Secoiridoid Glucosides from the Twigs of Syringa oblata var. dilatata and Their Neuroprotective and Cytotoxic Activities

Article abstract of DOI:10.1248/cpb.c16-00804

Phytochemical investigation of the twigs of Syringa oblata var. diatata led to the isolation of two new secoiridoid glucosides, dilatioside A-B (1-2), along with thirteen known ones (3-15). The structures were determined by spectroscopic methods including

Full text of DOI:10.1248/cpb.c16-00804

Vol. 65, No. 4
Chem. Pharm. Bull. 65, 359–364 (2017)
359
Regular Article
Secoiridoid Glucosides from the Twigs of Syringa oblata var. dilatata and
Their Neuroprotective and Cytotoxic Activities
Kyoung Jin Park,a Won Se Suh,a Lalita Subedi,b Sun Yeou Kim,b Sang Un Choi,c and
,a
Kang Ro Lee*
a Natural Products Laboratory, School of Pharmacy, Sungkyunkwan University; Suwon 16419, Republic of Korea:
b Laboratory of Pharmacognosy, College of Pharmacy, Gachon University; Incheon 21936, Republic of Korea: and
c Korea Research Institute of Chemical Technology; Daejeon 34114, Republic of Korea.
Received October 12, 2016; accepted January 25, 2017
Phytochemical investigation of the twigs of Syringa oblata var. diatata led to the isolation of two new
secoiridoid glucosides, dilatioside A–B (1–2), along with thirteen known ones (3–15). The structures were
determined by spectroscopic methods including one and two dimensional (1- and 2D-) NMR techniques, high
resolution (HR)-FAB-MS, and chemical methods. The isolated compounds (1–15) were tested for the induc-
tion of nerve growth factor (NGF) secretion in a C6 rat glioma cell line and their cytotoxicity against four
human cancer cell lines (A549, SK-OV-3, SK-MEL-2, HCT15) in vitro using a sulforhodamine B bioassay.
Compounds 5, 7, 8, 10, and 14 were found to induce upregulation of NGF secretion without causing signifi-
cant cell toxicity.
Key words Syringa oblata var. dilatata; Oleaceae; secoiridoid glucoside; nerve growth factor; cytotoxicity
Current research is focused on finding neurotrophin structures of the isolated compounds were determined by their
signaling-mediated neuroprotection against neurodegenera- NMR spectroscopic data, MS, and methanolysis. The isolated
tive diseases.1) Phytochemicals that trigger the production of compounds were tested for their ability to induce NGF secre-
neurotrophins such as nerve growth factor (NGF) will protect tion and for their cytotoxic activities. This paper describes the
neurons against neuronal degeneration. NGF mainly acts in isolation and structural elucidation of the two new compounds
the growth, development, and survival of neurons, and has (1–2) as well as their neuroprotective and antiproliferative
been suggested to play an important role in neurodegenerative activities.
diseases. NGF not only helps the survival and growth but also
the differentiation of sensory and sympathetic neurons in the
central and peripheral nervous systems. NGF signaling serves
Results and Discussion
The methanol extract of the twigs of S. oblata var. dilatata
neuroprotective and repair functions.2) Higher NGF production was partitioned successively with n-hexane, CHCl3, EtOAc,
ensures the protection of axons and myelin sheathes against and n-BuOH. Repeated column chromatographic purification
inflammation via modulation of the immune system and re- of the EtOAc-soluble fraction afforded two new secoiridoid
duction of endotoxin- or inflammation-induced toxicity in the glucosides (1–2), together with thirteen known secoiridoid
brain.3) Moreover, deficiency in NGF results in neurodegen- glucoside derivatives (3–15) (Fig. 1).
erative disorders including Alzheimer’s disease, Parkinson’s
Compound 1 was obtained as an amorphous gum. The
disease, and diabetic polyneuropathy.4,5) Thus, regulation of molecular formula of 1 was determined to be C37H48O16 on
NGF secretion or treatment with NGF mimetics is one method the basis of a [M+Na]+ peak at m/z 771.2844 (Calcd for
for the prevention and repair of neurodegenerative disorders. C37H48NaO16, 771.2840) in positive high-resolution (HR)-
The immune modulatory and strong neuroprotective efficacy FAB-MS. The 1H-NMR spectrum displayed the typical
of NGF has resulted in it becoming a potential target for the oleoside methyl ester moiety11); two olefinic protons [δH 7.56
screening of phytochemicals for neurodegenerative diseases.
(1H, s, H-3), 6.13 (1H, m, H-8)], one hemiacetalic proton [δH
Syringa genus has long been used for the treatment of 5.99 (1H, brs, H-1)], one methine [δH 4.03 (1H, dd, J=9.0,
asthma, inflammation, and liver and intestinal disorders.6–9) 4.6Hz, H-5)], one methylene [δH 2.53 (1H, dd, J=14.0, 9.0Hz,
Previous phytochemical investigations regarding this genus H-6a), 2.75 (1H, dd, J=14.0, 4.6Hz, H-6b), one methyl [δH
have reported the isolation of secoiridoid glucosides, lignans, 1.70 (3H, dd, J=7.1, 1.4Hz, H-10)], one methyl ester group
and phenolic compounds with anti-radical scavenging and cy- [δH 3.71 (3H, s, 11-OCH3)], and one glucopyranosyl unit [δH
totoxic activity.10) However, only a few phytochemical studies 4.83 (1H, d, J=7.8Hz, H-1′), 3.32 (1H, m, H-2′), 3.42 (1H, t,
on S. oblata var. dilatata have been reported. We found that J=9,1Hz, H-3′), 3.27 (1H, m, H-4′), 3.33 (1H, overlap, H-5′),
the MeOH extract of the twig of S. oblata var. dilatata in- 3.62 (1H, dd, J=12.0, 6.4Hz, H-6′a), 3.88 (1H, dd, J=11.9,
1
duces increases in the levels of endogenous NGF in C6 glioma 2.1Hz, H-6′b) ]. Furthermore, the H-NMR data exhibited the
cells. Thus, we investigated the bioactive constituents of the presence of a (−)-secoisolariciresinol12); six aromatic protons
aerial parts of S. oblata var. dilatata. The column chromato- [δH 6.69 (2H, d, J=8.0Hz, H-5‴, 5⁗), 6.61 (1H, d, J=1.8Hz,
graphic purification of the EtOAc-soluble fraction led to the H-2⁗), 6.58 (1H, dd, J=8.0, 1.8Hz, H-6‴), 6.55 (1H, dd,
isolation of two new secoiridoid glucosides, dilatioside A–B J=8.0, 1.8Hz, H-6⁗), and 6.53 (1H, d, J=1.8Hz, H-2‴)],
(1–2), along with thirteen known ones (3–15). The chemical two oxygenated methylenes [δH 4.29 (1H, dd, J=11.1, 5.8Hz,
*To whom correspondence should be addressed. e-mail: krlee@skku.edu
© 2017 The Pharmaceutical Society of Japan

360
Chem. Pharm. Bull.
Vol. 65, No. 4 (2017)
Fig. 1. Chemical Structures of Compounds 1 and 2
H-4″b), 3.93 (1H, dd, J=11.1, 6.3Hz, H-4″a), 3.70 (1H, dd, 6.3Hz, H-8″)], one oxygenated methylene [δH 4.78 (1H, m,
J=10.9, 6.7Hz, H-1″b), 3.53 (1H, m, H-1″a)], two methylenes H-9″a, 4.68 (1H, m, H-9″b)], two methoxy groups [δH 3.90
[δH 2.71 (1H, dd, J=13.9, 6.5Hz, H-7‴b), 2.57 (1H, dd, J=13.9, (6H, s, 3′,5′-OCH3)]; δC 154.57 (C-3″,5″), 136.48 (C-4″), 135.08
8.7Hz, H-7‴a), 2.62 (2H, m, H-7⁗)], two methines [δH 2.13 (C-7″), 134.64 (C-1″), 124.51 (C-8″), 105.88 (C-2″,6″) and 66.39
(1H, m, H-3″), 1.94 (1H, m, H-2″) and two aromatic methoxy (C-9″), and a glucopyranose unit [δH 4.92 (1H, d, J=6.5Hz,
groups [δH 3.76, 3.75 (3H each, s, 3‴,3⁗-OCH3)] (Table 1). The H-1‴), 3.50 (1H, m, H-2‴), 3.45 (1H, m, H-3‴), 3.44 (1H, m,
13C-NMR spectrum showed signals for an oleoside methyl H-4‴), 3.25 (1H, m, H-5‴), 3.69 (1H, m, H-6‴a), 3.81 (1H, m,
ester moiety and (−)-secoisolariciresinol12) (Table 1). The H-6‴b)], together with the 13C-NMR data [δC 105.4 (C-1‴),
connectivity between C-7 and C-4″ (oleoside dimethyl ester 75.9 (C-2‴), 78.0 (C-3‴), 71.5 (C-4‴), 78.6 (C-5‴), 62.8 (C-6‴)]
(1a) and (−)-secoisolariciresinol (1b)) was confirmed through (Table 1). The connectivity between C-7 and C-9″ (oleoside
heteronuclear multiple bond connectivity (HMBC) correlation dimethyl ester (1a) and syringin (2a)) was confirmed through
(Fig. 2). The HMBC correlation of H-1′ to C-1 indicated that HMBC correlation (Fig. 2). The HMBC correlation from H-1‴
the glucopyranose unit was linked to the oxygen at C-1, and to C-4″ showed that the D-glucopyranose unit was located
the J value of the anomeric proton (J=7.8Hz) confirmed it as at C-4″, and the J value of the anomeric proton (J=6.5Hz)
the β-configuration.13) This gross structure was confirmed by confirmed it as β-D-glucopyranose.13) This gross structure
1
1
1
analysis of the H–1H correlation spectroscopy (COSY), H- was confirmed by analysis of the H–1H COSY, HMQC, and
detected heteronuclear multiple quantum coherence (HMQC), HMBC spectra (Fig. 2). The configuration of the oleoside
and HMBC spectra (Fig. 2). The configuration of the oleoside dimethyl ester moiety (1a) was assumed by comparison of the
dimethyl ester moiety (1a) was identified by comparison of the NMR and physical data with previously isolated oleoside di-
NMR and physical data with oleoside dimethyl ester (3) previ- methyl ester (3).11) The stereochemistry in 2 was reconfirmed
ously isolated from Fraxinus excelsior.11) (−)-Secoisolarici- through NOESY correlations and biosynthetic aspect14) (Fig.
resinol moiety (1b) was confirmed by comparison the optical 3). Alkaline methanolysis of 2 afforded oleoside dimethyl
rotation and H-NMR data with previously reported values.12) ester (1a) and syringin (2a). 2a was identified by comparison
1
The stereochemistry in 1 was reconfirmed through nuclear of the co-TLC; (CHCl3–MeOH–H2O=3:1:0.1, Rf=0.38) and
Overhauser effect spectroscopy (NOESY) correlations and 1H-NMR data with the reported values, respectively.16) Thus,
biosynthetic aspect14) (Fig. 3). Alkaline methanolysis of 1 af- the structure of 2 was determined as shown in Fig. 1, and
forded oleoside dimethyl ester (1a) and (−)-secoisolariciresinol named dilatioside B. The thirteen known secoiridoid glucoside
(1b).12) 1a was identified by comparison of their 1H-NMR derivatives were identified as oleoside dimethyl ester (3),11) li-
data and 1b was identified as (2″R,3″R)-secoisolarisiresinol by gustroside (4),11) oleuropein (5),17) (2″R)-2″-methoxyoleuropein
1
comparison with their H-NMR data and negative specific ro- (6),18) fraxamoside (7),19) hydroxyframoside A (8),20) syrin-
tation {[α]D25 −20.0 (c=0.05, MeOH)} with the reported values, galactone A (9),21) syringalactone B (10),21) (8E)-nüzhenide
respectively.12) And the configurations at C-2″ and C-3″ in 1b (11),22) (8Z)-nuezhenide
A
(12),23) jaspolyanoside (13),13)
were confirmed by comparison of negative Cotton effects at jaspolyoside (14),24) oleonuezhenide (15)25) by comparison
228 and 287nm in the circular dichroism (CD) spectrum.12) of their spectroscopic data with the data reported in the lit-
Acid hydrolysis of 3(1a) afforded D-glucopyranose, which was eratures.
identified by co-TLC with authentic samlple (CHCl3–MeOH–
H2O=2:1:0.1, Rf=0.3) and specific optical rotation {[α]D25 evaluated by determining their effects on NGF secretion in C6
+105.0 (c=0.04, MeOH)}.15)
cells (Table 2). Of the tested compounds at 50µM, compounds
Thus, the structure of 1 was established as shown in Fig. 1, 5, 7, 8, 10, and 14 were potent stimulants of NGF release, with
and this compound was named dilatioside A. stimulation levels of 201.58 4.41, 207.48 15.41, 205.64 4.84,
The neuroprotective activities of the isolates (1–15) were
Compound 2 was isolated as an amorphous gum with the 196.85 4.71, and 171.64 1.61%, respectively (the positive
molecular formula C34H46O19 based on the positive HR-FAB- control 6-shogaol was 168.58 7.16%), while compounds 6,
MS data (m/z 781.2534 [M+Na]+, Calcd for C34H46NaO19, 12, and 13 exhibited moderate activities (Table 2). Interest-
1
781.2531). The H- and 13C-NMR spectra displayed the typical ingly, structural differences in the tested compounds displayed
1
oleoside methyl ester11) moiety. Moreover, the H-NMR data different NGF secretion stimulatory levels; that is, although
exhibited the presence of a syringin moiety16); aromatic pro- the structures of 4, 9, and 13 are quite similar to those of
tons [δH 6.80 (2H, s, H-2″, 6″)], one trans-substituted double 5, 10, and 14, with the exception of the presence of the hy-
bond [δH 6.64 (1H, d, J=15.7Hz, H-7″), 6.30 (1H, dt, J=15.9, droxy group at C-3 in the aromatic ring, their effects on NGF

Vol. 65, No. 4 (2017)
Chem. Pharm. Bull.
361
Table 1. 1H- (700MHz) and 13C- (175MHz) NMR Spectral Data of 1 and 2 in CD3OD (δ in ppm)a)
1
2
Position
δH
δC
δH
δC
1
3
4
5
6
5.99 brs
7.56 s
95.1
155.3
109.5
32.2
5.97 s
7.56 s
95.6
155.3
109.5
32.1
4.03 dd (9.0, 4.6)
2.53 dd (14.0, 9.0)
2.75 dd (14.0, 4.6)
4.05 m
2.58 m
41.5
41.3
2.79 d (13.6)
7
173.4
124.9
130.8
13.8
173.1
125.1
130.7
13.8
8
6.13 m
6.13 m
9
10
1.70 dd (7.1, 1.4)
1.75 brs
11
168.8
52.1
168.8
52.1
11-OCH3
3.71 s
4.83 d (7.8)
3.32 m
3.74 s
4.78 brs
3.31 m
3.43 m
3.31 m
3.32 m
3.66 m
3.89 m
1′
2′
3′
4′
5′
6′
100.8
74.9
101.2
74.9
3.42 t (9.1)
78.6
78.1
3.27 m
71.7
71.6
3.33 overlap
3.62 dd (12.0, 6.4)
3.88 dd (11.9, 2.1)
3.53 m
78.1
78.5
62.9
62.9
1″
63.0
134.64
3.70 dd (10.9, 6.7)
1.94 m
2″
3″
4″
44.4
40.5
66.4
6.80 s
105.88
154.57
136.48
2.13 m
3.93 dd (11.1, 6.3)
4.29 dd (11.1, 5.8)
5″
6″
7″
8″
9″
154.57
105.88
135.08
124.51
66.39
6.80 s
6.64 d (15.7)
6.30 dt (15.9, 6.3)
4.78 m
4.68 m
1‴
2‴
3‴
4‴
5‴
6‴
133.8
113.4
148.9
145.7
116.0
122.9
4.92 d (6.5)
3.50 m
105.4
75.9
78.0
71.5
78.6
62.8
6.53 d (1.8)
3.45 m
3.44 m
6.69 d (8.0)
3.25 m
6.58 dd (8.0, 1.8)
3.69 m
3.81 m
7‴
2.57 dd (13.9, 8.7)
2.71 dd (13.9, 6.5)
35.9
1⁗
2⁗
3⁗
4⁗
5⁗
133.3
113.6
149.0
145.8
116.0
123.0
36.1
6.61 d (1.8)
6.69 d (8.0)
6.55 dd (8.0, 1.8)
2.62 m
6⁗
7⁗
3′-OCH3
5′-OCH3
3‴-OCH3
3⁗-OCH3
3.90 s
3.90 s
57.3
57.3
3.76 s
3.75 s
56.4
56.4
a) Assignments were based on 2D-NMR including HMQC and HMBC. Well-resolved couplings are expressed with coupling patterns and coupling constants in Hz in pa-
rentheses.
synthesis induction is so different (Fig. S1, Table 2). These evaluated by the determination of their inhibitory effects on
data suggest that the presence of a 2-(3,4-dihydroxylphenyl)- four tumor cell lines including A549 (lung adenocarcinoma),
ethoxycarbonyl moiety may be important for NGF induction.
SK-OV-3 (ovarian cancer cells), SK-MEL-2 (skin melanoma),
The antiproliferative activities of compounds 1–15 were and HCT15 (colon cancer cells) using the SRB bioassay.26)

362
Chem. Pharm. Bull.
Vol. 65, No. 4 (2017)
1
Fig. 2. Key HMBC and H–1H COSY Correlations of 1 and 2
Fig. 3. Key NOESY Correlations of 1 and 2
Table 2. Effects of Compounds 1–15 on NGF Secretion and Cell Vi-
were measured on a JASCO P-1020 polarimeter. IR spectra
were recorded on a JASCO FT/IR-4600 spectrometer, and
NMR spectra were recorded on a Bruker AVANCEШ 700
NMR spectrometer operating at 700MHz (1H) and 175MHz
(13C) with chemical shifts given in ppm (δ). FAB and HR-
FAB mass spectra were obtained on a JEOL JMS700 mass
spectrometer, and preparative HPLC was performed using a
Gilson 306 pump with a Shodex refractive index detector and
a Phenomenex Luna 10µm column (250×10mm). Silica gel 60
(Merck, Darmstadt, 70–230, and 230–400 mesh) and RP-C18
silica gel (Merck, 230–400 mesh) were used for column chro-
matography. Ion exchange resin (Dowex® 50WX8 hydrogen
form, Sigma-Aldrich, U.S.A.) was used for alkali elimination.
TLC was performed using Merck pre-coated silica gel F254
plates and RP-18F254s plates. Spots were detected under UV
light or by heating after spraying with 10% H2SO4 in EtOH
(v/v). Low-pressure liquid chromatography was performed
over Merck LiChroprep Lobar-A Si gel 60 (240×10mm) with
an FMI QSY-0 pump (ISCO).
ability in C6 Cellsa)
Compounds
NGF secretion (%)
Cell viabilityb) (%)
1
97.36 0.77
103.24 2.70
112.47 7.27
97.28 0.81
90.20 4.78
95.28 0.81
102.44 4.40
101.56 1.38
103.43 1.52
108.62 3.31
97.11 0.04
99.08 1.31
108.59 2.10
111.70 1.42
107.22 0.56
106.27 1.15
125.80 0.93
2
93.20 15.02
3
69.32 16.90
4
94.79 10.36
5
201.58 4.41***
138.16 4.80*
207.48 15.41***
205.64 4.84***
92.72 10.72
6
7
8
9
10
196.85 4.71***
78.47 10.27
11
12
139.33 10.58*
114.40 2.62
13
14
171.64 1.61***
72.39 9.48
15
6-Shogaolc)
168.58 7.16***
Plant Material The twigs of S. oblata var. dilatata were
collected at Suwon, Korea in June 2014. The plant was identi-
fied by one of the authors (K. R. Lee). A voucher specimen
(SKKU-NPL 1404) was deposited in the herbarium of the
School of Pharmacy, Sungkyunkwan University, Suwon, Re-
public of Korea.
a) C6 cells were treated with 50µg/mL compounds 1–15. After 24h, NGF secre-
tion in C6-conditioned media was measured by ELISA and is expressed as a percent-
age of the untreated control. The data shown represent the mean standard deviation
(S.D.) of three independent experiments performed in triplicate. b) Cell viability
after treatment with 50µg/mL each extract was determined by an MTT assay and
is expressed as a percentage of the untreated control (%). The results are the aver-
age of three independent experiments, and the data are expressed as the mean S.D.
c) 6-Shogaol as a positive control. *p<0.05, and ***p<0.001 indicate statistically
significant differences in comparison to untreated control group.
Extraction and Isolation The air-dried and pulverized
twigs of S. oblata var. dilatata (6.9kg) were extracted with
Compounds 5 and 6 exhibited antiproliferative activity against 80% MeOH three times at room temperature. The resultant
the SK-MEL-2 cells, with IC50 values of 10.86 and 14.64µM, MeOH extract (450g) was suspended in distilled water and
respectively, and compound 10 showed activity against the successively partitioned with n-hexane, CHCl3, EtOAc and
SK-OV-3 and SK-MEL-2 cells, with IC50 values of 16.83 and n-BuOH, yielding 15, 25, 48 and 213g, respectively. The
10.45 µM, respectively, but most of the compounds were inac- EtOAc soluble fraction (48.0g) was separated over a silica
tive (IC50>30.0µM) against the human tumor cell lines tested.
gel column (230–400 mesh, 350g) eluted with CHCl3–MeOH
[20:1 (1.5L), 15:1 (1.5L), 10:1 (1.5L), 6:1 (1.0L), 3:1 (1.0L)
and 1:1 (1.5L)] to afford ten fractions [Fr. A, 20:1, 1.0L; Fr.
Experimental
General Experimental Procedures Optical rotations B, 20:1, 0.5L; Fr. C, 15:1, 1.0L; Fr. D, 15:1, 0.5L; Fr. E,

Vol. 65, No. 4 (2017)
Chem. Pharm. Bull.
363
1
10:1, 0.5L; Fr. F, 10:1, 0.5L; Fr. G, 10:1, 0.5L; Fr. H, 6:1, son of H-NMR. The reaction mixtures of 2 were evaporated
1.0L; Fr. I, 3:1, 1.0L; Fr. J, 1:1, 1.5L]. Fraction F (1.5g) was under reduced pressure to yield a mixture of oleoside dimethyl
1
separated over an RP-C18 silica gel column (230–400 mesh, ester and syringin (0.6mg), identified by H-NMR and co-TLC
80g, 50% MeOH) to give seven subfractions [Fr. F1–F7 (each as dimethyl ester (1a) and syringin (2a) (CHCl3–MeOH–
1.0L)]. Fraction F2 (274mg) was purified by semi-preparative H2O=3:1:0.1, Rf=0.38).13)
reversed-phase HPLC (flow rate; 2mL/min, 40% MeCN)
Acid Hydrolysis of Compound 3 (=1a) and Sugar Analy-
to yield 3 (7mg, tR=21.2min). Fraction F4 (567mg) was sis Compound 3 (=1a) (1.1mg) was refluxed with 1mL of
separated by semi-preparative reversed-phase HPLC (flow 1N HCl (aq.) at 90°C for 3h. The hydrolysate was extracted
rate; 2mL/min, 30% MeCN), as described above, to yield 4, with EtOAc, and the aqueous layer was neutralized using an
(130mg, tR=22.0min). Fraction G (7.6g) was chromatographed Amberlite IRA-67 column to yield the sugar. The H2O layer
over an RP-C18 silica gel column (230–400 mesh, 400g, 50% yielded D-glucose which was identified with authentic sample
MeOH) to give nine subfractions [Fr. G1–G9 (each 1.0L)]. (Sigma-Aldrich) using silica gel co-TLC (CHCl3–MeOH–
Fraction G3 (3.9g) was separated over a silica gel column H2O=2:1:0.1, Rf=0.3).27)
(230–400 mesh, 20g, CHCl3–MeOH–H2O=8.5:1:0.1) to
NGF and Cell Viability Assays In this study, C6 glioma
give six subfractions [Fr. G31–G36]. Fraction G34 (2.5g) was cells were used to measure the induction of NGF release into
separated by preparative reversed-phase HPLC (flow rate; the culture medium. C6 cells were seeded onto 24-well plates
2mL/min, 25% MeCN) to yield 5 (8mg, tR=27.8min). Frac- at a density of 1×105 cells/well, and after 24h, the cells were
tion G32 (2.5g) was separated by preparative reversed-phase treated with serum-free Dulbecco’s modified Eagle’s medium
HPLC (flow rate; 2mL/min, 25% MeCN) to yield 7 (4mg, (DMEM) containing different concentrations of compound
tR=25.2min) and 9 (4mg, tR=18.6min). Fraction G6 (70mg) for an additional 24h. The medium was collected from the
was purified by preparative reversed-phase HPLC (flow rate; cultured plates and the NGF level was measured using an
2mL/min, 30% MeCN) to yield 1 (11mg, tR=22.3min). Frac- enzyme-linked immunosorbent assay (ELISA) kit. Cell viabil-
tion H (1.8g) was separated over an RP-C18 silica gel column ity was measured using an 3-(4,5-dimethylthiazol-2-yl)-2,5-
(230–400 mesh, 80g, 45% MeOH) to give twelve subfractions diphenyltetrazolium bromide (MTT) assay, as described
[Fr. H1–H12 (each 1.0L)]. Fraction H3 (72mg) was further previously.26) The results are expressed as a percentage of
separated using a Lobar-A Si gel 60 (240×10mm) column the control group (untreated cells). 6-Shogaol was used as the
(CHCl3–MeOH–H2O=5:1:0.1) to yield 10 (6mg) and 11 positive control.28)
(3mg). Fraction H5 (94mg) was separated by preparative
Cytotoxicity Assay A sulforhodamine B bioassay (SRB)
reversed-phase HPLC (flow rate; 2mL/min, 27% MeCN) to was used to determine the cytotoxicity of each compound
yield 6 (3mg, tR=21.5min) and 12 (10mg, tR=18.9min). Frac- against four cultured human cancer cell lines.29) The cell lines
tion H6 (166mg) was separated using a Lobar-A Si gel 60 (National Cancer Institute, Bethesda, MD, U.S.A.) used were
(240×10mm) column (CHCl3–MeOH–H2O=6:1:0.1) to yield A549 (non-small cell lung adenocarcinoma), SK-OV-3 (ovarian
2 (4mg). Fraction H9 (91mg) was separated by preparative malignant ascites), SK-MEL-2 (skin melanoma), and HCT-15
reversed-phase HPLC (flow rate; 2mL/min, 30% MeCN) to (colon adenocarcinoma). Cisplatin (Sigma Chemical Co.,
yield 14 (6mg, tR=18.2min) and 15 (6mg, tR=15.7 min). Frac- ≥98%) was used as a positive control. The cytotoxic activi-
tion H10 (102mg) was separated by semi-preparative reversed- ties of cisplatin against the A549, SK-OV-3, SK-MEL-2, and
phase HPLC (flow rate; 2mL/min, 30% MeCN) to yield 8 HCT15 cell lines showed IC50 values of 1.96, 2.11, 1.17, and
(3mg, tR=24.3min) and 13 (5mg, tR=26.1 min).
Dilatioside A (1)
3.04 µM, respectively. Tested compounds were demonstrated
to be pure as evidenced by NMR and HPLC analysis (purity
Amorphous gum; [α]D25 −181.8 (c=0.56, MeOH); IR (KBr): ≥95%).
νmax 3419, 3185, 2965, 2845, 1710, 1633, 1516, 1270, 1033,
Statistical Analysis All results are presented as the
1009 cm−1; UV λmax (MeOH) nm (logε): 204 (1.7), 231 (0.7), mean standard error of the mean (S.E.M.). Significant dif-
1
282 (0.3) nm; H- (700MHz) and 13C- (175MHz) NMR data, ferences between experimental groups were determined using
see Table 1; HR-FAB-MS: m/z=771.2844 [M+Na]+ (Calcd for one-way ANOVA followed by a Newman–Keuls post hoc test
C37H48NaO16, 771.2840).
using GraphPad Prism 5 (GraphPad Software Inc., La Jolla,
Dilatioside B (2)
CA, U.S.A.). p<0.05 was considered statistically significant.
Amorphous gum; [α]D25 −47.7 (c=1.11, MeOH); IR (KBr):
νmax 3417, 3192, 2968, 2360, 2336, 1054, 1032, 1012cm−1;
Acknowledgments This research was supported by the
UV λmax (MeOH) nm (logε): 224 (0.11), 243 (0.05), 270 Basic Science Research Program through the National Re-
1
(0.04) nm; H- (700MHz) and 13C- (175MHz) NMR data, see search Foundation of Korea (NRF), funded by the Ministry
Table 1; HR-FAB-MS: m/z=781.2534 [M+Na]+ (Calcd for of Education, Science and Technology (2016R1A2B2008380),
C34H46NaO19, 781.2531).
and we would like to acknowledge financial support from
Alkaline Methanolysis of Compounds 1 and 2 Com- the Bio & Medical Technology Development Program of
pounds 1 and 2 (each 1.0mg) were hydrolyzed with 0.5mol/L the NRF funded by the Korean government, MSIP (NRF-
KOH in MeOH (1mL) at room temperature for 1h. The mix- 2014M3A9B6069338). We are thankful to the Korea Basic
ture was subsequently eluted using an ion exchange column Science Institute (KBSI) for the measurements on the mass
(Dowex® 50WX8 hydrogen form, Sigma-Aldrich) in 100% spectra.
MeOH to remove KOH. The reaction mixtures of 1 were sepa-
rated through semi-prep. HPLC (30% MeCN) to give 1a (=3)
Conflict of Interest The authors declare no conflict of
(0.3mg) and 1b (0.4mg), which were identified as oleoside di- interest.
methyl ester (1a) and (−)-secoisolariciresinol (1b) by compari-

364
Chem. Pharm. Bull.
Vol. 65, No. 4 (2017)
Koike K., Bioorg. Med. Chem. Lett., 21, 6426–6429 (2011).
15) Kim C. S., Kwon O. W., Kim S. Y., Lee K. R., J. Braz. Chem. Soc.,
25, 907–912 (2014).
16) Sugiyama M., Nagayama E., Kikuchi M., Phytochemistry, 33,
1215–1219 (1993).
17) Kadowaki E., Yoshida Y., Baba N., Nakajima S., Z. Naturforsch.,
58, 441–445 (2003).
18) Tanahashi T., Sakai T., Takenaka Y., Nagakura N., Chen C. C.,
Chem. Pharm. Bull., 47, 1582–1586 (1999).
19) Takenaka Y., Tanahashi T., Shintaku M., Sakai T., Nagakura N.,
Parida, Phytochemistry, 55, 275–284 (2000).
20) Iossifova T., Vogler B., Kostova I., Phytochemistry, 49, 1329–1332
(1998).
21) Shen Y. C., Chen C. F., Gao J., Zhao C., Chen C. Y., J. Chin. Chem.
Soc., 47, 367–372 (2000).
22) Machida K., Kaneko A., Hosogai T., Kakuda R., Yaoita Y., Kikuchi
M., Chem. Pharm. Bull., 50, 493–497 (2002).
23) Sung S. H., Kim E. S., Lee K. Y., Lee M. K., Kim Y. C., Planta
Med., 72, 62–64 (2006).
24) Tanahashi T., Takenaka Y., Nagakura N., Phytochemistry, 41, 1341–
1345 (1996).
25) Fukuyama Y., Koshino K., Hasegawa T., Yamada T., Nakagawa K.,
Planta Med., 53, 427–431 (1987).
26) Reif D. W., McCreedy S. A., Arch. Biochem. Biophys., 320, 170–176
(1995).
27) Jang S. W., Suh W. S., Kim C. S., Kim K. H., Lee K. R., Arch.
Pharm. Res., 38, 1943–1951 (2015).
28) Kim C. S., Subedi L., Park K. J., Kim S. Y., Choi S. U., Kim K. H.,
Lee K. R., Fitoterapia, 106, 147–152 (2015).
29) Skehan P., Storeng R., Scudiero D., Monks A., McMahon J., Vistica
D., Warren J. T., Bokesch H., Kenney S., Boyd M. R., J. Natl. Can-
cer Inst., 82, 1107–1112 (1990).
Supplementary Materials The online version of this
article contains supplementary materials. NMR spectra and
HR-MS of 1 and 2 are available as supplementary materials.
References
1) Lin B., Pirrung M. C., Deng L., Li Z., Liu Y., Webster N. J. G., J.
Pharmacol. Exp. Ther., 322, 59–69 (2007).
2) Sofroniew M. V., Howe C. L., Mobley W. C., Annu. Rev. Neurosci.,
24, 1217–1281 (2001).
3) Colafrancesco V., Villoslada P., Arch. Ital. Biol., 149, 183–192
(2011).
4) Priyanka A. W., Swati R. D., Vilasrao J. K., World J. Pharm. Sci., 3,
1087–1094 (2015).
5) Fernyhough P., Diemel L. T., Brewster W. J., Tomlinson D. R., Neu-
roscience, 62, 337–344 (1994).
6) Wang Q. H., Ao W. L. J., Wu Q., Ying X., Helv. Chim. Acta, 95,
1680–1685 (2012).
7) Xu Q., Li Q., Liu Y., Feng Y., Yang S., Li X., Chem. Nat. Compd.,
46, 366–369 (2010).
8) Yin W. P., Zhao T. Z., Zhang H. Y., J. Asian Nat. Prod. Res., 10,
95–100 (2008).
9) Park H. J., Lee M. S., Lee K. T., Sohn I. C., Han Y. N., Miyamoto
K. I., Chem. Pharm. Bull., 47, 1029–1031 (1999).
10) El-Desouky S. K., Gamal-Eldeen A. M., Pharm. Biol., 47, 872–877
(2009).
11) Bai N., He K., Ibarra A., Bily A., Roller M., Chen X., Ruhl R., J.
Nat. Prod., 73, 2–6 (2010).
12) Xie L. H., Akao T., Hamasaki K., Deyama T., Hattori M., Chem.
Pharm. Bull., 51, 508–515 (2003).
13) Tanahashi T., Takenaka Y., Akimoto M., Okuda A., Kusunoki Y.,
Suekawa C., Nagakura N., Chem. Pharm. Bull., 45, 367–372 (1997).
14) Bi X., Li W., Sasaki T., Li Q., Mitsuhata N., Asada Y., Zhang Q.,