1.61, 1.44, m
5
3.25, m
3.25, m
6
3.90, m
3.90, m
Ara
1
1.67, m
1.68, m
5.60, d (4.7)
3.86, m
5.60, d (4.4)
3.86, m
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
1.94, m
5.39, m
1.94, m
5.39, m
3
3.90, m
3.90, m
4
3.85, m
3.85, m
5
3.93, 3.56, m
3.92, 3.55, m
Rha
1
1.76, 1.48, m
1.75, 1.48, m
5.16, br s
4.03, m
5.17, br s
4.03, m
4.53, br t (3.0)
4.53, br t (3.0)
2
3
3.86, m
3.86, m
3.07, m
3.06, m
4
3.67, m
3.70, m
2.34, 1.09, m
2.33, 1.08, m
5
3.82, m
3.83, m
6
1.33, d (6.2)
1.35, d (5.9)
1.98, 1.19, m
1.96, 1.80, m
1.01, br s
0.82, br s
1.00, br s
0.83, br s
1.42, s
1.98, 1.18, m
1.95, 1.80, m
1.01, br s
0.81, br s
0.99, br s
0.82, br s
1.41, s
t-Xyl
1
4.51, d (7.6)
3.30, m
4.51, d (7.6)
3.30, m
2
3
3.36, m
3.36, m
4
3.53, m
3.53, m
5
3.90, 3.26, m
3.90, 3.26, m
t-Ara
1
5.36, br s
4.12, m
5.37, br s
4.11, m
0.92, s
0.91, s
2
1.00, br s
0.99, br s
3
3.83, m
4.02, m
GlcNAc
4
4.14, m
3.88, m
1
2
3
4
5
6
4.49, d (8.8)
3.76, m
4.49, d (8.5)
3.76, m
5
4.32, 4.18, m
1.98, br s
2.09, s
3.76, 3.67, m
1.96, br s
CH CONH
3
3.65, m
3.64, m
−OCOCH3
3.63, m
3.63, m
3.42, m
3.42, m
3.90, 3.68, m
3.90, 3.67, m
a
d = doublet, dd = doublet of doublets, br t = broad triplet, br s = broad singlet, s = singlet, m = multiplet. t = terminal. GlcNAc = N-
acetylglucosamine. Glc = glucose. Ara = arabinose. Rha = rhamnose. Xyl = xylose.
seeds of this plant has been observed to “make your hair fall
out” by people of the Yankunytjatjara language group in
Central Australia. This observation of possible toxicity
following consumption of the plant made A. ligulata interesting
for cytotoxic screening and chemical investigations. Preliminary
screening studies of crude extracts from different devel-
1
3
B
DOI: 10.1021/acs.jnatprod.7b00437
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 2. 13C NMR Spectroscopic Data (CD OD, 600 MHz)
for Ligulatasides A (1) and B (2)
opmental stages of seeds and pods from A. ligulata showed a
cytotoxic effect of pod extracts, particularly against melanoma
3
a
14
cancer cells (SK-MEL28). The extract of the mature pods
14
position 1 δ
C
, type 2 δ
C
, type
position
t-Glc
1 δ
, type 2 δ , type
C C
showed the best activity, with an IC50 value at 40.4 μg/mL.
1
2
40.3, CH2
27.4, CH2
40.3, CH2
27.4, CH2
Further, metabolite screening indicated a similar metabolite
profile between the different pod extracts, suggesting the active
component(s) was the same in all pod extracts.
This paper describes the bioassay-guided isolation, structural
elucidation, and cytotoxic analysis of the new triterpenoid
saponin compounds 1 and 2 from the mature pods of A.
ligulata.
1
105.1, CH
105.1,
CH
3
4
5
6
7
91.7, CH 91.8, CH
40.5, C 40.5, C
2
3
4
5
6
78.7, CH 78.7, CH
78.4, CH 78.4, CH
71.9, CH 71.9, CH
75.5, CH 75.5, CH
57.6, CH 57.6, CH
20.1, CH2 20.1, CH2
35.0, CH2 35.0, CH2
62.6, CH2
62.5,
CH2
RESULTS AND DISCUSSION
8
41.3, C
41.3, C
Ara
1
■
The pods of A. ligulata were extracted using 80% (v/v) aqueous
ethanol. The crude extract was subjected to bioassay-guided
fractionation using cytotoxicity in SK-MEL28 cells. The extract
was initially fractionated by normal-phase solid-phase extraction
followed by radial centrifugal TLC and column chromatog-
raphy to yield an active fraction. The two major compounds in
this fraction were purified by RP-HPLC. They were identified
as triterpenoid saponins 1 and 2 and given the trivial names
ligulataside A and ligulataside B, respectively, with their
structures elucidated on the basis of 1D and 2D NMR, GC-
MS, LC-MS/MS, and saccharide linkage analysis studies
9
48.7, CH 48.7, CH
94.8, CH 94.8, CH
75.4, CH 75.7, CH
72.5, CH 72.6, CH
68.2,CH 68.2, CH
10
11
12
13
38.4, C
38.4, C
2
25.0, CH
25.0, CH
3
2
2
124.2, CH 124.1, CH
4
145.2, C
145.2, C
5
65.0, CH
65.1,
CH2
2
1
1
4
5
43.2, C
43.2, C
Rha
1
36.94,
CH2
36.94,
CH2
101.6, CH
101.6,
CH
1
1
1
1
2
6
7
8
9
0
75.1, CH 75.1, CH
50.8, C 50.8, C
2
3
4
5
6
72.8, CH 72.9, CH
81.9, CH 82.1, CH
79.6, CH 79.4, CH
69.5, CH 69.6, CH
42.7, CH 42.7, CH
48.3, CH2 48.3, CH2
(
Figures S1−S27, Supporting Information).
1 13
The H and C NMR data are shown in Tables 1 and 2,
31.8, C
31.8, C
18.8, CH3
18.9,
CH3
respectively, and the spectra are given in the Supporting
Information (Figures S1−S16).
21
22
37.0, CH2 36.9, CH2 t-Xyl
Compound 1 was obtained as a white, amorphous solid. The
molecular formula, C H NO , was deduced from the
32.5, CH2 32.5, CH2
1
106.8, CH
106.8,
CH
6
7
107
31
+
23
29.1, CH3
17.6, CH3
29.1, CH3
17.6, CH3
2
3
4
5
75.7, CH 75.7, CH
78.4, CH 78.3, CH
71. 6, CH 71.6, CH
HRESIMS data (m/z 1422.6950 [M + H] , calcd 1422.6900)
Figure S17, Supporting Information). Compound 1 showed
one olefinic proton in the H NMR spectrum (δ 5.39 ppm)
2
4
(
1
25
16.7, CH3 16.7, CH3
18.5, CH3 18.5, CH3
H
2
6
67.5, CH2
67.5,
CH2
and seven methyl singlets (δ 0.82, 0.83, 0.92, 1.00, 1.00, 1.01,
H
and 1.42 ppm) with one of them (δ 1.42 ppm) showing an
H
2
7
8
27.7, CH3 27.7, CH3 t-Ara
HMBC correlation to the quaternary olefinic carbon at δC
2
177.4, C
177.4, C
1
111.4, CH
111.4,
CH
1
45.2. The 13C NMR spectrum also had two characteristic
signals at δ 23.6 and 173.8 ppm, which indicated the presence
C
29
33.9, CH
25.5, CH
3
33.9, CH
25.5, CH
3
3
2
3
4
5
84.7, CH 84.7, CH
79.6, CH 85.7, CH
82.4, CH 79.0, CH
15
1
of a CH CONH group in the compound. Further, the H
3
30
3
NMR shifts of protons at δ 5.60 (d, J = 4.7 Hz), 5.36 (br s),
H
GlcNAc
1
5
.16 (br s), 4.51, (d, J = 7.6 Hz); 4.49, (d, J = 8.8 Hz); and 4.46
105.4, CH 105.4, CH
65.8, CH2
63.5,
CH2
23.6
ppm (d, J = 8.0 Hz) together with their correlations in the
2
3
4
5
6
57.8, CH 57.8, CH
74.6, CH 74.6, CH
81.6, CH 81.6, CH
76.8, CH 76.8, CH
CH CONH
23.6
173.8
21.2
HSQC spectrum to oxygenated methines at δ 94.8, 111.4,
3
C
CH CONH
173.8
1
01.6, 106.8, 105.4, and 105.1 ppm, respectively, suggested
3
them to be anomeric protons.
−OCOCH3
−OCOCH3
173.0
LC-MS/MS analysis of 1 in the positive-ion mode indicated
a sequenced loss of masses corresponding to saccharide units,
which could be observed in the fragmentation spectra (Figure
S18 A and B, Supporting Information). The presence of three
pentoses (neutral loss of −132), one methyl pentose (−146),
one hexose (−162), an N-acetylglucosamine (−204), and an
acetyl group (−42) was indicated by the fragments observed at
63.0, CH2
63.0, CH2
a
t = terminal. GlcNAc = N-acetylglucosamine. Glc = glucose. Ara =
arabinose. Rha = rhamnose. Xyl = xylose.
and indicated the presence of D-glucose, L-rhamnose, D-xylose,
and L-arabinose (two units) (Figures S23 and S24, Supporting
Information). An N-acetylglucosamine residue could not be
detected by initial GC-MS possibly due to degradation caused
by the strong HCl treatment. In summary, the presence of six
+
+
m/z 1422 [M + H] , 1290 [M − 132 + H] , 1248 [M − 132−
+
+
4
1
1
2 + H] , 1116 [M − 132 − 132 − 42 + H] , 970 [M − 132 −
+
32 − 42 − 146 + H] , 838 [M − 132 − 132 − 42 − 146 −
+
32 + H] , 366 [M − 132 − 132 − 42 − 146 − 132 − 472 +
+
1
H] , and 204 [M − 132 − 132 − 42 − 146 − 132 − 472 − 162
anomeric protons in the H NMR spectrum and the mass
+
+
H] . The presence of a hexose linked to an N-acetylglucos-
spectrometric fragmentation pattern in the LC-MS/MS
suggested a sequential loss of six sugar moieties. This was
also supported by the GC-MS analysis of an acidic hydrolyzed
sample that confirmed the saccharide moieties. Together, this
evidence suggested that compound 1 possesses six saccharide
units.
amine residue was strongly suggested by the fragments of m/z
16,17
366, 221, 204, and 186.
Acid hydrolysis of 1 with HCl
followed by GC-MS suggested the aglycone was echinocystic
acid [(3β,16α)-3,16-dihydroxyolean-12-en-28-oic acid] (3)
(Figures S21 and S22, Supporting Information) (Figure 1)
C
DOI: 10.1021/acs.jnatprod.7b00437
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 1. Hydrolysis of ligulataside A (1). Acid hydrolysis resulted in the aglycone echinocystic acid and free saccharide units analyzed by GC-MS.
Alkaline hydrolysis gave the prosaponin (4), analyzed by LC-MS, indicating the tetrasaccharide chain to be attached at C-28.
Figure 2. Key HMBC and ROESY correlations for 1.
The identity of the aglycone as echinocystic acid was further
which was analyzed by LC-MS (Figure S25, Supporting
Information) with fragments found at m/z 838 corresponding
to echinocystic acid (m/z 472), the N-D-acetylglucosamine-
glucose (m/z 366), suggesting the two saccharides to be linked
at C-3.
confirmed by comparison of the NMR data with previously
18,19
reported echinocystic acid-based triterpenoid saponins.
The NMR data were consistent with 1 being a bisdesmosidic
triterpenoid glycoside with two linkage sites (C-3 and C-28), as
indicated by the chemical shifts of C-3 (δ 91.7) and C-28 (δC
A tetrasaccharide chain was therefore proposed at C-28
through an ester bond, which was supported by the HMBC
data showing a linkage between the anomeric proton of the
arabinose unit of the chain and C-28 of the echinocystic acid
unit (Figure 2). The connectivities of each of the saccharides
and the aglycone were established further using information
C
177.4) of the aglycone. This was also shown by the HMBC
cross-correlation peaks between the doublet at δ 4.49 (H-1 of
H
GlcNAc) and δ 91.7 (C-3 of the aglycone) as well as between
C
the doublet at δ 5.60 (H-1 of Ara) and δ 177.4 (C-28 of the
H
C
aglycone) (Figure S11, Supporting Information). Alkaline
hydrolysis of 1 (Figure 1) produced the prosaponin (4),
1
13
from the H, C, DEPT, COSY, HSQC, HMBC, ROESY, and
D
DOI: 10.1021/acs.jnatprod.7b00437
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
1
1
13
H− H TOCSY experiments. The C NMR spectroscopic data
and 3.76, 3.67 ppm in 2). Further, the chemical shifts of the
suggested one of the arabinose residues to be in the furanose
acetyl group of 1 at δ 21.2, 173.0 and δ 2.09 ppm could not
C H
form, as indicated by the chemical shift of the anomeric carbon
be observed for 2. LC-MS/MS analysis of 2 in the positive-ion
mode gave similar sequential loss of saccharide masses to those
observed for 1. The mass loss of m/z 42 corresponding to the
acetyl unit seen for 1 (Figure S18, Supporting Information) was
also absent (Figure S20, Supporting Information), which
suggested that 2 is a deacetylated derivative of 1. GC-MS and
linkage analysis of the hydrolyzed compound indicated
echinocystic acid as the aglycone and the type of saccharides
and their linkages to be identical to 1 (Figures S22−S27,
Supporting Information). Accordingly, the complete structure
of 2 was determined as 3-O-β-D-glucopyranosyl-(1→4)-2-
acetamido-2-deoxy-β-D-glucopyranosylechinocystic acid 28-O-
α-L-arabinofuranosyl-(1→4)-[β-D-xylopyranosyl-(1→3)]-α-L-
rhamnopyranosyl-(1→2)-α-L-arabinopyranosyl ester (ligulata-
side B).
The cytotoxicity of the two purified ligulatasides was tested
against the SK-MEL28 cells using the sulforhodamine B (SRB)
assay to determine the IC50 values of the pure compounds.
Compounds 1 and 2 gave IC50 values of 25.0 and 15.0 μM,
respectively. These values indicated a lack of significant
cytotoxicity (IC50 > 10 μM). The compounds were further
tested against noncancerous diploid human foreskin fibroblast
3
(
δ 111.37 ppm), while the remaining five monosaccharides
C
20
are in the pyranose form. The coupling constants of each of
the anomeric protons, the glucose, N-acetylglucosamine, and
the xylose were consistent with a beta-configuration (δ 4.46
H
(d, J = 8.0 Hz), 4.49 (d, J = 8.8 Hz), and 4.51 (d, J = 7.6 Hz)),
while the signals arising from the two arabinose units and the
rhamnose residue (δ 5.60 (d, J = 4.7 Hz), 5.36 (br s), and 5.16
H
(
br s), respectively) were consistent with an alpha config-
21,22
uration.
The saccharide sequences and linkage sites were obtained
from TOCSY, HMBC, and ROESY experiments (Figures S11,
S13, S15, Supporting Information). The occurrence of the
esterified saccharide substituent at C-28 was inferred from the
following significant C−H correlations obtained in the HMBC
spectrum: H-1 (δ 5.60) of Ara with C-28 (δ 177.4) of the
H
C
aglycone; H-1 (δ 5.16) of Rha with C-2 (δ 75.4) of Ara; H-1
H
C
(
δ 4.51) of Xyl with C-3 (δ 81.9) of Rha; and H-1 (δ 5.36)
H C H
of the terminal Ara with C-4 (δC 79.6) of Rha. Further
supporting evidence came from the ROESY analysis, as
illustrated in Figure 2.
The sequence of the disaccharide chain at the C-3 position
was indicated by the long-range coupling in the HMBC
spectrum of H-1 (δH 4.46) of Glc with C-4 (δC 81.6) of
(
HFF) cells. It was shown that the activity toward the HFF cells
was similar to the activity against SK-MEL28 cells, with IC50
values of 13.4 and 14.2 μM for 1 and 2, respectively. This
suggested no selectivity between the cancerous and non-
cancerous cell lines tested.
GlcNAc as well as the key correlation between H-1 (δ 4.49) of
H
GlcNAc with C-3 (δC 91.7) of the aglycone, which further
supported the attachment of the disaccharide moiety at C-3 of
1
The most common aglycone for saponins found in Acacia
sensu stricto has been acacic acid, which is structurally related to
echinocystic acid, except for the presence of an additional
glycosylation site. Acacic acid-based saponins are not restricted
to the genus Acacia, but have been found in the closely related
genera Albizia, Entada, and Archidendron from the Mimosoi-
the aglycone. The H NMR spectrum of 1 also showed a signal
for an acetyl methyl group (δH 2.09). HMBC correlations
between H-5 of the furanosyl Ara and the carbonyl carbon (δC
173.0) of this acetyl group indicated that it was attached to the
C-5 oxygen of the Ara saccharide unit.
The types of glycosidic linkages were confirmed by
permethylation analysis and GC-MS. Based on the MS
fragmentation pattern of partially methylated alditol acetates
23
deae subfamily of the Fabales. To the best of our knowledge
echinocystic acid-based saponins have only been described in
24
Acacia sensu stricto from Acacia pulchella, but the complete
structures of any saponins from this species have not been
published. Saponins also consisting of echinocystic acid and
with N-acetylglucosamine attached at C-3 similar to ligulata-
(
Figure S26, Supporting Information), the glycosidic linkages in
the samples were identified to be terminal arabinofuranose (t-
Araf), terminal xylopyranose (t-Xylp), terminal glucopyranose
(t-Glcp), 2-linked arabinopyranose (2-Arap), 3,4-linked rham-
25−28
sides A (1) and B (2) have been found in Albizia.
According to recent molecular studies, the genera Albizia and
nopyranose (3,4-Rhmp), and 4-linked N-acetylglucosaminopyr-
anose (4-GlcNAcp). Due to the experimental conditions during
the methylation (strong alkaline environment caused by the
presence of sodium hydroxide), the acetyl group attached to the
t-Araf residue of compound 1 was cleaved off during the
analysis. The deacetylated compounds 1 and 2 had the same
glycosidic linkage types (Figure S27, Supporting Information).
The complete structure of 1 was assigned as 3-O-β-D-
glucopyranosyl-(1→4)-2-acetamido-2-deoxy-β-D-glucopyrano-
sylechinocystic acid 28-O-(5-O-acetyl)-α-L-arabinofuranosyl-
29
Acacia sensu stricto are closely related. In contrast to acacic
acid-based saponins, echinocystic acid-based triterpenoid
saponins have been found to be more widely distributed
throughout the kingdom Plantae in families such as
30−32
Campanulaceae, Asteraceae, and Sapindaceae
not restricted to the Mimosoideae.
and are
In conclusion, we here report the first compounds isolated
and identified from A. ligulata and the first fully elucidated
echinocystic-type triterpenoid saponins from Acacia sensu
stricto.
(1→4)-[β-D-xylopyranosyl-(1→3)]-α-L-rhamnopyranosyl-(1→
2)-α-L-arabinopyranosyl ester (ligulataside A).
Compound 2 was obtained as a white, amorphous solid. The
EXPERIMENTAL SECTION
molecular formula, C H NO , was deduced from the
■
6
5
105
30
+
General Experimental Procedures. Optical rotations were
recorded on an Anton Paar MCP 100 polarimeter at 25 °C. UV
spectroscopy was conducted using an Agilent Technologies Cary 6
UV−vis spectrophotometer. The 1D and 2D NMR data were acquired
HRESIMS data (m/z 1380.6851 [M + H] , calcd 1380.6794)
Figure S19, Supporting Information), which was 42 mass units
less than 1. The 1D NMR data for 2 indicated this compound
(
1
13
displays identical H NMR and C NMR signals as 1, except
on a Varian INOVA 600 MHz spectrometer in CD OD (Sigma-
3
for the chemical shifts assigned to C-3, C-4, and C-5 of the
Aldrich, St Louis, MO, USA). Chemical shifts in the NMR spectra
were assigned by reference to the signals from the residual solvents.
LC-MS/MS was performed on a Dionex UltiMate 3000 Quaternary
Rapid Separation UHPLC+ focused system (Thermo Fisher Scientific,
terminal Ara in 1 (δ 79.6, 82.4, and 65.8 ppm in 1 and δ 85.7,
C
C
7
3
9.0, and 63.5 ppm in 2) as well as the attached protons (δH
.83, 4.14, and 4.32, 4.18 ppm in 1 compared to δ 4.02, 3.88,
H
E
DOI: 10.1021/acs.jnatprod.7b00437
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Germering, Germany). Separation was achieved on a Kinetex 1.7 μm
XB-C18 column (100 × 2.1 mm, 1.7 μm, 100 Å, Phenomenex). For
larger fractions (ALP1−7), of which each was tested in the SRB assay.
Fractions ALP4 (1130.2 mg), ALP5 (1155.7 mg), and ALP6 (4017.7
mg) all showed activity against the SK-MEL28 cells.
eluting 0.05% (v/v) formic acid in H O and MeCN [supplied with
2
0
.05% (v/v) formic acid] were employed as mobile phases A and B,
Fraction ALP5 was further separated by radial centrifugal thin-layer
chromatography using a Chromatotron centrifugal thin-layer chroma-
tograph (T-squared Technology, San Bruno, CA, USA) with a 1 mm
silica plate (silica gel 60, 2−25 μm, Sigma-Aldrich). Fractions were
eluted from the plate using increasing amounts of MeOH in a
CH Cl /MeOH, 0.1% (v/v) acetic acid system (100 mL per
respectively. Gradient conditions were as follows: 0.0−1.0 min 10% B;
1
8
.0−7.0 min 10−40% B; 7.0−20.0 min 40−80% B, 20.0−26.0 min
0% B, 26.0−27.0 min 80−100%, 27.0−30.0 min 100% B, 30.0−31.0
min 100−10% B, and 31.0−34.0 min 10% B. The flow rate of the
mobile phase was 300 μL/min. The column temperature was
maintained at 30 °C. UV spectra for each sample were acquired at
2
2
concentration increasing from 0 to 25% MeOH in 5% increments,
then 200 mL 30%), with the collection of approximately 5 mL
fractions. The remaining compounds were washed off with 100%
acetone. Based on the TLC analysis of every second fraction, fractions
were pooled into 10 larger fractions (ALP5.1−10). Fraction ALP5.10
(266 mg) showed the most potent activity against SK-MEL28 cells.
The TLC analysis indicated that fraction ALP5.10 consisted of more
than one compound, and a further purification using a short bed
column was performed. A short glass column (⦶ 10 mm, packing
height 200 mm) was wet loaded (in CH Cl ) with Merck silica
2
05, 210, 250, and 310 nm. The UHPLC was coupled to a Compact
micrOTOF-Q mass spectrometer (Bruker, Bremen, Germany)
equipped with an electrospray ion source (ESI) operated in the
positive or negative ionization modes. The ion spray voltage was
maintained at 4500 or 3900 V in the positive and negative ionization
modes, respectively. The dry temperature was set to 200 °C, and the
dry gas flow was set to 8 L/min. Nitrogen was used as the dry gas,
nebulizing gas, and collision gas. The nebulizing gas was set to 2.5 bar
and collision energy to 15 eV. Na-formate clusters were used for
calibration. HRESIMS and MS/MS spectra were acquired in a m/z
range from 50 to 3000 amu at a sampling rate of 3 Hz. All files were
automatically calibrated by postprocessing. The free echinocystic acid
fraction and acidic and alkaline hydrolyzed fractions were analyzed
using GC-MS. The GC-MS consisted of an Agilent 7890A GC and an
Agilent 5975C series MSD (Agilent Technologies, Glostrup, Den-
mark). GC separation was performed on an Agilent HP-5MS column
2
2
(0.040−0.063 mm, Merck, Darmstadt, Germany). The compounds
were eluted with increasing amounts of MeOH in a CH Cl /MeOH,
2
2
0.1% (v/v) acetic acid system (100 mL per concentration increasing
from 0 to 25% MeOH in 5% increments and 200 mL 70%) with
approximately 5 mL fractions being collected. The remaining
compounds were washed off with 100% MeOH. A TLC analysis of
every second fraction from the column showed that at least three
major compounds were eluted from ALP5.10.
(
30 m × 250 μm × 0.25 μm) using hydrogen as carrier gas at a
constant flow rate of 1.2 mL/min. Details of sample preparation and
experimental conditions for the GC-MS are provided in the
Supporting Information.
A final purification by reversed-phase HPLC using a mobile phase of
CH CN/H O (3:7) with 0.1% (v/v) formic acid was performed with
3
2
manual collection of eluting peaks. This yielded 13.8 mg of compound
1 as a white, amorphous solid, 15.6 mg of compound 2 as a white,
amorphous solid, and minor fractions of other impure compounds.
All solvents used for extractions from plant tissues and separations
were analytical grade (Merck, Darmstadt, Germany, and Univar, Ajax
Finechem, Auckland, New Zealand) or HPLC grade (Merck) for
HPLC separations. TLC plates (normal-phase silica gel 60 F254) were
from Merck. Silica gel 60 (70−230 mesh/0.063−0.200 mm, Merck),
silica gel 60 (2−25 μm, Sigma-Aldrich), and silica gel 60 (230−400
mesh/0.040−0.063 mm, Merck) were used for solid-phase extraction
Ligulataside A (1): white, amorphous solid; [α]2 −17.7 (c 0.12,
5
D
1
13
MeOH); UV (MeOH) λmax (log ε) 261 (2.21) nm (sh); H and
C
NMR data (CD OD, 600 MHz), see Table 1 and Table 2; HRESIMS
3
+
m/z 1422.6950 [M + H] (calcd for C H107NO31, 21422.6900).
67
5
Ligulataside B (2): white, amorphous solid; [α] −32.7 (c 0.16,
MeOH); UV (MeOH) λmax (log ε) 279 (2.57) nm; H and C NMR
D
1
13
(
SPE), radial centrifugal thin-layer chromatography (TLC), and
column chromatography, respectively. All HPLC experiments were
carried out on a Shimadzu SIL 20-AHT UFLC system with an
autoinjector using a Phenomenex Gemini 3 μm NX-C18 column (250
mm × 4.40 mm, 3 μm, 110 Å, Phenomenex, Torrance, CA, USA).
Plant Material. Mature pods of Acacia ligulata were collected in
the Southern Flinders Ranges, South Australia, in December 2013.
The pods were collected from more than 20 individual trees. A
voucher specimen was recorded and lodged at the State Herbarium of
South Australia (AD271669), and species identity authenticated by
botanist Mr. Martin O’Leary. Fresh plant material was stored at −20
data (CD OD, 600 MHz), see Table 1 and Table 2; HRESIMS m/z
3
+
1380.6851 [M + H] (calcd for C H NO , 1380.6794).
65
105
30
Acid and Alkaline Hydrolysis. In order to identify the aglycone
and the attached saccharides of the saponins in A. ligulata pod extracts,
HPLC-purified saponin fractions were subjected to acid and alkaline
hydrolysis. Details of these procedures are given in the Supporting
Information.
Methylation-GC-MS Analysis. Pure compounds 1 and 2 (∼0.5
mg) were subjected to glycosidic linkage analysis by permethylation
and GC-MS. The compounds were consecutively methylated in the
presence of iodomethane and sodium hydroxide in dimethyl sulfoxide,
hydrolyzed with trifluoroacetic acid, reduced with sodium borodeu-
teride, and peracetylated using acetic anhydride to produce partially
°C until used.
Extraction and Isolation. A. ligulata mature pods (120.9 g,
separated from seeds) were ground and extracted at room temperature
in 80% (v/v) aqueous EtOH (Univar, Ajax Finechem, Auckland, New
Zealand) using a solvent to plant material ratio of 5:1. After 24 h, the
extract was decanted and filtered through a Whatman No. 1 filter
paper (GE Health Life Sciences, UK). A second equivalent portion of
33
methylated alditol acetates (PMAAs), as described previously.
PMAAs from neutral saccharides and amino saccharides were analyzed
on an Agilent 6890/5973 GC-MS system (Agilent Technologies, Santa
Clara, CA, USA) fitted with an SP-2380 capillary column (30 m × 0.25
mm i.d., Sigma-Aldrich) and an Agilent CP-Sil 5 CB capillary column
8
0% (v/v) EtOH was added to the plant material and allowed to
3
4
extract for an additional 24 h before being decanted and filtered. The
filtered extract was concentrated using a rotary evaporator (Buchi,
Switzerland) with a water bath temperature of 40 °C, then freeze-dried
(30 m × 0.25 mm i.d.; Agilent Technologies), respectively.
Experiments were conducted in duplicate.
Cytotoxicity Assays. The SK-MEL28 human melanoma cancer
cell line (ATCC HTB-72) (ATCC, Manassas, VA, USA) was cultured
in RPMI-1640 GlutaMAX medium (Gibco, ThermoFisher, Australia)
supplemented with 10% (v/v) fetal bovine serum (FBS) (Thermo
Scientific, MA, USA). The diploid human fibroblast cell line HFF
(ATCC SCRC-1041) was cultured in DMEM (SA Pathology,
Adelaide, Australia) supplemented with 2 mM glutamine and 10%
(v/v) FBS (Thermo Scientific). All cells were cultured in humidity at
37 °C with 5% CO2.
(
Christ Alpha 2-4LD). The dried extract was stored at −20 °C until
further fractionation.
An initial separation of 10 g of the crude extract was performed by
normal-phase solid-phase extraction (NP-SPE) using a 150 mL
cartridge (⦶ 3.8 cm, length 15 cm, Grace Davison Discovery Sciences,
Deerfield, IL, USA) and silica gel 60 (0.063−0.200 mm, Merck). The
column was attached to a vacuum manifold for each separation step.
The compounds were eluted with increasing amounts of MeOH in an
EtOAc/MeOH system (90 mL 0%, 90 mL 20%, 240 mL 40%, 90 mL
The cytotoxic activity of fractions and compounds was measured
using a sulforhodamine B-based cell viability assay kit (TOX6, Sigma-
Aldrich). Cells were distributed in 96-well plates (Sarstedt, Australia)
6
0%, 90 mL 80%, and 150 mL 100%) with collection of 10 mL
fractions. Based on TLC analysis the eluted fractions were pooled into
F
DOI: 10.1021/acs.jnatprod.7b00437
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
5
at a density of 2 × 10 cells/mL and preincubated for 24 h. The
(9) Lassak, E. V.; McCarthy, T. Australian Medicinal Plants, 2nd ed.;
Reed New Holland: Chatswood, Australia, 2011; p 55.
(10) Latz, P. K. Bushfires & Bushtucker; IAD Press: Alice Springs,
Australia, 1995.
following day, cells were treated with fractions or compounds in
various concentrations and incubated for 48 h. Controls for cell
growth (without treatment) in the maximum DMSO concentration
(
0.4% for all assays), fraction absorbance controls of all concentrations,
(11) O’Connell, J. F.; Latz, P. K.; Barnett, P. Econ. Bot. 1983, 37, 80−
and medium controls were also included. A positive control of
staurosporine as a known growth inhibitor was also included. After 48
h, the cells were fixed and stained following the manufacturer’s
protocol. Absorbance was measured using an Ascent Multiskan plate
reader (Labsystems, Helsinki, Finland) at 540 nm and at 690 nm
109.
(12) Maslin, B.; Thomson, L.; McDonald, M.; Hamilton-Brown, S.
Edible Wattle Seeds of Southern Australia: A Review of Species for Use in
Semi-arid Regions; CSIRO: Melbourne, 1998.
(13) Goddard, C.; Everard, P.; Kalotas, A. Punu: Yankunytjatjara
Plant Use; IAD Press: Alice Springs, Australia, 1985.
(
background). The concentration of each fraction or pure compound
that inhibited cell viability by 50% compared to the control (IC ) was
̈
(14) Jæger, D.; Simpson, B. S.; Ndi, C. P.; Jager, A. K.; Crocoll, C.;
50
determined by dose−response curve fitting using the GraphPad Prism
software (version 6 for Mac OS X, GraphPad, San Diego, CA, USA).
Møller, B. L.; Weinstein, P.; Semple, S. J. Nat. Prod. Res. 2017, 1.
(15) Seo, Y.; Hoch, J.; Abdel-Kader, M.; Malone, S.; Derveld, I.;
Adams, H.; Werkhoven, M.; Wisse, J. H.; Mamber, S. W.; Dalton, J. M.
J. Nat. Prod. 2002, 65, 170−174.
ASSOCIATED CONTENT
■
(
16) Domon, B.; Costello, C. E. Glycoconjugate J. 1988, 5, 397−409.
*
S
Supporting Information
(17) Cui, M.; Song, F.; Zhou, Y.; Liu, Z.; Liu, S. Rapid Commun. Mass
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.7b00437.
Spectrom. 2000, 14, 1280−1286.
18) Koz, O.; Bedir, E.; Masullo, M.; Alankus-Caliskan, O.; Piacente,
S. Phytochemistry 2010, 71, 663−668.
19) Pertuit, D.; Larshini, M.; Brahim, M. A.; Markouk, M.; Mitaine-
(
(
Experimental details, GC-MS analysis, the acid and
alkaline hydrolysis of saccharides, NMR spectra, mass
spectra, GC-MS traces, and MS fragmentation pattern of
PMAAs (PDF)
Offer, A. C.; Paululat, T.; Delemasure, S.; Dutartre, P.; Lacaille-Dubois,
M. A. Phytochemistry 2017, 139, 81−87.
(
(
(
20) Gorin, P. A.; Mazurek, M. Can. J. Chem. 1975, 53, 1212−1223.
21) Agrawal, P. K. Phytochemistry 1992, 31, 3307−3330.
22) Bush, C. A. Bull. Magn. Reson. 1988, 10, 73−78.
AUTHOR INFORMATION
(23) Lacaille-Dubois, M. A.; Pegnyemb, D. E.; Note,
Offer, A. C. Phytochem. Rev. 2011, 10, 565−584.
24) Alexander, R.; Croft, K. D.; Kagi, R. I.; Shea, S. Aust. J. Chem.
978, 31, 2741−2744.
25) Orsini, F.; Pelizzoni, F.; Verotta, L. Phytochemistry 1991, 30,
4111−4115.
26) Carpani, G.; Orsini, F.; Sisti, M.; Verotta, L. Phytochemistry
989, 28, 863−866.
27) Melek, F.; Miyase, T.; Ghaly, N.; Nabil, M. Phytochemistry 2007,
8, 1261−1266.
28) Miyase, T.; Melek, F.; Ghaly, N.; Warashina, T.; El-Kady, M.;
Nabil, M. Phytochemistry 2010, 71, 1375−1380.
29) Kyalangalilwa, B.; Boatwright, J. S.; Daru, B. H.; Maurin, O.;
Bank, M. Bot. J. Linn. Soc. 2013, 172, 500−523.
30) Lee, K. T.; Choi, J.; Jung, W. T.; Nam, J. H.; Jung, H. J.; Park, H.
J. J. Agric. Food Chem. 2002, 50, 4190−4193.
31) Lunga, P. K.; Qin, X.-J.; Yang, X. W.; Kuiate, J.-R.; Du, Z. Z.;
́
O. P.; Mitaine-
■
Corresponding Author
(
1
(
*
Tel: +61 8 830 22395. Fax: +61 8 830 21087. E-mail: Susan.
Semple@unisa.edu.au.
ORCID
(
1
(
6
Susan J. Semple: 0000-0001-5988-3993
Notes
The authors declare no competing financial interest.
(
ACKNOWLEDGMENTS
■
(
This work was partially funded through a Sansom Institute for
Health Research, University of South Australia, Collaborative
Grant and University of South Australia Scholarship to D.J. B.
Dueholm is thanked for his assistance with the collection of
plant material. P. Clements (University of Adelaide) is thanked
for running the NMR experiments. H. C. Lam and A. Day
(
(
Gatsing, D. BMC Complementary Altern. Med. 2014, 14, 369.
(32) Shao, Y.; Poobrasert, O.; Ho, C.-T.; Chin, C.-K.; Cordell, G. A.
Phytochemistry 1996, 43, 195−200.
(
University of Adelaide) are thanked for assistance in
measuring the specific rotation.
(33) Pettolino, F. A.; Walsh, C.; Fincher, G. B.; Bacic, A. Nat. Protoc.
2
012, 7, 1590−1607.
34) Melida, H.; Sandoval-Sierra, J. V.; Dieg
Bulone, V. Eukaryotic Cell 2013, 12, 194−203.
(
́
́
uez-Uribeondo, J.;
REFERENCES
■
(
(
1) Maslin, B. Gard. Bull. Singapore 2015, 67, 231−250.
2) Haridas, V.; Higuchi, M.; Jayatilake, G. S.; Bailey, D.; Mujoo, K.;
Blake, M. E.; Arntzen, C. J.; Gutterman, J. U. Proc. Natl. Acad. Sci. U. S.
A. 2001, 98, 5821−5826.
(
3) Jayatilake, G. S.; Freeberg, D. R.; Liu, Z.; Richheimer, S. L.; Blake,
M. E.; Bailey, D. T.; Haridas, V.; Gutterman, J. U. J. Nat. Prod. 2003,
6, 779−783.
4) Mujoo, K.; Haridas, V.; Hoffmann, J. J.; Wac
6
(
̈
hter, G. A.; Hutter,
L. K.; Lu, Y.; Blake, M. E.; Jayatilake, G. S.; Bailey, D.; Mills, G. B.
Cancer Res. 2001, 61, 5486−5490.
(
5) Seigler, D. S. Conservation Sci. Western Australia 2002, 4, 109−
1
16.
(
(
(
6) Chapman, A.; Maslin, B. Nuytsia 1992, 8, 249−283.
7) Webb, L. J. Mankind 1969, 7, 137−146.
8) Barr, A.; Chapman, J.; Smith, N.; Wightman, G.; Knight, T.;
Mills, L.; Andrews, M.; Alexander, V. Traditional Aboriginal Medicines
in the Northern Territory of Australia by Aboriginal Communities of the
Northern Territory; Conservation Commission of the Northern
Territory of Australia: Darwin, Australia, 1993.
G
DOI: 10.1021/acs.jnatprod.7b00437
J. Nat. Prod. XXXX, XXX, XXX−XXX