Eudesmane Glycosides from Ambrosia artemisiifolia (Common Ragweed) as Potential Neuroprotective Agents

Article abstract of DOI:10.1021/acs.jnatprod.8b00841

In Alzheimer's disease, amyloid-β (Aβ) accumulation in the brain results in neuronal cell death and is one of the major causes of dementia. Because the current therapeutic agents are not yet sufficiently effective or safe, there have been attempts to find

Full text of DOI:10.1021/acs.jnatprod.8b00841

Article
pubs.acs.org/jnp
Cite This: J. Nat. Prod. XXXX, XXX, XXX−XXX
Eudesmane Glycosides from Ambrosia artemisiifolia (Common
Ragweed) as Potential Neuroprotective Agents
Jin-Pyo An,† Thi Kim Quy Ha,† Hyun Woo Kim,† Byeol Ryu,† Jinwoong Kim,† Junsoo Park,‡
Chul Ho Lee,§ and Won Keun Oh*,†
†Korea Bioactive Natural Material Bank, Research Institute of Pharmaceutical Sciences, College of Pharmacy, Seoul National
University, Seoul 08826, Republic of Korea
‡Division of Biological Science and Technology, Yonsei University, Wonju 220-100, Republic of Korea
§Laboratory Animal Resource Center, Korea Research Institute of Bioscience and Biotechnology (KRIBB), Daejeon 34141,
Republic of Korea
S
* Supporting Information
ABSTRACT: In Alzheimer’s disease, amyloid-β (Aβ) accumulation in the brain
results in neuronal cell death and is one of the major causes of dementia. Because
the current therapeutic agents are not yet sufficiently effective or safe, there have
been attempts to find new neuroprotective chemicals against Aβ-induced
cytotoxicity. A 70% EtOH extract of whole plants of Ambrosia artemisiifolia
(common ragweed) was selected after the screening of a natural extract library.
Seven new eudesmane-type glycosides (1−7) and seven known compounds (8−
14) were obtained through bioactivity-guided fractionation from the aerial parts of
this plant. Their structures were determined on the basis of their nuclear magnetic
resonance spectra, high-resolution electrospray ionization mass spectrometry
analysis, and electronic circular dichroism calculations. Among them, compounds
1, 2, 4−6, 8, 9, 11, 13, and 14 showed protective effects against Aβ-induced
cytotoxicity in Aβ42-transfected HT22 cells. The most active compounds, 5 and 6,
exhibited moderate protective activity dose-dependently (10, 20, and 40 μM).
lzheimer’s disease (AD) is a neurodegenerative disorder
Aβ aggregate-induced neurotoxicity, oxidative stress, and
inflammatory reactions are related to the etiology of AD.8
Several studies have reported that the cytotoxic effect of Aβ is
attributed to its aggregation. Human amyloid precursor protein
(hAPP) was transferred to mice, and increased levels of Aβ
altered the neuronal expression of synaptic-activity-regulated
genes;9 an Aβ-induced model of AD in rodents showed deficits
in learning and memory abilities.10 To date, the U.S. Food and
Drug Administration (FDA) has approved only five drugs for
AD: donepezil, galantamine, rivastigmine, tacrine, and
memantine.11 However, these medications only temporarily
improve cognition and delay the disease progression for a short
period.12
Three hundred extracts were selected from the KBNMB
(Korea Bioactive Natural Material Bank) and screened for
their inhibitory activities against Aβ-induced NO production.
As a result, 10 μg/mL of the ethyl acetate extract of Ambrosia
artemisiifolia L. (Asteraceae) inhibited ∼50% of Aβ-induced
NO production (data not shown). The genus Ambrosia
consists of approximately 50 species worldwide.13 A.
artemisiifolia, native to the United States, spreads with
competitive seeds and can grow rapidly; it is now recognized
A
that features declines of judgment, cognition, memory,
and behavior.1 AD is the leading cause of dementia, becoming
one of the largest public health concerns. Currently, 30 million
people suffer from AD, and the World Health Organization has
reported that the number of patients will triple over the next
20 years.2 Owing to the steady growth of the aging population,
an increasing prevalence of dementia is expected in both
developing and developed countries.
The accumulation of Aβ peptide into senile plaques has
been considered one of the major causes of AD.3 Two types of
Aβ peptides, Aβ40 and Aβ42, can exist outside as well as inside
nerve cells. Aβ has been found to be a normal metabolite of
amyloid precursor protein (APP), produced by the cleavage of
APP by β-secretase and γ-secretase.4 The senile plaque mainly
consists of Aβ40 and Aβ42, which can produce several types of
oligomeric structures such as protofibrils, fibrils, and plaques
according their extent of oligomerization.5 Several Aβs linked
through hydrogen bonds form a protofibril, and the
aggregation of several protofibrils generates a fibril. Further
aggregation of fibrils forms plaques that are highly toxic to
neurons.6 In AD, Aβ deposits as diffuse and compact plaques;
the latter consisting of Aβ fibrils are associated with a change
of pathology in the brain and surrounding parenchyma,
whereas diffuse plaques are not involved in degenerative
changes.7
Received: October 11, 2018
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.8b00841
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
as an invasive species in Korea. Previous chemical studies of
this plant have reported several sesquiterpenes, triterpenes, and
caffeic acid derivatives.14,15 In this study, the isolation and
structure elucidation of seven new eudesmane-type glycosides
(1−7) along with seven previously described compounds (8−
14) are reported. All compounds were also examined for their
protective activities against Aβ-induced cytotoxicity in Aβ42-
transfected HT22 cells.
eudesmane-type sesquiterpene, with the corresponding carbon
signals at C-3 (δC 122.0), C-4 (δC 137.0), C-12 (δC 22.7), C-
13 (δC 22.2), C-14 (δC 12.7), and C-15 (δC 24.1).15 The
COSY correlations between H-1/H-2, H-2/H-3, H-3/H-5, H-
5/H-6, H-6/H-7, H-7/H-8, H-8/H-9, and H-9/H-14 together
with the heteronuclear multiple bond correlation (HMBC)
between H-1/C-14 confirmed the A/B ring system of an
eudesmane sesquiterpene. The HMBCs from two doublet
methyl groups to one tertiary carbon (H3-12/C-11 and H3-13/
C-11) indicated an isopropyl group, and its position was
confirmed by the HMBC from H-11 (δH 2.07) to C-7 (δC
43.2). Other proton signals of two pairs of methylene protons
(δH 2.63 (2H, m) and 2.61 (2H, m)) and a singlet methyl
group at δH 1.38 (3H, s) were predicted to be from an 3′′-
hydroxy-3′′-methylglutaryl (HMG) moiety, with correspond-
ing carbon signals at C-2′′ (δC 46.6), C-4″ (δC 46.3), and C-6″
(δC 27.9). The chemical structure of compound 1 is similar to
1α-cinnamolyoxy-6β-hydroxy-5,10-bis-epi-eudesm-3-ene iso-
lated from A. artemisioides15 but additionally showed the
presence of an HMG moiety and a glucopyranosyl group along
with the absence of a cinnamoyl group. The linkage of an
RESULTS AND DISCUSSION
■
Compound 1 was isolated as an amorphous powder with [α]D25
6.1 (c 0.5, MeOH). The high-resolution electrospray ionization
mass spectrometry (HRESIMS) data (m/z 559.2773 [M −
H]−, calcd for C27H43O12, 559.2760) established the molecular
formula as C27H44O12. The infrared (IR) bands at 3395, 1725,
and 1625 cm−1 were indicative of hydroxy, carbonyl, and
1
olefinic groups. The H NMR data showed an olefinic proton
at δH 5.28 (1H, br s), an allylic methyl group at δH 1.91 (3H,
br s), a singlet methyl group at δH 0.92 (3H, s), and two
doublet methyl moieties (δH 1.04 (3H, d, J = 6.5 Hz) and 0.90
(3H, d, J = 6.5 Hz)), suggesting the occurrence of an
B
DOI: 10.1021/acs.jnatprod.8b00841
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 1. (a) Selected HMBC (H → C) correlations for compounds 1−7. (b) Selected NOESY (
compounds 1 and 3.
) correlations for aglycones of
Figure 2. Absolute configuration of the HMG moiety in compound 1. Steps applied were amination (a), reduction (b), and acetylation (c).
Synthesis of key intermediate 1b: (a) (S)-1-phenylethylamine, PyBOP, Et3N, HOBt, DMF of 9 h; (b) LiBH4, THF of 24 h; and (c) Ac2O, pyridine
of 24 h.
C
DOI: 10.1021/acs.jnatprod.8b00841
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 3. Comparison of experimental and calculated ECD spectra of compounds 1 (a), 2 (b), and 3 (c).
HMG moiety was determined at C-9 by the HMBC from H-9
(δH 4.95) to C-1″ (δC 172.3). The glycosylation position was
established by HMBC cross-peaks between H-1′ (δH 4.44) and
C-6 (δC 77.6) (Figure 1a). A concerted method that consists of
the process of amination, reduction, and acetylation revealed
the absolute configuration of the HMG group.16 (S)-1-
Phenylethylamine was used to obtain the aminated compound
1a. Subsequently, compound 1a was reduced with LiBH and
then acetylated with Ac2O to give 5-O-acetyl-1-[(S)-
phenylethyl]mevalonamide (1b) (Figure 2). The proton
D
DOI: 10.1021/acs.jnatprod.8b00841
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
NMR information for 1b was consistent with (3R)-5-O-acetyl-
1-[(S)-phenylethyl]mevalonamide but not (3S)-5-O-acetyl-1-
[(S)-phenylethyl]mevalonamide (Table S1).17 The relative
configuration of the aglycone of 1 was determined by the
interpretation of its nuclear Overhauser effect spectroscopy
(NOESY) data (Figure 1b and Figure S6). In the NOESY
spectrum, H-1 (δH 3.93) correlated with H-5 (δH 3.01),
whereas H-14 (δH 0.92) was correlated to H-6 (δH 4.43), H-7
(δH 1.59), and H-9 (δH 4.95), as shown in Figure 1b. With
respect to the 13C NMR chemical shifts, a significant difference
was suggested for C-1 according to the configuration of the
hydroxy group found in previous research for eudesmane-type
sesquiterpenes.18−21 The chemical shift of C-1 (δC 72.1) with
an α-oriented hydroxy group is relatively upfield compared
with C-1 (δC 79.4) with a β-oriented hydroxy group.18,19 The
HMG moiety of 1 was removed by acid hydrolysis to yield 1c,
and the similar carbon chemical shift of C-1 (δC 71.7) as that
of previous reported compounds (δC 72.1) with an α-oriented
hydroxy group confirmed the configuration at C-1. To define
the absolute configuration of 1, electronic circular dichroism
(ECD) calculations were performed for (1S,5R,6R,7S,9S,10S)-
1 and (1R,5S,6S,7R,9R,10R)-1 by time-dependent density
functional theory (TDDFT) at the B3LYP/def-SV(P) level
(Table S2). Conformers for (1S,5R,6R,7S,9S,10S)-1 were
searched using the molecular mechanics force field, and four
major conformers occupied 99% of the Boltzmann population
(Figure S52). The absolute configuration of compound 1 was
established based on the good agreement between the
calculated ECD spectrum of (1S,5R,6R,7S,9S,10S)-1 and the
experimental ECD spectrum (Figure 3a). Therefore, com-
pound 1 was defined as 1α,6β,9β-trihydroxy-5,10-bis-epi-
eudesm-3-ene-9-O-[(S)-3″-hydroxy-3′′-methylglutaryl]-6-O-β-
D-glucopyranoside.
1β,6β-dihydroxy-trans-eudesm-3-en-6-O-β-D-glucopyranoside
isolated from Aster koraiensis,22 except for the presence of the
HMG moiety and the absence of a hydroxycinnamoyl group.
The HMBC from H-1 (δH 4.78) to C-1′′ (δC 172.5) and from
an anomeric proton (δH 4.43) to C-6 (δC 75.2) suggested that
the HMG unit and the glucopyranosyl group are located at C-1
and C-6, respectively. The relative configuration of compound
3 was suggested based on the NOESY cross-peaks (Figure 1b
and Figure S17). In the NOESY spectrum, H-1 (δH 4.78)
showed a correlation with H-5 (δH 2.57), whereas H-14 (δH
0.99) gave correlations with H-6 (δH 4.41) and H-7 (δH 1.90),
as depicted in Figure 1b. Acid hydrolysis of 3 yielded 3a. The
chemical shift at C-1 (δC 77.3) of compound 3a (Table S5)
was well matched with previous studies for compounds in
which C-1 (δC 79.4) with a β-oriented hydroxy group is shifted
downfield compared with C-1 (δC 72.1) in analogues with an
α-oriented hydroxy group.18,19 Thus H-1 was deduced to be α-
oriented, whereas H-6 and H-7 were found to have a β-
orientation. Furthermore, the β-oriented hydroxy group at C-1
showed a γ-gauche shielding effect over H3-14, which, in turn,
resulted in an upfield chemical shift of C-14. The absolute
configuration of compound 3 was determined by ECD
calculations (Table S6). The experimental ECD spectrum of
compound 3 was in good agreement with the calculated
spectrum of 1R,5S,6S,7R,10R of 3 (Figure 3c). Therefore,
compound 3 was assigned as 1β,6α-dihydroxy-7-epi-eudesm-3-
ene-1-O-[(S)-3″-hydroxy-3′′-methylglutaryl]-6-O-β-D-gluco-
pyranoside.
Compound 4, an amorphous powder with [α]2D5 31.8 (c 0.3,
MeOH), was determined to have a molecular formula of
C29H46O13, as suggested by the HRESIMS peaks at m/z
601.2876 [M − H]− (calcd for C29H45O13, 601.2866). The IR
absorption of compound 4 indicated the presence of hydroxy
(3389 cm−1), carbonyl (1731 cm−1), and olefinic (1615 cm−1)
groups. The NMR pattern of 4 was similar to that of 1, and
compound 4 was found to possess an acetyl group linked to
the glucopyranoside moiety. The HMBC from the signals of
H-6a′ (δH 4.41) and the acetyl group (δH 2.04) to the acetate
carbonyl carbon (δC 172.7) suggested that this acetyl group is
attached to C-6′ of the glucose. The 6-O-acetyl-β-D-
glucopyranosyl unit was determined to be at C-6 of the
aglycone by the HMBC from the anomeric proton (δH 4.43) to
C-6 (δC 78.0). In addition, the position of the HMG moiety
was deduced by the HMBC from H-9 (δH 4.99) to C-1′′ (δC
172.8). The relative configuration of 4 was assigned based on
the NOESY cross-peaks from H-1 (δH 3.91) to H-5 (δH 3.03)
and from H-14 (δH 0.92) to H-6 (δH 4.40), H-7 (δH 1.54), and
H-9 (δH 4.99). Experimental ECD and calculated spectra gave
Compound 2 was isolated as an amorphous powder with
[α]2D5 13.8 (c 0.3, MeOH), and its molecular formula of
C27H44O12 was analyzed based on the HRESIMS (m/z
1
559.2794 [M − H]−, calcd for C27H43O12, 559.2760). Its H
and 13C NMR data revealed that compound 2 possesses a
similar structure to compound 1. The major difference found
for compound 2 was in the attachment of HMG to C-1 (δC
75.7), which was confirmed by a HMBC from H-1 (δH 5.17)
to C-1′′ (δC 172.6). The HMBC from the anomeric proton
(δH 4.43) to C-6 (δC 77.9) was used to establish the
glycosylation position. NOESY correlation peaks of H-1/H-5,
H-14/H-6, H-14/H-7, and H-14/H-9 revealed the relative
configuration of 2. To compare the chemical shift of C-1 with
those of other analogues, compound 2 was hydrolyzed to
remove the HMG moiety from C-1 and yielded 2a (Table S3).
The chemical shift of C-1 in 2a proved to be similar to that of
previously reported compounds that possess an α-oriented
hydroxy group.18,20 To determine the absolute configuration of
2, ECD calculations were applied (Figure S53 and Table S4) in
which the calculated ECD spectrum of (1S,5R,6R,7S,9S,10R)-2
and the experimental ECD spectrum were well matched
(Figure 3b). Therefore, compound 2 was determined
structurally as 1α,6β,9β-trihydroxy-5,10-bis-epi-eudesm-3-ene-
1-O-[(S)-3″-hydroxy-3′′-methylglutaryl]-6-O-β-D-glucopyrano-
side.
the absolute configuration of compound
4 as
1S,5R,6R,7S,9S,10S (Figure S56 and Table S7). Thus
compound 4 was characterized as 1α,6β,9β-trihydroxy-5,10-
bis-epi-eudesm-3-ene-9-O-[(S)-3″-hydroxy-3′′-methylglutaryl]-
6-O-(6′-O-acetyl)-β-D-glucopyranoside.
Compound 5 was obtained as a colorless gum with [α]D25
−2.0 (c 0.5, MeOH) and analyzed to have the molecular
formula of C26H44O12 based on the HRESIMS ion peak at m/z
593.2812 [M + HCOO]− (calcd for C27H45O14, 593.2815).
The IR absorption of compound 5 indicated the presence of
hydroxy (3410 cm−1) and olefinic (1620 cm−1) groups. The
chemical structure of 5 proved to be similar to that of
compound 2, except for the attachment of the arabinopyr-
anosyl unit at C-1. This was confirmed by the HMBCs from H-
1′ (δH 4.37) to C-1 (δC 82.1) and from H-1 (δH 4.01) to C-1′
Compound 3, an amorphous powder with [α]2D5 1.7 (c 0.5,
MeOH) had a molecular formula of C27H44O11 from the
HRESIMS ion peak at m/z 543.2820 [M − H]− (calcd for
C27H43O11, 543.2811). The NMR data of compound 3 showed
similar patterns to those of 9β-O-(E-p-hydroxycinnamoyl)-
E
DOI: 10.1021/acs.jnatprod.8b00841
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 4. (a) Neuroprotective effects of isolated compounds from A. artemisiifolia against cytotoxicity induced by Aβ42 plasmid transfection into
HT22 cells. After 10 h of transfection, the transfected cells were exposed to the test compounds at a concentration of 40 μM, except for compounds
10 and 12 at 5 μM. (b) Compounds 5 and 6 at various concentrations. The cultures were incubated continuously for 24 h, and cell viability was
measured using the MTT method. Data are expressed as the means SD (n = 3); *p < 0.05, **p < 0.01, ***p < 0.001, compared with the Aβ42
transfection without treatment group.
(δC 105.9). The rotating frame nuclear Overhauser effect
spectroscopy (ROESY) spectrum showed correlations between
H-1 (δH 4.01) and H-5 (δH 2.97) and between H-6 (δH 4.43),
H-7 (δH 1.46), and H-14 (δH 0.91) (Figure S28). Acid
hydrolysis of 5 gave 5a, for which the 13C NMR data were
identical to those of 1c (Tables S8 and S9). Reversed-phase
high-performance liquid chromatography (HPLC) analysis
(coinjection) also demonstrated that 5a is the same as 1c
(Figure S49). Therefore, compound 5 was defined as
1α,6β,9β-trihydroxy-5,10-bis-epi-eudesm-3-ene-1-O-α-L-arabi-
nopyranosyl-6-O-β-D-glucopyranoside.
Compound 6 was isolated as an amorphous powder with
[α]2D5 1.7 (c 0.2, MeOH). HRESIMS analysis suggested a
molecular formula of C21H36O8 based on the [M + HCOO]−
peak at m/z 461.2382, calcd for C22H37O10 (m/z 461.2392).
The IR absorption of compound 6 indicated the occurrence of
hydroxy (3413 cm−1) and olefinic (1615 cm−1) groups. The
13C NMR data of compound 6 were identical to those of 1c,
and coinjection by reversed-phase HPLC analysis confirmed
that compound 6 is identical to 1c (Figure S50). Therefore,
compound 6 was determined to be 1α,6β,9β-trihydroxy-5,10-
bis-epi-eudesm-3-ene-6-O-β-D-glucopyranoside.
Compound 7 was purified as a colorless gum with [α]D25
−1.6 (c 0.5, MeOH) and had a molecular formula of
C26H44O11 from the HRESIMS ion at m/z 577.2850 [M +
HCOO]− (calcd for C27H45O13, 577.2866). The broad IR
peaks at 3403 and 1630 cm−1 indicated hydroxy and olefinic
groups, respectively. The NMR of compound 7 resembled
compound 3, except for the presence of an apiofuranosyl group
and the absence of an HMG moiety in 7. The anomeric proton
at δH 4.99 (d, J = 1.9 Hz) was shown to belong to an apiose
moiety from a series of pentose proton signals ((δH 3.88, d, J =
1.8 Hz; δC 78.1), (δH 3.90, d, J = 9.6 Hz and δH 3.74, d, J = 9.6
Hz; δC 75.0), and (δH 3.58, s; δC 66.0)) that included two
methylene groups based on the COSY spectrum. HMBC cross-
peaks between H-1′ (δH 4.43) and C-6 (δC 75.7), and between
H-6′ (δH 3.93) of the glucose and C-1″ (δC 110.7) of the
apiose suggested the relative positions of the two sugar units.
The NOESY spectrum exhibited cross-peaks from H-1 (δH
3.49) to H-5 (δH 2.45) and from H-14 (δH 0.87) to H-6 (δH
4.38) and H-7 (δH 1.86). The sugars were confirmed as D-
F
DOI: 10.1021/acs.jnatprod.8b00841
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 5. (a) Effects of compounds 5 and 6 (40 μM) on the condensation of green fluorescence in Aβ42-transfected cells. (b) HT22 cells were
transfected with pEGFP-C1/Aβ42 plasmid using Lipofectamine 2000 reagent for 10 h. The transfected cells were then incubated with test
compounds for 24 h, and fluorescence images were captured using a fluorescence microscope. Fluorescence intensity of the transfected cells was
successfully analyzed using flow cytometry. After the Aβ42-transfected cells were exposed to the test compounds for 24 h, the cells were collected
and investigated by flow cytometry. Results are presented as the means SD (n = 3); ###p < 0.001 compared with the nontransfection group; **p
< 0.01 compared with the Aβ42 transfection without treatment group.
glucose and D-apiose using the method of Tanaka et al.23 The
arabinofuranosyl moiety was removed by acid hydrolysis to
afford 7a, and reversed-phase HPLC on coinjection proved
that 7a is identical to 3a (Figure S51). Accordingly, compound
7 was structurally determined to be 1β,6α-dihydroxy-7-epi-
eudesm-3-ene-6-O-β-D-apiofuranosyl-(1→6)-β-D-glucopyrano-
side.
By comparison of the spectroscopic data with the literature
data, the seven known isolated compounds were identified as
1β,6α-dihydroxyeudesman-4(15)-ene-1-O-β-D-glucopyrano-
side (8),24 3-deacetoxylhymenolane (9),25 3α,4β-diacetylhy-
menoratin (10),26 ambrosic acid (11),27 cumambrin B (12),28
N1,N5,N10-tri-p-coumaroylspermidine (13),29 and N1,N5-di-p-
coumaroyl-N10-caffeoyl spermidine (14).30
AD is a fatal degenerative brain disorder that affects the
elderly, and approximately 60−80% of AD patients progress to
dementia.31 Many studies have demonstrated that the Aβ
peptide is the major protein composition of neuritic plaques.32
Thus many attempts have been made to find new chemical
entities to protect the brain from Aβ-induced cytotoxicity. In
the present study, bioassay-guided fractionation with a 70%
EtOH extract of A. artemisiifolia led to the isolation of a total of
14 compounds from the EtOAc-soluble partition, which
showed neuroprotective activity. To investigate the protective
effect against the cytotoxicity induced by Aβ, HT22 cells were
used and were transfected with Aβ42 plasmids using Lipofect-
amine 2000 reagent. After 10 h of transfection, 40 μM of
compounds 1−9, 11, 13, and 14 and 5 μM of compounds 10
and 12 were added because compounds 10 and 12 showed
significant cytotoxicity at 40 μM (Figure S57). Cell viability
was measured by a 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl
tetrazolium bromide (MTT) assay. Compounds 1, 2, 4−6, 8,
9, 11, 13, and 14 showed protective effects against Aβ-induced
cytotoxicity, and the activity of compound 5 was comparable
to that of andrographolide as a positive control (Figures S61
and S62). In addition, compounds 5 and 6 presented dose-
dependent activity (10, 20, and 40 μM) in Aβ42-transfected
cells (Figure 4a,b). Thus flow cytometry analysis was
performed with the selected isolates, 5 and 6. As shown in
Figure 5, compounds 5 and 6 showed moderate decreases in
fluorescence intensity in Aβ42-transfected HT22 cells at a
concentration of 40 μM when compared with a vehicle.
Structure−activity relationship (SAR) analysis revealed that
the neuroprotective effects can be attributed to the 5,10-bis-
epi-eudesm-3-ene-6-O-β-D-glucopyranosyl moiety in Aβ42-
transfected cells. Among all isolates, compounds possessing
the 5,10-bis-epi-eudesm-3-ene-6-O-β-D-glucopyranosyl moiety
increased cell viability against Aβ-induced cytotoxicity.
EXPERIMENTAL SECTION
■
General Experimental Procedures. Analytical-grade solvents
were purchased from Fisher Scientific (Pittsburgh, PA). Solvents for
extraction were obtained from Daejung Chemicals & Metals
(Siheung, Korea). Optical rotations were measured on a JASCO P-
2000 polarimeter (JASCO International, Tokyo, Japan). ECD and UV
spectra were determined with a Chirascan ECD spectrometer
(Applied Photophysics, Leatherhead, U.K.). IR spectra (KBr) were
obtained using a Nicolet 6700 FT-IR spectrometer (Thermo Fisher
Scientific, Waltham, MA). The 1H and 13C NMR spectra were
acquired on Bruker Avance-800 (Billerica, MA), JNM-ECA-600
(Tokyo, Japan), and Bruker Avance-400 spectrometers at Seoul
National University, Seoul, Korea. HRESIMS data of all compounds
were determined with an Agilent 6530 Q-TOF mass spectrometer
equipped with an Agilent 1260 Infinity HPLC (Agilent Technologies,
Santa Clara, CA). Silica gel (63−200 μm, Merck, Darmstadt,
Germany) and RP-C18 (40−63 μm, Merck) columns were applied
G
DOI: 10.1021/acs.jnatprod.8b00841
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
1
Table 1. H NMR Spectroscopic Data (CD3OD) of Compounds 1−7
a
b
b
c
c
b
c
no.
1
2
3
4
5
6
7
1
3.93, dd
(9.8, 7.0)
2.25, m
1.91, m
5.17, dd
(8.8, 7.6)
2.61, m
1.97, m
4.78, dd
3.91, dd
(10.0, 6.9)
2.28, m
1.94, ddd,
(10.1, 7.1, 2.2)
5.28, br s
−
4.01, dd
(9.6, 7.2)
2.42, m
2.05, m
4.07, dd
(9.4, 7.4)
2.35, m
1.89, m
3.49, dd
(9.5, 7.3)
2.37, d (15.9)
1.93, m
(9.8, 6.7)
2.25, m
1.95, m
2
3
4
5
6
5.28, br s
−
3.01, br s
5.28, br s
−
3.09, br s
5.25, br s
−
2.57, m
4.41, dd
(11.8, 4.8)
1.90, m
1.77, d (14.4)
1.65, m
5.27, br s
−
2.97, br s
5.27, br s
−
2.99, br s
5.25, br s
−
2.45,br d (10.2)
4.38, dd
(11.1, 4.6)
1.86, m
1.68, m
3.03, br s
4.40, m
4.43, m
4.46, m
4.43, m
4.44, m
7
8
1.59, m
2.25, m
2.12, m
1.47, m
2.27, dt
(14.6, 7.1)
1.65, dd
(14.9, 9.7)
3.82, m
1.54, m
2.20, dt
1.46, m
2.25, m
1.68, ddd
(15.1, 9.7, 2.0)
1.42, m
2.28, m
1.62, ddd
(15.0, 9.8, 1.5)
(15.2, 6.4)
1.79, ddd
(16.0, 8.0, 0.8)
4.99, dd
(6.7, 1.4)
−
1.80, m
9
4.95, br d (5.9)
1.44, d (12.2)
1.30, m
−
2.24, m
1.05, d (6.0)
0.97, d (6.0)
0.99, s
1.93, br s
4.43, d (7.7)
3.22, m
3.35, m
3.23, m
3.21, m
3.65, dd
(11.8, 5.7)
3.87, dd
(11.9, 1.7)
−
4.07, dd
(7.3, 2.0)
−
3.93, d (6.5)
1.57, m
1.31, td (13.6, 4.8)
−
10
11
12
13
14
15
1′
2′
3′
4′
5′
−
−
−
2.07, m
1.04, d (6.5)
0.90, d (6.5)
0.92, s
2.03, m
2.08, m
2.04, m
1.99, m
2.21, m
1.01, d (6.5)
0.93, d (6.5)
0.97, s
1.88, br s
4.43, d (7.7)
3.20, m
3.35, m
3.33, m
3.24, m
3.79, m
1.03, d (6.6)
0.91, d (6.6)
0.92, s
1.89, br s
4.43, d (7.8)
3.24, m
3.36, m
3.41, dd (10.0, 6.9)
3.21, m
4.41, dd
(11.8, 2.1)
4.12, dd
(11.8, 7.0)
−
0.94, d (6.7)
1.02, d (6.6)
0.91, s
1.86, br s
4.37, d (7.6)
3.81, m
3.45, m
3.32, m
3.22, m3.15, m
0.93, d (6.5)
1.01, d (6.5)
0.83, s
1.85, br s
4.43, d (7.8)
3.17, m
3.32, m
3.32, m
3.23, m
3.79, dd
(11.8, 2.5)
3.68, dd
(11.8, 5.1)
0.99, d (6.8)
1.07, d (6.8)
0.87, s
1.91, br s
4.43, d (7.7)
3.21, m
3.31, m
3.27, m
3.31, m
3.93, dd
(11.1, 1.7)
3.60, dd
(11.1, 5.9)
4.99, d (1.9)
3.88, d (1.8)
−
3.90, d (9.6)
3.74, d (9.6)
3.58, s
1.91, br s
4.44, d (7.6)
3.20, m
3.34, m
3.31, m
3.22, m
3.80, dd
(11.0, 2.2)
3.70, dd
(10.8, 4.5)
−
6′
3.70, dd
(11.8, 4.9)
1′′
2′′
3′′
4′′
−
4.45, d (7.6)
3.17, m
3.32, m
2.63, m
−
2.61, m
2.65, m
−
2.63, m
2.64, m
−
2.61, m
2.67, m
−
2.63, m
3.54, m
5′′
6′′
−
1.38, s
−
1.37, s
−
1.35, s
−
1.42, s
3.24, m
3.81, m
3.68, dd
(11.9, 5.3)
AcO
2.04, s
a
b
c
NMR 400 MHz. NMR 600 MHz. NMR 800 MHz.
for column chromatography. RP-18 TLC plates and silica gel 60 F254
were used for TLC analysis. HPLC was performed on a Gilson HPLC
system using an OptimaPak C18 column (10 mm × 250 mm, 10 μm;
RS Tech, Seoul, Korea) and a C18 column (30 mm × 250 mm, 40−63
μm; Isu Industry, Seoul, Korea). Medium-pressure liquid chromatog-
raphy (MPLC) (Biotage-Isolera One, Biotage, Charlotte, NC) was
carried out with C18 SNAP cartridges (KP-C18-HS; 120 g, Biotage).
A Conflex 7 (Conflex, Tokyo, Japan) was used for the computed
conformer searches.33
Plant Material. The whole plants of A. artemisiifolia were
collected in September 2015 in Wonju, Gangwon Province, Korea.
The samples were identified botanically by W. K. Oh. A voucher
specimen (SNU2015-10) was deposited at the Medicinal Plant
Garden of Seoul National University, Seoul, Korea.
(3 × 3 L), and H2O. The EtOAc-soluble partition (96 g) was loaded
onto a silica gel column (9.5 × 60 cm) and eluted with n-hexane/
acetone (1:0 to 0:1) to obtain seven fractions (A−G). Fraction A (21
g) was chromatographed via an open RP-18 column (2.5 × 40 cm)
with MeOH/H2O (4:6 to 10:0) as solvent to give six subfractions.
Fraction A.3 (100 mg) was crystallized from MeOH/H2O (6:4) to
yield compound 11 (24 mg). Fraction A.4 (90 mg) was crystallized
from MeOH/H2O (6:4) to yield compound 12 (10 mg). Fraction A.6
(320 mg) was purified by semipreparative HPLC (43% aqueous
CH3CN, flow rate 3 mL/min) to isolate compounds 9 (6.0 mg; tR =
48.5 min) and 10 (5.5 mg; tR = 54.0 min). Fraction C (15 g) was
subjected to passage over an open RP-18 column (4.5 × 40 cm) with
MeOH/H2O (7:13 to 20:0) as solvent to give seven subfractions.
Further purification of fraction C.5 (450 mg) using semipreparative
HPLC (26% aqueous CH3CN, flow rate 3 mL/min) yielded
compound 8 (8.0 mg; tR = 28.2 min). Fraction D (10 g) was
subjected to the open RP-18 column (4.5 × 40 cm) separation using a
CH3CN/H2O gradient (7:13 to 20:0) to give eight subfractions.
Extraction and Isolation. The air-dried and powdered whole
plants of A. artemisiifolia (4 kg) were extracted with 70% EtOH (20
L) three times (each 1 day) at room temperature. The extract (405 g)
was partitioned with n-hexane (3 × 3 L), EtOAc (3 × 3 L), n-BuOH
H
DOI: 10.1021/acs.jnatprod.8b00841
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 2. 13C NMR Spectroscopic Data (CD3OD) of Compounds 1−7
a
b
b
c
c
b
c
no.
1
2
3
4
5
6
7
1
2
3
4
71.5, CH
33.4, CH2
122.0, C
137.0, C
46.6, C
75.7, CH
31.1, CH2
120.8, C
136.4, C
46.4, C
80.0, CH
30.6, CH2
121.7, C
138.1, C
47.1, C
70.9, CH
33.4, CH2
121.8, C
136.8, C
46.6, C
82.1, CH
33.0, CH2
121,7, C
136.4, C
46.2, C
71.6, CH
33.9, CH2
121.5, C
136.1, C
46.6, C
77.3, CH
33.5, CH2
122.5, C
137.8, C
47.3, C
5
6
7
8
9
77.6, CH
43.2, CH
29.1, CH2
74.1, CH
42.7, C
77.9, CH
43.5, CH
32.0, CH2
69.9, CH
42.0, C
75.2, CH
41.2, CH
22.1, CH2
32.1, CH2
40.1, C
78.0, CH
43.5, CH
29.4, CH2
73.9, CH
42.6, C
77.8, CH
43.6, CH
31.8, CH2
70.2, CH2
43.0, C
78.4, CH
43.7, CH
32.4, CH2
70.3, CH
42.9, C
75.7, CH
41.6, CH
22.7, CH2
32.1, CH2
41.3, C
10
11
12
13
14
15
1′
2′
3′
4′
5′
6′
1″
2″
3″
4″
5″
6″
AcO
29.2, CH
22.7, CH3
22.2, CH3
12.7, CH3
24.1, CH3
104.1, CH
75.4, CH
78.2, CH
71.7, CH
77.4, CH
62.8, CH2
172.3, C
46.6, CH2
70.9, C
30.1, CH
22.2, CH3
21.7, CH3
13.9, CH3
22.7, CH3
104.9, CH
75.6, CH
78.1, CH
71.5, CH
77.8, CH
62.6, CH2
172.6, C
46.7, CH2
70.8, C
26.3, CH
24.1, CH3
22.8, CH3
12.7, CH3
25.8, CH3
100.0, CH
75.7, CH
78.2, CH
72.2, CH
78.0, CH
63.1, CH2
172.5, C
46.5, CH2
70.7, C
29.4, CH
22.6, CH3
22.2, CH3
12.8, CH3
23.9, CH3
104.4, CH
75.1, CH
78.1, CH
71.9, CH
75.1, CH
65.2, CH2
172.8, C
46.9, CH2
70.9, C
46.6, CH2
175.0, C
28.0, CH3
172.7, C 20.8, CH3
29.9, CH
21.9, CH3
22.3, CH3
13.7, CH3
26.1, CH3
105.9, CH
77.9, CH
75.5, CH
71.2, CH
70.0, CH2
30.2, CH
21.6, CH3
22.1, CH3
12.7, CH3
22.6, CH3
104.9, CH
75.6, CH
78.1, CH
71.4, CH
77.8, CH
62.6, CH2
26.5, CH
22.9, CH3
24.1, CH3
11.6, CH3
26.1, CH3
100.0, CH
75.7, CH
78.2, CH
71.9, CH
76.8, CH
68.2, CH2
110.7, CH
78.1, CH
80.6, CH
75.0, CH
66.0, CH2
104.1, CH
75.7, CH
78.2, CH
71.6, CH
78.1, CH
62.7, CH2
46.3, CH2
172.3, C
27.9, CH3
46.4, CH2
175.2, C
27.8, CH3
46.1, CH2
175.5, C
27.8, CH3
a
b
c
NMR 100 MHz. NMR 150 MHz. NMR 200 MHz.
Fraction D.4 (800 mg) was subjected to Sephadex LH-20 column
chromatography with a MeOH/H2O system (17:3) to yield two
subfractions. Compounds 2 (7.0 mg; tR = 42.0 min) and 7 (3.6 mg; tR
= 17.0 min) were isolated from fraction D.4.1 using semipreparative
HPLC (26% aqueous CH3CN, flow rate 3 mL/min). Fraction D.5
(430 mg) was purified using semipreparative HPLC (26% aqueous
CH3CN, flow rate 3 mL/min) to afford compound 3 (6.8 mg; tR =
35.5 min). After fraction D.6 (1.3 g) was separated by MPLC with a
gradient system (5% CH3CN/H2O; 10 min/60% CH3CN/H2O; 60
min) to yield three fractions, subfraction D.6.1 (360 mg) was purified
by semipreparative HPLC (26% aqueous CH3CN, flow rate 3 mL/
min) to furnish compound 4 (2.3 mg; tR = 65.2 min). Compounds 5
(2.4 mg; tR = 44.8 min) and 6 (2.1 mg; tR = 52.0 min) were obtained
by semipreparative HPLC (26% aqueous CH3CN, flow rate 3 mL/
min) from fraction D.6.2 (460 mg). Fraction D.6.3 (490 mg) was
purified by semipreparative HPLC with a gradient system (30%
CH3CN/H2O; 8 min/45% CH3CN/H2O; 33 min) to provide
compounds 13 (6.0 mg; tR = 27.0 min) and 14 (5.0 mg; tR = 25.0
min). Fraction E (9 g) was separated using an open RP-18 column
(4.5 × 40 cm; 150 μm particle size) with a MeOH/H2O (4:6 to 10:0)
solvent system to produce 10 subfractions. Fraction E.6 (300 mg) was
further purified by semipreparative HPLC (26% aqueous CH3CN,
flow rate 3 mL/min) to obtain compound 1 (2.0 mg; tR = 28.0 min).
1α,6β,9β-Trihydroxy-5,10-bis-epi-eudesm-3-ene-9-O-[(S)-3″-hy-
droxy-3′′-methylglutaryl]-6-O-β-D-glucopyranoside (1). amorphous
powder; [α]2D5 6.1 (c 0.5, MeOH); UV (MeOH) λmax (log ε) 217
13C NMR data, Tables 1 and 2; HRESIMS m/z 559.2794 [M − H]−
(calcd for C27H43O12, 559.2760).
1β,6α-Dihydroxy-7-epi-eudesm-3-ene-1-O-[(S)-3″-hydroxy-3′′-
methylglutaryl]-6-O-β-D-glucopyranoside (3). amorphous powder;
[α]2D5 1.7 (c 0.5, MeOH); UV (MeOH) λmax (log ε) 216 (3.17) nm;
1
IR (KBr) νmax 3403, 2920, 1720, 1620 cm−1; H and 13C NMR data,
Tables 1 and 2; HRESIMS m/z 543.2820 [M − H]− (calcd for
C27H43O11, 543.2811).
1α,6β,9β-Trihydroxy-5,10-bis-epi-eudesm-3-ene-9-O-[(S)-3″-hy-
droxy-3′′-methylglutaryl]-6-O-(6′-O-acetyl)-β-D-glucopyranoside
(4). amorphous powder; [α]2D5 31.8 (c 0.3, MeOH); UV (MeOH) λmax
(log ε) 217 (3.20) nm; IR (KBr) νmax 3389, 2913, 1731, 1615, 1435,
1
1085 cm−1; H and 13C NMR data, Tables 1 and 2; HRESIMS m/z
601.2876 [M − H]− (calcd for C29H45O13, 601.2866).
1α,6β,9β-Trihydroxy-5,10-bis-epi-eudesm-3-ene-1-O-α-L-arabi-
nopyranosyl-6-O-β-D-glucopyranoside (5). colorless gum; [α]D25
−2.0 (c 0.5, MeOH); UV (MeOH) λmax (log ε) 217 (2.72) nm; IR
1
(KBr) νmax 3410, 2910, 1620 cm−1; H and 13C NMR data, Tables 1
and 2; HRESIMS m/z 593.2812 [M + HCOO]− (calcd for
C27H45O14, 593.2815).
1α,6β,9β-Trihydroxy-5,10-bis-epi-eudesm-3-ene-6-O-β-D-gluco-
pyranoside (6). amorphous powder; [α]2D5 1.7 (c 0.2, MeOH); UV
(MeOH); λmax (log ε) 217 (2.78) nm; IR (KBr) νmax 3413, 2909,
1
1615 cm−1; H and 13C NMR data, Tables 1 and 2; HRESIMS m/z
461.2382 [M + HCOO]− (calcd for C22H37O10, 461.2392).
1β,6α-Dihydroxy-7-epi-eudesm-3-ene-6-O-β-D-apiofuranosyl-
(1→6)-β-D-glucopyranoside (7). colorless gum; [α]2D5 −1.6 (c 0.5,
MeOH); UV (MeOH); λmax (log ε) 217 (2.68) nm; IR (KBr) νmax
1
(3.16) nm; IR (KBr) νmax 3395, 2925, 1725, 1625 cm−1; H and 13C
NMR data, Tables 1 and 2; HRESIMS m/z 559.2773 [M − H]−
(calcd for C27H43O12, 559.2760).
1α,6β,9β-Trihydroxy-5,10-bis-epi-eudesm-3-ene-1-O-[(S)-3″-hy-
droxy-3′′-methylglutaryl]-6-O-β-D-glucopyranoside (2). amorphous
powder; [α]2D5 13.8 (c, 0.5, MeOH); UV (MeOH) λmax (log ε) 217
(3.16) nm; IR (KBr) νmax 3399, 2926, 1720, 1455, 1083 cm−1; 1H and
1
3403, 2911, 1630 cm−1; H and 13C NMR data, Tables 1 and 2;
HRESIMS m/z 577.2850 [M + HCOO]− (calcd for C27H45O13,
577.2866).
Determination of the Absolute Configuration of Sugars in
Compounds 5 and 7. Compound 5 (0.9 mg) was hydrolyzed with 1
N HCl (1.0 mL) for 8 h at room temperature, and compound 7 (0.9
I
DOI: 10.1021/acs.jnatprod.8b00841
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
mg) was hydrolyzed by 2 N HCl (1.0 mL) at 60 °C for 3 h. For each,
after neutralization with Na2CO3 and concentration under reduced
pressure, L-cysteine methyl ester hydrochloride in anhydrous pyridine
(0.5 mg) was added, followed by heating for 1 h at 60 °C. Then,
phenyl isothiocyanate (0.1 mL) was added, and the mixture was
heated to 60 °C for 1 h. The final solution was analyzed by LC−MS
(column: an INNO C18 column, 5 μm, 120 Å, 4.6 × 250 mm; 10−
80% CH3CN/H2O, 0−30 min; detection wavelength: 203 nm; flow
rate: 1.0 mL/min). The retention times of the sugar derivatives were
established by comparison with those of the derivatives of commercial
L-arabinose (tR = 7.21 min), D-glucose (tR = 6.95 min), and D-apiose
(tR = 7.90 min), respectively. Thus it was determined to be the D-
configuration of the glucose and apiose and the L-configuration of
arabinose moieties in compounds 5 and 7.
Absolute Configuration of HMG in Compound 1. (S)-1-
Phenylethylamine (1.52 μL, 12.0 μmol), benzotriazol-1-yl-oxy-
tripyrrolidinophosphonium hexafluorophosphate (PyBOP, Sigma-
Aldrich) (4.6 mg, 9.2 μmol), Et3N (2.28 μL, 18.0 μmol), and
hydroxybenzotriazole (HOBt, Sigma-Aldrich) (1.6 mg, 12.0 μmol)
were mixed with a solution containing compound 1 (3.9 mg, 6.0
μmol) and then cooled in ice bath. The mixture was stirred at 25 °C
for 9 h, and dilute aqueous HCl was added to quench the reaction.
The mixture was partitioned with EtOAc and consequently separated
over a silica gel (CHCl3/MeOH, 19:1, 10:1, and 1:1) to yield 1a (3.0
mg).34 The ESIMS analysis showed a [M + HCOO]− peak at m/z
708. LiBH4 (2.2 mg, 67.5 μmol) was mixed with a 1a-containing
solution (3.0 mg, 4.5 μmol) and THF (0.3 mL). After 24 h of stirring
at 25 °C, the solution was mixed with 10 μL of HCl (0.2 N) and then
partitioned with EtOAc. The resulting extract was separated over silica
gel (CHCl3/MeOH, 19:1, 10:1, and 1:1) and yielded a colorless oil.
For acetylation, the sample was stirred at 25 °C for 24 h with Ac2O
(2.1 μL, 22.5 μmol) in pyridine (30 μL) and diluted using H2O. The
mixture was partitioned with EtOAc to yield 1b (0.5 mg).
(td, J = 13.6, 4.8 Hz, H-9b), 1.07 (d, J = 6.8 Hz, H3-13), 0.99 (d, J =
6.8 Hz, H3-12), 0.87 (s, H3-14); 13C NMR (CD3OD, 200 MHz) δC
137.8 (C-4), 122.5 (C-3), 100.0 (C-1′), 78.2 (C-3′), 77.9 (C-5′), 77.3
(C-1), 75.7 (C-6), 75.7 (C-2′), 72.1 (C-4′), 63.0 (C-6′), 47.3 (C-5),
41.6 (C-7), 41.3 (C-10), 33.5 (C-2), 32.1 (C-9), 26.4 (C-11), 26.0
(C-15), 24.1 (C-13), 22.9 (C-12), 22.7 (C-8), 11.4 (C-14); ESIMS
m/z 445 [M + HCOO]−.
Measurement of Cell Viability. The cell cytotoxicity was
evaluated using the MTT reagent (Sigma). In brief, HT22 cells,
immortalized mouse hippocampal neuronal cells, were maintained in
Dulbecco’s modified Eagle’s medium (DMEM) (HyClone, Logan,
UT) containing 10% fetal bovine serum (FBS) (HyClone), 100 U/
mL penicillin, and 100 μg/mL streptomycin. Then, 96-well plates
were used to seed cells (4000 cells/well) and incubated for 24 h. Test
compounds were dissolved in dimethyl sulfoxide (DMSO) (Junsei,
Japan) and diluted in serum-free medium (final DMSO concentration
0.1 to 0.2% (v/v)). After 24 h of incubation, a 2 mg/mL MTT
solution (20 μL) was added to each well, and incubation was
continued for 4 h in the dark. The supernatants were removed, and
formazan crystals were dissolved in DMSO. The absorbance was
recorded with a microplate reader (VersaMax, San Jose, CA) at 550
nm.
Cytotoxicity Assay and Fluorescence Images of Aβ42-
Transfected HT22 Cells. HT22 cells were seeded into 96-well
plates at 4000 cells/well and incubated for 2 to 3 h at 37 °C under a
5% CO2 atmosphere. The cells in each well were transfected with 0.2
μg of pEGFP-C1/Aβ42 plasmid (a gift from Professor Junsoo Park,
Yonsei University, Korea) with Lipofectamine 2000 reagent
(Invitrogen, Carlsbad, CA). After 10 h of transfection, the compounds
were dissolved in the medium and used to treat the cells. Afterward,
the cells were continually incubated for 24 h, and 20 μL of MTT
solution (2 mg/mL) was added. Similarly, HT22 cells were grown on
sterilized glass coverslips and transfected with Aβ42 plasmid for 10 h.
The cells were replaced with fresh medium containing 2% FBS with or
without test compounds. After 24 h of incubation, the cells were
washed once with phosphate-buffered saline (PBS) (Takara, Japan).
Fluorescence microscopy was applied to capture both fluorescence
and bright-field images (Olympus IX70, Olympus, Tokyo, Japan).
Flow Cytometry Analysis of Fluorescence Intensity. For
flow-cytometry analysis, the HT22 cells were seeded into six-well
plates for 2 to 3 h. After being transfected with the Aβ42 plasmid for
10 h, the cells were exposed to the test compounds using medium
supplemented with 2% FBS. The cells were then incubated for 24 h at
37 °C under a 5% CO2 atmosphere. Afterward, all cells (both dead
and live) were collected by a trypsinization method. The cells were
resuspended in PBS, and a flow cytometer (BD, FACS Calibur, San
Jose, CA) was used for analysis. For each test compound, at least
10 000 events were counted, and data were calculated using gates.
Statistical Analysis. Data were processed through variance
analysis (ANOVA) to determine the significance of differences
between each group followed by Tukey’s or Duncan’s post hoc test. A
p value <0.05 was taken as a significant difference (* p < 0.05, ** p <
0.01, and *** p < 0.001).
3(R)-5-O-Acetyl-1-[(S)-phenylethyl]-mevalonamide (1b). color-
1
less oil; H NMR (CDCl3, 600 MHz) δH 7.26−7.36 (5H, m, Ph),
5.15 (1H, q, J = 7.1 Hz, H-1′), 4.23 (2H, m, H-5), 2.41, 2.29 (each
1H, d, J = 14.8 Hz, H-2), 2.04 (3H, s, Ac), 1.85−1.87 (2H, m, H-4),
1.51 (3H, d, J = 6.9 Hz, H-2′), 1.24 (3H, s, H-6); ESIMS m/z 294.2
[M + H]+.
Computational Chemistry for ECD Calculation. Conforma-
tional analysis was performed by means of the CONFLEX software
using the MMFF94 force field. Geometry optimization (B3LYP/6-
31G(d) in vacuo) and TDDFT calculations were carried out with
Turbomole.33 Boltzmann distributions were calculated from the
B3LYP/6-31g(d) energy, and the ECD spectra were generated as the
sum of Gaussians. Ri and ΔEi represent the strengths of rotation, and
the value of σ in the equation was 0.10 eV.
A
2
2
1
1
2πσ
i
Δε(E) =
ΔE R e[−(E−ΔE ) /(2σ) ]
∑
i
i
2.297 × 10−39
i
Each major conformer of 1−4 occupied 99% of the Boltzmann
population, and the spectra of each simulation matched well to the
corresponding experimental spectra. This suggested that the absolute
configurations for the aglycones were 1S,5R,6R,7S,9S,10S (1 and 4),
1S,5R,6R,7S,9S,10R (2), and 1R,5S,6S,7R,10R (3), respectively.
Hydrolysis of 1−3, 5, and 7. Hydrolysis of compounds 1−3 and
5 were performed to confirm the configuration of C-1. Acid hydrolysis
of 1 by stirring at room temperature for 4 h in 3 mL of 1 N HCl
yielded 1c (Table S9). Acid hydrolysis of compounds 2, 5, and 7 by
stirring for 8 h in 3 mL of 1 N HCl yielded 2a, 5a, and 7a, respectively
(Tables S3, S8, and S9). Alkaline hydrolysis of compound 3 by
stirring for 24 h in 3 mL of 3 N KOH yielded 3a (Table S5).
1β,6α-Dihydroxy-7-epi-eudesm-3-ene-6-O-β-D-glucopyranoside
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.8b00841.
NMR spectra and HRESIMS (PDF)
AUTHOR INFORMATION
Corresponding Author
*Tel/Fax: +82-02-880-7872. E-mail: wkoh1@snu.ac.kr.
■
1
(3a). H NMR (CD3OD, 800 MHz) δH 5.25 (m, H-3), 4.45 (d, J =
7.6 Hz, H-1′), 4.39 (dd, J = 11.2, 4.5 Hz, H-6), 3.77 (m, H-6a′), 3.68
(dd, J = 11.9, 4.1 Hz, H-6b′), 3.49 (dd, J = 9.8, 6.7 Hz, H-1), 3.33 (m,
H-3′), 3.30 (m, H-5′), 3.26 (m, H-4′), 3.22 (m, H-2′), 2.45 (m, H-5),
2.24 (m, H-2a), 2.21 (m, H-11), 1.94 (m, H-2b), 1.91 (br s, H3-15),
1.86 (m, H-7), 1.80 (m, H-8a), 1.68 (m, H-8b), 1.57 (m, H-9a), 1.31
ORCID
Jinwoong Kim: 0000-0001-9579-738X
Won Keun Oh: 0000-0003-0761-3064
J
DOI: 10.1021/acs.jnatprod.8b00841
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(27) Oberti, J. C.; Silva, G. L.; Sosa, V. E.; Kulanthaivel, P.; Herz, W.
Notes
Phytochemistry 1986, 25, 1355−1358.
The authors declare no competing financial interest.
́
́
́
(28) Vajs, V.; Todorovic, N.; Bulatovic, V.; Menkovic, N.; Macura,
́
́
S.; Juranic, N.; Milosavljevic, S. Phytochemistry 2000, 54, 625−633.
(29) Ma, C. M.; Nakamura, N.; Hattori, M. Chem. Pharm. Bull.
2001, 49, 915−917.
ACKNOWLEDGMENTS
■
This work was supported financially in part by grants from the
Bio & Technology Development (2016M3A9A5919548) and
the Basic Science Research Program (NRF-
2017R1E1A1A01074674) through the National Research
Foundation of Korea funded by the Ministry of Science, ICT
and Planning.
(30) Walters, D.; Meurer-Grimes, B.; Rovira, I. FEMS Microbiol. Lett.
2001, 201, 255−258.
(31) Femminella, G. D.; Thayanandan, T.; Calsolaro, V.; Komici, K.;
Rengo, G.; Corbi, G.; Ferrara, N. Int. J. Mol. Sci. 2018, 19, 3702.
(32) Hong, S.; Dissing-Olesen, L.; Stevens, B. Curr. Opin. Neurobiol.
2016, 36, 128−34.
(33) Yang, J. L.; Dhodary, B.; Quy Ha, T. K.; Kim, J.; Kim, E.; Oh,
W. K. Tetrahedron 2015, 71, 4651−4658.
(34) Huh, J.; Ha, T. K. Q.; Kang, K. B.; Kim, K. H.; Oh, W. K.; Kim,
J.; Sung, S. H. J. Nat. Prod. 2017, 80, 2818−2824.
REFERENCES
■
(1) Zhou, Y.; Wang, Z. F.; Li, W.; Hong, H.; Chen, J.; Tian, Y.; Liu,
Z. Y. J. Cell. Biochem. 2018, 119, 5437−5448.
(2) Tangaro, S.; Fanizzi, A.; Amoroso, N.; Bellotti, R. Phys. Med.
2017, 38, 36−44.
(3) Singh, A.; Hasan, A.; Tiwari, S.; Pandey, L. M. CNS Neurol.
Disord.: Drug Targets 2018, 17, 571−589.
(4) Habib, A.; Sawmiller, D.; Tan, J. J. Neurosci. Res. 2017, 95, 973−
991.
(5) Palop, J. J.; Mucke, L. Nat. Neurosci. 2010, 13, 812−818.
(6) Takahashi, R. H.; Nagao, T.; Gouras, G. K. Pathol. Int. 2017, 67,
185−193.
(7) Baltes, C.; Princz-Kranz, F.; Rudin, M.; Mueggler, T. Methods
Mol. Biol. 2011, 711, 511−533.
(8) Wang, S. W.; Wang, Y. J.; Su, Y. J.; Zhou, W. W.; Yang, S. G.;
Zhang, R.; Zhao, M.; Li, Y. N.; Zhang, Z. P.; Zhan, D. W.; Liu, R. T.
NeuroToxicology 2012, 33, 482−490.
(9) Palop, J. J.; Chin, J.; Roberson, E. D.; Wang, J.; Thwin, M. T.;
Bien-Ly, N.; Yoo, J.; Ho, K. O.; Yu, G. Q.; Kreitzer, A.; Finkbeiner, S.;
Noebels, J. L.; Mucke, L. Neuron 2007, 55, 697−711.
(10) Zhang, L.; Fang, Y.; Lian, Y.; Chen, Y.; Wu, T.; Zheng, Y.;
Zong, H.; Sun, L.; Zhang, R.; Wang, Z.; Xu, Y. PLoS One 2015, 10,
No. e0122415.
(11) Tan, C. C.; Yu, J. T.; Wang, H. F.; Tan, M. S.; Meng, X. F.;
Wang, C.; Jiang, T.; Zhu, X. C.; Tan, L. J. Alzheimer's Dis. 2014, 41,
615−631.
(12) Chopra, K.; Misra, S.; Kuhad, A. Expert Opin. Pharmacother.
2011, 12, 335−350.
(13) Ambrosia  The Plant List. http://www.theplantlist.org/1.1/
browse/A/Compositae/Ambrosia/ (accessed December 13, 2018).
(14) Tamura, Y.; Hattori, M.; Konno, K.; Kono, Y.; Honda, H.;
Ono, H.; Yoshida, M. Chemoecology 2004, 14, 113−118.
(15) Jakupovic, J.; Jaensch, M.; Bohlmann, F.; Dillon, M. O.
Phytochemistry 1988, 27, 3551−3556.
(16) Eom, H. G.; Lee, D.; Lee, S.; Noh, H. J.; Hyun, J. W.; Yi, P. H.;
Kang, K. S.; Kim, K. H. J. Agric. Food Chem. 2016, 64, 7171−7178.
(17) Kamo, T.; Hirai, N.; Matsumoto, C.; Ohigashi, H.; Hirota, M.
Phytochemistry 2004, 65, 2517−2520.
(18) Hu, J. F.; Lu, Y.; Zhao, B.; Feng, X. Z.; Hamann, M. T.
Spectrosc. Lett. 2001, 34, 75−81.
́
(19) Chernenko, G. F.; Gamilov, Y. V.; Shmidt, E. N. Chem. Nat.
Compd. 1994, 30, 328−331.
(20) Zhang, H. J.; Tan, G. T.; Santarsiero, B. D.; Mesecar, A. D.;
Hung, N. V.; Cuong, N. M.; Doel Soejarto, D.; Pezzuto, J. M.; Fong,
H. H. S. J. Nat. Prod. 2003, 66, 609−615.
(21) Mahmoud, A. A. Phytochemistry 1997, 45, 1633−1638.
(22) Lee, J.; Lee, Y. M.; Lee, B. W.; Kim, J. H.; Kim, J. S. J. Nat. Prod.
2012, 75, 267−270.
(23) Tanaka, T.; Nakashima, T.; Ueda, T.; Tomii, K.; Kouno, I.
Chem. Pharm. Bull. 2007, 55, 899−901.
(24) Huang, R.; Ding, W.; Zhou, Z.; Li, Y. Biochem. Syst. Ecol. 2014,
57, 363−366.
(25) Gao, F.; Wang, H.; Mabry, T. J.; Watson, W. H.; Kashyap, R. P.
Phytochemistry 1990, 29, 551−560.
(26) Ahmed, A. A.; Spring, O.; Razek, M. H.; Hussein, N. S.; Mabry,
T. J. Phytochemistry 1995, 39, 1127−1131.
K
DOI: 10.1021/acs.jnatprod.8b00841
J. Nat. Prod. XXXX, XXX, XXX−XXX