Proanthocyanidins from spenceria ramalana and their effects on AGE formation in vitro and hyaloid-retinal vessel dilation in larval zebrafish in vivo

Article abstract of DOI:10.1021/np400442b

Three new A-type proanthocyanidins (1-3), ent-epiafzelechin- (2α→O→7,4α→8)-ent-afzelechin 3′-O-β-d- glycopyranoside (1), ent-epiafzelechin-(2α→O→7,4α→8)- ent-epiafzelechin-(2α→O→7,4α→8)-ent-afzelechin (2), and ent-epiafzelechin-(2α→O→7,4α→8)-ent-epicatechin- (2α→O→7,4α→8)-ent-afzelechin (3), and three known compounds (4-6) were isolated from the whole plant of Spenceria ramalana. The structures of the new proanthocyanidins were established by spectroscopic and chemical studies. The inhibitory effects of compounds 1-6 on the formation of advanced glycation end products were examined in vitro. Compounds 3 and 6 showed the strongest inhibition, with IC₅₀ values of 17.4 ± 0.5 and 14.1 ± 1.6 μM, respectively. The effects of these isolates on the dilation of hyaloid-retinal vessels induced by high glucose (HG) in larval zebrafish were also investigated. Compound 3 reduced the dilation of HG-induced hyaloid-retinal vessels most effectively. This compound reduced the diameters of HG-induced hyaloid-retinal vessels by about 157.7% and 164.1% at 10 and 20 μM, respectively, versus the HG-treated control group.

Full text of DOI:10.1021/np400442b

Article
pubs.acs.org/jnp
Proanthocyanidins from Spenceria ramalana and Their Effects on
AGE Formation in Vitro and Hyaloid-Retinal Vessel Dilation in Larval
Zebrafish in Vivo
Ik-Soo Lee,†,∥ Song Yi Yu,†,∥ Seung-Hyun Jung,† Yu-Ri Lee,† Yun Mi Lee,† Joo-Hwan Kim,‡ Hang Sun,§
and Jin Sook Kim*,†
†KM-Based Herbal Drug Research Group, Herbal Medicine Research Division, Korea Institute of Oriental Medicine, Daejeon
305−811, Republic of Korea
‡Department of Life Science, Gachon University, Seongnam, Gyeonggi-do 461-701, Republic of Korea
§Laboratory of Biodiversity and Biogeography, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming, Yunnan
650204, People’s Republic of China
S
* Supporting Information
ABSTRACT: Three new A-type proanthocyanidins (1−3),
ent-epiafzelechin-(2α→O→7,4α→8)-ent-afzelechin 3′-O-β-D-
glycopyranoside (1), ent-epiafzelechin-(2α→O→7,4α→8)-
ent-epiafzelechin-(2α→O→7,4α→8)-ent-afzelechin (2), and
ent-epiafzelechin-(2α→O→7,4α→8)-ent-epicatechin-(2α→
O→7,4α→8)-ent-afzelechin (3), and three known compounds
(4−6) were isolated from the whole plant of Spenceria
ramalana. The structures of the new proanthocyanidins were
established by spectroscopic and chemical studies. The
inhibitory effects of compounds 1−6 on the formation of
advanced glycation end products were examined in vitro.
Compounds 3 and 6 showed the strongest inhibition, with
IC50 values of 17.4 0.5 and 14.1 1.6 μM, respectively. The effects of these isolates on the dilation of hyaloid-retinal vessels
induced by high glucose (HG) in larval zebrafish were also investigated. Compound 3 reduced the dilation of HG-induced
hyaloid-retinal vessels most effectively. This compound reduced the diameters of HG-induced hyaloid-retinal vessels by about
157.7% and 164.1% at 10 and 20 μM, respectively, versus the HG-treated control group.
Spenceria ramalana Trimen, a species in the genus Spenceria,
iabetic vascular complications are leading causes of end-
Dstage renal failure, acquired blindness, various neuro- belonging to the family Rosaceae, is a perennial herb, native to
China, distributed in a large cross-section of Shanyu Tibet,
Yunnan, and Sichuan. It has been used in traditional folk
medicine in China for the treatment of various ailments,
including lumbago, stomachache, and dysentery.6 To date,
there has been no study of the chemical constituents or
biological activities of this plant. In searching for new AGE
inhibitors from natural sources, we found that the EtOAc-
soluble fraction of the 80% EtOH extract of the whole plant of
S. ramalana had a considerable inhibitory effect on AGE
formation (IC50 = 19.92 0.13 μg/mL). Further phytochem-
ical study of this fraction resulted in the isolation of three new
(1−3) and three known dimeric A-type proanthocyanidins (4−
6). In this report, we describe the isolation and structural
elucidation of these compounds, as well as the characterization
of their inhibitory effects on AGE formation. The effects of
these isolates on the dilation of high-glucose (HG)-induced
hyaloid-retinal vessels in larval zebrafish were also investigated.
pathies, and accelerated atherosclerosis. Chronic hyperglycemia
plays a major role in the initiation of diabetic vascular
complications through various hyperglycemia-induced meta-
bolic and hemodynamic derangements, including increased
advanced glycation end-product (AGE) formation, increases in
the polyol pathway, activation of protein kinase C isomers, and
increases in the hexosamine pathway.1−4 Of these factors
implicated in the pathogenesis of diabetic vascular complica-
tions, enhanced formation and accumulation of AGE,
heterogeneous molecules derived from nonenzymatic glycation
between amino acid residues and oxidative derivatives of
glucose or pentose, is irreversible, causing structural and
functional changes in proteins, such as collagen, elastin, and
albumin, leading to the development of diabetic vascular
complications, including diabetic retinopathy, neuropathy, and
nephropathy.5 Because of the potential role of AGE in diabetic
vascular complications, the development of pharmacological
compounds that inhibit the formation of these glycated
products may provide a therapeutic approach for delaying
and preventing diabetic vascular complications.
Received: June 4, 2013
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
dx.doi.org/10.1021/np400442b | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
6.09), and two pairs of A2B2 aromatic system protons [δH 7.51
and 6.83 (each d, J = 9.0 Hz) and δH 7.34 and 6.84 (each d, J =
8.4 Hz)]. Additionally, an isolated AB system [δH 4.15 and 4.31
(each d, J = 3.6 Hz)] in the heterocyclic proton region was
observed, characteristic of the C-ring protons of A-type
proanthocyanidins, consisting of two flavanyl units.11 Such a
structural type was supported by the typical acetal carbon
resonance at δC 100.7 in the 13C NMR spectrum (Table 1).
The interflavonoid linkage between the two flavanyl units was
established by methylation of 1 with diazomethane, yielding
pentamethyl ether (1a). The 1H NMR spectrum of 1a
exhibited, together with four methoxy signals at δH 3.72−
3.82, one upfield methoxy signal at δH 3.31 assignable to MeO-
5 due to the anisotropic shielding of the E-ring in the lower
flavanyl unit.12,13 This suggests that the two flavanyl units were
linked through the C-4 and C-8′ positions. The existence of a
(2→O→7′) rather than a (2→O→5′) ether linkage was
deduced from the NOE correlation between a methoxy signal
at δH 3.75 (3H, s, MeO-5′) and H-4′ methylene signals (δH
2.72−2.92, F-ring) in 1a. The (4→8′, 2→O→7′) interflavonoid
linkage of 1 was further supported by observation of the
HMBC cross-peaks of H-3 with C-8′ and of H-4 with C-7′, C-
8′, and C-9′ (Figure 1). The absolute configuration at C-4 on
the heterocyclic C-ring was determined from the sign of the
Cotton effect in the region 220−240 nm.14 A negative Cotton
effect around 230 nm ([Θ]232 −9230) in the electronic circular
dichroism (ECD) spectrum of 1 leads to assignment of a 4S-
configuration of the C-ring, thus confirming the (2α,4α)-
configuration of the C-ring because the linkage of the two units
of A-type proanthocyanidins must be cis.15 The NOESY cross-
peak between H-3 (δH 4.15, C-ring) and H-6′ (δH 5.96, D-ring)
indicated the 3,4-trans configuration of the C-ring (Figure 2).
On the basis of these data, the absolute configuration of the C-
ring was deduced to be 2R,3S,4S, and thus the upper flavanyl
unit of 1 was identified as ent-epiafzelechin. The absolute
configuration at C-2′ of the F-ring was established from the
ECD spectrum. Flavan-3-ols with 2R and 2S absolute
configuration give rise to negative and positive Cotton effects,
respectively, in the 260−280 nm region of their ECD spectra.16
Thus, the weak positive Cotton effect around 270 nm ([Θ]271
+580) observed in 1 indicated that the absolute configuration at
C-2 of the F-ring differed from that of the C-ring, indicating a
2S-configuration of the F-ring.7,15 The large coupling constant
(J = 8.0 Hz) between H-2′ and H-3′ of the F-ring reflected a
relative trans-configuration.17 Consequently, the absolute
configurations at C-2 and C-3 of the F-ring were deduced to
be S and R, respectively, and the lower flavanyl unit of 1 was
thus identified as ent-afzelechin. The carbon signals at δC 103.9,
78.2, 77.8, 75.4, 71.7, and 62.9 and an anomeric proton signal at
δH 4.12 in 1 were typical of a glucopyranosyl unit. Acid
hydrolysis of 1 yielded 4 as the aglycone and a monosaccharide
unit identified as D-glucose upon GC analysis. When the 13C
NMR spectrum of 1 was compared with that of 4, a downfield
shift was observed for C-3′ of the F-ring by about 7 ppm,
indicating the linkage position of a glucopyranosyl moiety at C-
3′ of the lower flavanyl unit, further supported by the HMBC
cross-peak between the anomeric proton (δH 4.12) and the
aglycone carbon (δC 75.3) for C-3 of the F-ring (Figure 1).
Moreover, the large coupling constant (J = 7.5 Hz) of the
anomeric proton indicated that the glucopyranosyl unit was
linked in a β-configuration. Thus, the structure of 1 was
determined as ent-epiafzelechin-(2α→O→7,4α→8)-ent-afzele-
chin 3′-O-β-D-glycopyranoside.
RESULTS AND DISCUSSION
■
An 80% (v/v) EtOH extract of the whole S. ramalana plant was
suspended in H2O and successively partitioned using n-hexane,
EtOAc, and n-BuOH. The EtOAc-soluble fraction was
subjected to a series of chromatographic steps, resulting in
the isolation of six proanthocyanidins (1−6). Of these, three
were identified as ent-epiafzelechin-(2α→O→7,4α→8)-ent-
afzelechin (4),7 ent-epiafzelechin-(2α→O→7,4α→8)-afzelechin
(5),8 and ent-epiafzelechin-(2α→O→7,4α→8)-catechin (6),9
by comparison of observed and published physicochemical
data.
Compound 1 was obtained as a white powder, with the
molecular formula C36H34O15, as established by HRESIMS,
based on the [M + H]+ peak (m/z 707.1970). The IR spectrum
exhibited absorption bands for hydroxy (3450 cm−1) and
aromatic (1610 and 1520 cm−1) functional groups. A positive
vanillin−sulfuric acid test and a UV absorption maxima
observed at 272 nm suggested the presence of a flavan
structure in 1.10 The 1H NMR spectrum (Table 1) of 1
exhibited characteristic peaks for two meta-coupled doublets
[δH 6.07 and 5.96 (each d, J = 2.2 Hz)], an aromatic singlet (δH
B
dx.doi.org/10.1021/np400442b | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Table 1. NMR Data for Compounds 1−3 in Methanol-d4 (500 MHz for H NMR, 125 MHz for 13C NMR)
Article
1
1
2
3
C(ring)
δH (J in Hz)
4.15 d (3.6)
δC
δH (J in Hz)
4.28 d (3.5)
δC
δH (J in Hz)
4.28 d (3.5)
δC
2(C)
3(C)
4(C)
5(A)
100.7
67.8
101.5
67.3
101.5
68.2
4.31 d (3.6)
5.96 d (2.2)
6.07 d (2.2)
29.4
4.37 d (3.5)
5.92 d (2.2)
6.04 d (2.2)
29.8
4.38 d (3.5)
5.93 d (2.2)
6.04 d (2.2)
29.8
156.8
98.3
157.0
98.6
156.9
98.6
6(A)
7(A)
158.4
96.8
158.5
96.8
158.5
96.8
8(A)
9(A)
154.3
104.2
131.8
129.7
115.7
159.1
115.7
129.7
81.4
154.2
103.6
131.0
129.7
116.0
159.5
116.0
129.7
100.8
67.6
154.2
103.6
131.0
129.7
116.0
159.5
116.0
129.7
100.8
67.6
10(A)
11(B)
12(B)
13(B)
14(B)
15(B)
16(B)
2′(F)
3′(F)
4′(F)
7.51 d (9.0)
6.83 d (9.0)
7.67 d (9.0)
6.90 d (9.0)
7.67 d (9.0)
6.91 d (9.0)
6.83 d (9.0)
7.51 d (9.0)
5.15 d (8.0)
6.90 d (9.0)
7.67 d (9.0)
6.91 d (9.0)
7.67 d (9.0)
4.38 ddd (5.6, 8.0, 8.8)
75.3
4.18 d (3.5)
4.36 d (3.5)
4.18 d (3.5)
4.36 d (3.5)
a
ax: 2.78 dd (8.8, 16.4)
26.9
29.6
29.6
b
eq: 2.87 dd (5.6, 16.4)
5′(D)
6′(D)
7′(D)
8′(D)
9′(D)
10′(D)
11′(E)
12′(E)
13′(E)
14′(E)
15′(E)
16′(E)
2″(I)
156.2
96.9
158.5
96.9
158.4
96.9
6.09 s
6.11 s
6.11 s
152.5
106.8
150.6
102.9
130.4
129.5
116.6
159.0
116.6
129.5
154.0
108.7
147.1
103.4
131.6
129.7
115.7
159.2
115.7
129.7
84.0
154.0
108.6
147.1
103.5
131.6
115.5
146.9
146.6
116.5
120.3
83.9
7.34 d (8.5)
6.84 d (8.5)
7.52 d (9.0)
6.84 d (9.0)
6.95 d (2.1)
6.84 d (8.5)
7.34 d (8.5)
6.84 d (9.0)
7.52 d (9.0)
4.78 d (8.0)
6.83 d (6.8)
6.84 dd (2.1, 6.8)
4.78 d (8.0)
3″(I)
4″(I)
4.12 ddd (5.5, 8.0, 9.0)
eq: 2.60 dd (9.0, 16.5)
ax: 3.06 dd (5.5, 16.5)
68.3
4.12 ddd (5.5, 8.0, 9.0)
eq: 2.60 dd (9.0, 16.5)
ax: 3.06 dd (5.5, 16.5)
67.3
29.5
29.5
5″(G)
6″(G)
7″(G)
8″(G)
9″(G)
10″(G)
11″(H)
12″(H)
13″(H)
14″(H)
15″(H)
16″(H)
Glc-1
157.1
98.6
157.0
98.6
5.91 s
5.91 s
150.2
108.5
149.8
103.9
130.3
130.1
116.5
159.1
116.5
130.1
150.2
108.5
149.6
103.7
131.0
129.7
115.7
159.2
115.7
129.7
7.35 d (8.5)
6.86 d (8.5)
7.52 d (8.5)
6.84 d (8.5)
6.86 d (8.5)
7.35 d (8.5)
6.84 d (8.5)
7.52 d (8.5)
4.12 d (7.5)
3.10
103.9
75.4
78.2
71.7
77.8
62.9
Glc-2
Glc-3
3.18
Glc-4
3.22
Glc-5
3.23
Glc-6
3.86, 3.64
a
b
Axial proton. Equatorial proton.
1
Compound 2 was obtained as a white powder and gave a
suggesting the presence of a triflavonoid structure. The H
NMR spectrum (Table 1) was similar to those of 4 and 5,
except that 2 had additional signals consistent with the
presence of a flavanyl unit [δH 7.35 and 6.86 (each d, J = 8.5
positive result in the vanillin−sulfuric acid test. Its molecular
formula, C45H34O15, was determined on the basis of its [M−
H]− peak (m/z 813.1826) in the HRESIMS and NMR data,
C
dx.doi.org/10.1021/np400442b | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 1. Key HMBC (H→C) correlations of compounds 1−3.
Figure 2. Key NOE (H↔H) correlations of compounds 1−3.
Hz) and 5.91 (s) in the aromatic region; δH 4.78 (d, J = 8.0
Hz), 4.12 (ddd, J = 5.5, 8.0, 9.0 Hz), 3.06 (dd, J = 5.5, 16.5 Hz),
and 2.60 (dd, J = 9.0, 16.5 Hz) in the heterocyclic region],
suggesting that 2 is a trimeric A-type proanthocyanidin.18 The
two doubly linked units were also demonstrated by the typical
acetal carbon resonances at δC 101.5 and 100.8 in the 13C NMR
spectrum (Table 1). The methylation of 2 with diazomethane
between H-3 (δH 4.15, C-ring) and H-6′ (δH 6.11, D-ring) and
between H-3′ (δH 4.15, F-ring) and H-6″ (δH 5.91, G-ring)
indicated the 3,4-trans configurations of the C- and F-rings
(Figure 2). Thus, both the upper and middle flavanyl units of 2
were identified as ent-epiafzelechin. The absolute configuration
at C-2‴ of the I-ring was determined from the sign of the
Cotton effect observed in the 260−280 nm region of the ECD
spectrum of 2. Compared to 1, the sign of the Cotton effect
around 270 nm was identical, but the amplitude in 2 ([Θ]271
+820) was larger than that in 1, [Θ]271 +580. This indicates a
2S-configuration of the I-ring of 2 because the absolute
configurations at C-2 of the C-ring and C-2′ of the F-ring of 2
are identical to those of 1. Because H-2″ and H-3″ of the I-ring
were trans-oriented (J2,3 = 8.0 Hz), the lower flavanyl unit was
characterized as ent-afzelechin. Thus, the structure of 2 was
determined to be ent-epiafzelechin-(2α→O→7,4α→8)-ent-
epiafzelechin-(2α→O→7,4α→8)]-ent-afzelechin.
Compound 3 gave a positive result in the vanillin−sulfuric
acid test and showed a molecular formula of C45H34O16, from
its [M − H]− peak (m/z 829.1771) in the HRESIMS, which
was assumed to be a trimeric flavonoid.17 The 1H NMR
spectrum (Table 1) displayed two isolated AB systems [δH
4.28, 4.38 and 4.18, 4.36 (each d, J = 3.5 Hz)] in the
heterocyclic proton region, ascribed as a diagnostic feature of
the C- and F-ring protons of A-type proanthocyanidins,15
indicating the presence of two doubly linked flavanyl units. The
location of the interflavanyl linkage of 3 was established by
analysis of the 1H NMR spectrum and NOE correlations of the
corresponding octamethyl ether (3a). The two upfield methoxy
signals (δH 3.31 and 3.38), assignable to the methoxy groups at
C-5 (A-ring) and C-5′ (D-ring), and the NOE correlations of a
methoxy signal at δH 3.38 (3H, s, MeO-5′) with H-4′ (δH 4.30,
1
yielded heptamethyl ether (2a). The H NMR spectrum of 2a
exhibited two upfield methoxy signals (δH 3.33 and 3.40),
assignable to the methoxy groups at C-5 (A-ring) and C-5′ (D-
ring), together with five methoxy signals at δH 3.70−3.83. The
unusual upfield shifts of the two methoxy groups indicated the
presence of a through-space interaction with the E- and H-
rings,12,13 suggesting that the two interflavonoid C−C linkages
were (4→8′) and (4′→8″). The NOESY cross-peaks between a
methoxy signal at δH 3.40 (3H, s, MeO-5′) and H-4′ (δH 4.30,
F-ring) and between a methoxy signal at δH 3.70 (3H, s, MeO-
5″) and H-4″ (δH 2.58−3.10, I-ring) of 2a indicated the
presence of (2→O→7′) and (2′→O→7″) ether linkages. The
unsubstituted phloroglucinol A-ring carbons C-6 (δC 98.6) and
C-8 (δC 96.8), C-6 (δC 96.9) of the D-ring, and C-6 (δC 98.6)
of the G-ring in 2, being similar to those of geranin D isolated
from Geranium niveum,19 further supported that the inter-
flavonoid linkage between the three flavanyl units was (4→8,
2→O→7′) and (4′→8″, 2′→O→7″). Although the 1H and 13
C
NMR spectra of 2 were found to be identical to those of
geranin D, 2 showed a Cotton effect with opposite sign around
240 nm ([Θ]240 −5820) relative to that of geranin D ([Θ]236
+7760), indicating that the C- and F-rings had 4S-
configurations in 2. The absolute configurations at C-2 of the
C-ring and C-2′ of the F-ring were thus automatically assigned
as R and S, respectively. Furthermore, the NOESY cross-peaks
D
dx.doi.org/10.1021/np400442b | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
F-ring) and a methoxy signal at δH 3.78 (3H, s, MeO-5″) with
H-4″ (δH 2.60−3.11, I-ring) indicated, as for 2, the presence of
(4→8, 2→O→7′) and (4′→8″, 2′→O→7″) interflavanyl
linkages in 3. This was confirmed by the unsubstituted
phloroglucinol A-ring carbons C-6 (δC 98.6) and C-8 (δC
96.8), C-6 (δC 96.9) of the D-ring, and C-6 (δC 98.6) of the G-
ring in 3. The 1H NMR features of 3 were similar to those of 2,
but there were marked differences between 3 and 2 in the
chemical shifts and splitting patterns of the aromatic protons of
the E-ring; an AMX system [δH 6.83 (d, J = 6.8 Hz), 6.84 (dd, J
= 2.1, 6.8 Hz), and 6.95 (dd, J = 2.1 Hz)] was observed in 3,
instead of an A2B2 system in 2. Thus, 3 differed from 2 in the
substitution pattern of the E-ring, substantiated by the 13C
NMR chemical shifts of C-4′ (δC 146.6) and C-3′ (δC 146.9) of
the E-ring, as well as analysis of the HMBC spectrum (Figure
1). The absolute configuration of 3 was assigned as being
identical to that of 2 on the basis of the Cotton effects and
NOE correlations (Figure 2). Thus, the upper and middle
flavanyl units of 3 were identified as ent-epiafzelechin and ent-
epicatechin, respectively, and the lower flavanyl unit was
characterized as ent-afzelechin, due to a positive Cotton effect
around 272 nm ([Θ]272 +780) in the ECD spectrum and the
trans relationship (J2,3 = 8.0 Hz) between H-2 and H-3 of the I-
ring. Consequently, the structure of 3 was assigned as ent-
epiafzelechin-(2α→O→7,4α→8)-ent-epicatechin-(2α→O→
7,4α→8)-ent-afzelechin.
unit(s) in flavonoids is important to show substantial
antioxidant activity, our result is consistent with previous
reports that the AGE formation inhibitory activities of several
flavonoids were closely associated with their antioxidant
capacity.22,23
Accumulation of AGE in retinal blood vessels plays an
important role in the onset and development of diabetic
retinopathy.24 Experimental studies have shown that specific
inhibitors of AGE formation effectively reduce retinal micro-
vascular lesions in diabetic animal models.25,26 Recently, it has
been reported that the retinal vessels of the hyperglycemic adult
zebrafish increase in diameter (compared to normal animals).27
Thus, to evaluate the effects of compounds 1−6 on retinopathy,
an in vivo diabetic zebrafish model was used. The change in the
extent of hyaloid-retinal vessel dilation was assessed using
zebrafish flk:EGFP transgenic embryos held under high-glucose
(30 mM) conditions. Dilation was reduced by treatment with
compounds 1−6; compound 3, which contains a catechol unit,
was most effective in this regard. Vessels treated with 3 were
significantly thinner than were those of the control group
(Figure 3B−D), and no cytotoxicity was evident at the
concentrations tested. Quantitative analysis revealed that 3
reduced the diameters of HG-induced hyaloid-retinal vessels by
157.7% and 164.1% when used at 10 and 20 μM, respectively,
compared to the HG-treated control group (Figure 3F).
Compound 6 also reduced HG-induced dilation of hyaloid-
retinal vessels, albeit to a lesser extent (104.0% inhibition at 20
μM) than did 3 (see Supporting Information). Our results
suggest that compound 3 inhibits development of experimental
retinopathy at early larval stages.
Proanthocyanidins represent a unique class of oligomeric and
polymeric end-products of the flavonoid biosynthetic pathway.
The materials are widespread throughout the plant kingdom
and exhibit a number of biological properties, including
antioxidant, antibacterial, antiviral, anticarcinogenic, anti-
inflammatory, and antiallergic activities.28−30 Our current
study of novel AGE inhibitors from S. ramalana resulted in
isolation of three new (1−3) and three known dimeric A-type
proanthocyanidins (4−6). This is the first report of the
occurrence of A-type proanthocyanidins in the genus Spenceria.
Compounds 1−6 notably inhibited AGE formation with IC50
values ranging from 14.1 to 39.8 μM. Compound 3 also
significantly reduced dilation of HG-induced hyaloid retinal
vessels in a diabetic zebrafish model. Our results suggest that S.
ramalana and active components thereof may be beneficial to
treat and prevent diabetic vascular complications and other
related diseases.
The inhibitory effects of the isolated compounds (1−6) on
AGE formation were examined in vitro according to a
previously described procedure20 using commercially available
aminoguanidine, the first AGE inhibitor for the treatment of
diabetic nephropathy,21 as a positive control. As shown in Table
2, all of the tested compounds exhibited considerable inhibition
Table 2. Inhibitory Effects of Compounds 1−6 from S.
a
ramalana against AGE Formation
b
compound
IC50 (μM)
1
2
3
4
5
6
32.1 0.5
39.8 3.6
17.4 0.5
34.7 2.8
33.5 2.0
14.1 1.6
c
aminoguanidine
965.9 15.5
a
b
Results are expressed as means SEM (n = 3). IC50 indicates the
concentration (μM) at which the inhibition percentage of the AGE
formation was 50%, and the values were determined by regression
c
analysis. Positive control.
EXPERIMENTAL SECTION
■
of AGE formation, with IC50 values ranging from 14.1 to 39.8
μM, compared with that of aminoguanidine (IC50 = 965.9
15.5 μM). Of the tested compounds, compounds 3 and 6, both
of which have a catechol unit in the molecule, were the
strongest inhibitors of AGE, with IC50 values of 17.4 0.5 and
General Experimental Procedures. Optical rotations were
measured on a JASCO P-2000 digital polarimeter. ECD spectra
were measured on a JASCO J-715 spectrometer. IR spectra were
recorded on a JASCO 100 IR spectrometer. All 1D (1H and 13C) and
2D (HMQC, HMBC, and NOESY) NMR spectra were obtained using
a Bruker Avance 500 NMR spectrometer with TMS as an internal
standard. HRESIMS were recorded on a Shimadzu LCMS-IT-TOF
spectrometer. Column chromatography was performed using silica gel
(70−230 mesh and 230−400 mesh, Merck) and Sephadex LH-20
(Amersham Pharmacia Biotech). Thin-layer chromatography (TLC)
was performed on precoated silica gel 60 F254 (0.25 mm, Merck) and
RP-18 F254s plates (0.25 mm, Merck). Spots were detected by UV light
(254 nm) and spraying with 10% H2SO4 followed by heating.
14.1
1.6 μM, respectively, and their activities were
approximately twice those of the other isolates (1, 2, 4, and
5) that do not contain a catechol unit, suggesting that the
presence of catechol unit(s) in these A-type proanthocyanidins
increases their AGE-inhibitory properties. We additionally
found that epicatechin and epicatechin-3-O-gallate, commer-
cially available compounds containing catechol units, notably
inhibited AGE formation at IC50 values of 13.6 0.3 and 15.1
0.2 μM, respectively. Because the presence of catechol
Plant Material. The whole plant of S. ramalana was collected from
an alpine meadow in Lijiang, China, in September 2008, and identified
E
dx.doi.org/10.1021/np400442b | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 3. Effect of compound 3 on dilation of hyaloid-retinal vessels from a high-glucose (HG)-induced diabetic retinopathy model. (A−D)
Hyaloid-retinal vessels of flk:EGFP transgenic zebrafish. (A) Untreated normal group (Nor). (B) HG-treated control group (HG). (C) 10 μM and
(D) 20 μM compound 3 in the HG-treated group. (E) Data are displayed as International System of Units (SI) for vessel diameters. (F) Percentage
inhibition (10 μM = 157.7%, 20 μM = 164.1%) of hyaloid-retinal vessel dilation of compound 3 in HG-induced embryos. The diameters of hyaloid-
retinal vessels were measured at locations proximal to the optic disc (yellow dotted circle). The hyaloid vessel diameter of each lens was measured
three times, and the experiment was performed in triplicate. ###p < 0.001 vs Nor, ***p < 0.001 vs HG.
by one of the authors (J.-H.K.). A voucher specimen (No. DiAB-2008-
063) has been deposited in the Herbarium of the Diabetic
Complications Research Team, Korea Institute of Oriental Medicine,
Republic of Korea.
Extraction and Isolation. The dried whole plant of S. ramalana
(4.0 kg) was extracted with EtOH (3 × 60 L) at room temperature for
7 days with maceration and filtering and then concentrated to give an
EtOH extract (330 g). The EtOH extract (320 g) was suspended in
H2O (4 L) and partitioned successively with n-hexane (3 × 4.0 L),
EtOAc (3 × 4.0 L), and n-BuOH (3 × 4.0 L) to afford n-hexane- (13.8
g), EtOAc- (198.1 g), and n-BuOH-soluble fractions (100.5 g),
respectively.
The EtOAc-soluble fraction (180.0 g), which significantly inhibited
AGE formation, was subjected to silica gel column chromatography
(70−230 mesh, 40 × 9.5 cm), eluting with a gradient solvent system
consisting of CHCl3−MeOH (100:1 → 0:1) to afford nine fractions
(A−I). Fraction B (18.8 g) was chromatographed on a silica gel
column (70−230 mesh, 50 × 10 cm) using a stepwise gradient of
CHCl3−MeOH (50:1 → 5:1) to yield nine fractions (B1−B9), one of
which, B7 (2.1 g), was further chromatographed on a silica gel column
(230−400 mesh, 40 × 7.5 cm), eluted with a CHCl3−MeOH gradient
(50:1 → 5:1), yielding seven subfractions (B2.1−B2.7). Subfractions
B2.2 (700 mg) and B2.4 (600 mg) were separately chromatographed
on a Sephadex LH-20 column (60 × 2.5 cm); elution of B2.2 with a
MeOH−H2O gradient (6:4 → 7:3) yielded compounds 4 (110 mg)
and 5 (7 mg), and elution of B.2.4 with MeOH−H2O (7:3) yielded
compound 6 (145 mg). Chromatography of fraction D (21.1 g) on a
silica gel column (70−230 mesh, 50 × 10 cm), eluting with a stepwise
gradient of CHCl3−MeOH (50:1 → 5:1), afforded six fractions (D1−
D6). Fraction D2 (600 mg) was further purified on a Sephadex LH-20
column (60 × 2.5 cm) using a MeOH−H2O gradient (65:35 →
70:30) to obtain compound 1 (25 mg). Fraction D3 (2.6 g) was
applied to a silica gel column (230−400 mesh, 40 × 7.5 cm) and
eluted using a CHCl3−MeOH gradient (50:1 → 10:1) to yield eight
subfractions (D3.1−D3.8). Compounds 2 (21 mg) and 3 (16 mg)
were isolated from fraction D3.8 (350 mg) using a Sephadex LH-20
column (60 × 2.5 cm) eluting with a MeOH−H2O gradient (60:40 →
70:30).
ent-Epiafzelechin-(2α→O→7,4α→8)-ent-afzelechin 3′-O-β-D-gly-
copyranoside (1): white powder; [α]2D5 −101 (c 0.1, MeOH); UV
(MeOH) λmax 224, 272 nm; ECD (c 0.0002 M, MeOH) [Θ] (nm)
−9230 (232), +580 (271); IR (KBr) νmax 3450, 1610, 1520, 1140, 830
cm−1; 1H and 13C NMR, see Table 1; HRESIMS m/z 707.1970 [M +
H]+ (calcd for C36H35O15+, 707.1970).
ent-Epiafzelechin-(2α→O→7,4α→8)-ent-epiafzelechin-(2α→O→
7,4α→8)]-ent-afzelechin (2): white powder; [α]2D5 −101 (c 0.13,
MeOH); UV (MeOH) λmax 226, 272 nm; ECD (c 0.0002 M, MeOH)
[Θ] (nm) −5820 (240), +447 (254), +820 (271); IR (KBr) νmax 3300
cm−1; 1H and 13C NMR, see Table 1; HRESIMS m/z 813.1826 [M −
H]− (calcd for C45H33O15−, 813.1825).
ent-Epiafzelechin-(2α→O→7,4α→8)-ent-epicatechin-(2α→O→
7,4α→8)-ent-afzelechin (3): white powder; [α]2D5 −90 (c 0.1, MeOH);
UV (MeOH) λmax 224, 274 nm; ECD (c 0.0002 M, MeOH) [Θ] (nm)
1
−5635 (240), +371 (254), +780 (272); H and 13C NMR, see Table
−
1; HRESIMS m/z 829.1771 [M − H]− (calcd for C45H33O16
829.1774).
,
Methylation of 1−3. The methylation of 1 (10 mg) was done
with an excess of CH2N2 in MeOH−Et2O at room temperature for 48
h. After a routine workup, the product was purified by HPLC [Agilent
1200 system; YMC-pack Pro C18 column (250 × 10 mm, i.d.);
MeOH−H2O (30:70, v/v); UV detection, 270 nm; flow rate, 3.0 mL/
min] to give pentamethyl ether (1a, 6 mg). The methylation of 2 (7
mg) and 3 (10 mg) was similarly performed to afford heptamethyl
ether (2a, 3 mg) and octamethyl ether (3a, 2.5 mg).
1a: white powder; 1H NMR (300 MHz, CDCl3) δH 2.72−2.92 (2H,
m, H-4′), 3.31 (3H, s, MeO-5), 3.72 (3H, s, MeO-7), 3.75 (3H, s,
MeO-5′), 3.80 (3H, s, MeO-14), 3.82 (3H, s, MeO-14′), 4.14 (1H, d, J
= 3.5 Hz, H-3), 4.25 (1H, d, J = 3.5 Hz, H-4), 5.02 (1H, d, J = 8.0 Hz,
H-2′), 6.10 (1H, d, J = 2.2 Hz, H-6), 6.20 (1H, s, H-6′), 6.24 (1H, d, J
= 2.2 Hz, H-8), 6.94−7.60 (8H, aromatic protons in rings B and E),
4.13 (1H, d, J = 7.5 Hz, glc-1), 3.12−3.85 (6H, m, glc-2, 3, 4, 5, 6);
ESIMS m/z 777.2 [M + H]+.
F
dx.doi.org/10.1021/np400442b | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
2a: white powder; 1H NMR (300 MHz, CDCl3) δH 2.58−3.10 (2H,
m, H-4″), 3.33 (3H, s, MeO-5), 3.40 (3H, s, MeO-5′), 3.70 (3H, s,
MeO-5″), 3.73 (3H, s, MeO-7), 3.78 (3H, s, MeO-14″), 3.81 (3H, s,
MeO-14′), 3.83 (3H, s, MeO-14), 4.04 (1H, m, H-3″), 4.12 (1H, d, J
= 3.5 Hz, H-3′), 4.25 (1H, d, J = 3.5 Hz, H-3), 4.30 (1H, d, J = 3.5 Hz,
H-4 or H-4′), 4.32 (1H, d, J = 3.5 Hz, H-4 or H-4′), 4.70 (1H, d, J =
8.0 Hz, H-2″), 5.98 (1H, d, J = 2.2 Hz, H-6″), 6.08 (1H, d, J = 2.2 Hz,
H-6), 6.18 (1H, d, J = 2.2 Hz, H-6′), 6.20 (1H, d, J = 2.2 Hz, H-8),
6.92−7.70 (12H, aromatic protons in rings B, E, and H); ESIMS m/z
911.2 [M − H]−.
Statistical Analysis. Hyaloid vessel diameter was used to calculate
the percentage inhibition of compound 3 in HG-treated embryos
according to the formula
percentage inhibition (%)
= [1 − (T − Cav)/(HGav − Cav)] × 100
n
where Cav = average of hyaloid-vessel diameter (SI) in the control
embryos, HGav = average of hyaloid-vessel diameter (SI) in the HG-
treated embryos, and Tn = hyaloid-vessel diameter (SI) of the HG-
treated embryos with compound 3. The results are expressed as means
standard error of the mean (SEM) from three independent
experiments. Statistical significance was assessed using one-way
analysis of variance (ANOVA) and Dunnett’s multiple comparison
tests with the GraphPad 5.0 Prism software (GraphPad, San Diego,
CA, USA).
1
3a: white powder; H NMR (300 MHz, CDCl3) δH 2.60−3.11
(2H, m, H-4″), 3.31 (3H, s, MeO-5), 3.38 (3H, s, MeO-5′), 3.68 (3H,
s, MeO-13′), 3.74 (3H, s, MeO-14′), 3.78 (3H, s, MeO-5″), 3.83 (3H,
s, MeO-7), 3.87 (3H, s, MeO-14″), 3.90 (3H, s, MeO-14), 4.03 (1H,
m, H-3″), 4.14 (1H, d, J = 3.5 Hz, H-3′), 4.26 (1H, d, J = 3.5 Hz, H-3),
4.30 (1H, d, J = 3.5 Hz, H-4′), 4.34 (1H, d, J = 3.5 Hz, H-4), 4.70 (1H,
d, J = 8.0 Hz, H-2″), 5.96 (1H, d, J = 2.2 Hz, H-6″), 6.07 (1H, d, J =
2.2 Hz, H-6), 6.19 (1H, d, J = 2.2 Hz, H-6′), 6.20 (1H, d, J = 2.2 Hz,
H-8), 6.90−7.72 (11H, aromatic protons in rings B, E, and H); ESIMS
ASSOCIATED CONTENT
■
S
* Supporting Information
m/z 941.3 [M − H]−
.
1H NMR, 13C NMR, NOESY, and mass spectra for the new
compounds 1−3 and zebrafish data of compound 6. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Acid Hydrolysis of 1. Compound 1 (5 mg) in 10% HCl−dioxane
(1:1, 1 mL) was heated at 80 °C for 3 h in a water bath. The mixture
was neutralized with Ag2CO3, filtered, and extracted with EtOAc (20
mL). The EtOAc layer was evaporated, and the residue subjected to
reversed-phase HPLC [Agilent 1200 system; YMC-pack Pro C18
column (250 × 10 mm, i.d.); MeOH−H2O (30:70, v/v); UV
detection, 270 nm; flow rate, 3.0 mL/min] to give ent-epiafzelechin-
(2α→O→7,4α→8)-ent-afzelechin (4, 2 mg), identified by comparing
spectroscopic data with literature data.7 The aqueous layer was
evaporated, and the residue was treated with L-cysteine methyl ester
hydrochloride (2 mg) in pyridine (0.5 mL) at 60 °C for 1 h. After the
reaction was completed, the solution was treated with Ac2O (3 mL) at
60 °C for 1 h. Authentic samples were prepared by the same
procedure. The acetate derivatives were subjected to gas chromatog-
raphy (GC) analysis. GC conditions: GC-2010 (Shimadzu) instru-
ment; detector, FID; column, TC-1 capillary column (0.25 mm × 30
m; GL Science, Inc.); column temperature, 230 °C; programmed
increase, 38 °C/min; carrier gas, N2 (1 mL/min); injection and
detector temperature, 270 °C. The sugar derivative thus obtained
showed a retention time of 21.30 min, identical to that of authentic D-
glucose.
AUTHOR INFORMATION
Corresponding Author
*Tel: +82-42-868-9465. Fax: +82-42-868-9471. E-mail: jskim@
kiom.re.kr.
■
Author Contributions
∥I.-S. Lee and S. Y. Yu contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was supported by a grant (K12040, K13040)
from the Korea Institute of Oriental Medicine. The NMR and
MS experiments were performed by the Korea Basic Science
Institute (KBSI). The authors thank Prof. C.-H. Kim
(Chungnam National University, Daejeon, Republic of
Korea) for providing zebrafish lines and for technical support.
Determination of AGE Formation. According to a well-
established method, the reaction mixture [bovine serum albumin (10
mg/mL, Sigma, St. Louis, MO, USA; 700 μL) in 50 mM phosphate
buffer (pH 7.4) with 0.02% sodium azide] was added to 0.2 M fructose
and glucose (100 μL). In screw cap tubes (1.5 mL), the reaction
mixture was mixed with 200 μL of serially diluted compounds and
aminoguanidine (Sigma). After incubating at 37 °C for 7 days, the
fluorescent reaction products (200 μL) were transferred to 96-well
plates and assayed on a spectrofluorometric detector (Bio-Tek,
Synergy HT, USA; excitation wavelength, 350 nm; emission
wavelength, 450 nm). The AGE assay was performed in triplicate.
The concentration of each test sample giving 50% inhibition of the
activity (IC50) was estimated from the least-squares regression line of
the logarithmic concentration plotted against remaining activity.
Measurement of Vessel Dilation in Larval Zebrafish. Adult
zebrafish were maintained under standard conditions at 28.5 °C under
a 14 h light/10 h dark cycle. Embryos were obtained from crosses
between flk:EGFP Tg (transgenic) fish and raised in egg water (sea
salt, 0.06 g/L). One-day flk:EGFP Tg embryos were placed into 24-
well plates (five embryos per well) and maintained in 2 mL volumes of
egg water with 30 mM glucose. HG-induced embryos were treated
with 10 or 20 μM of compounds from 1 day postfertilization (dpf) to
6 dpf. At 6 dpf, HG-induced embryos were fixed with 4% (v/v)
paraformaldehyde, and each lens containing hyaloid retinal vessels was
isolated and aligned so that the optic disc (OD) was facing upward.
Fluorescence images were obtained using an Olympus SZX16
stereomicroscope. The diameters of hyaloid vessels were measured
in 3−4 main branches of the OD, to the first sub-branch, using the
Image J software. All experiments were performed in triplicate.
REFERENCES
■
(1) Wei, M.; Gaskill, S. P.; Hanffner, S. M.; Stern, M. P. Diabetes Care
1998, 21, 1167−1172.
(2) Laakso, M. Diabetes 1999, 48, 937−942.
(3) Nishikawa, T.; Edelstein, D.; Du, X. L.; Yamagishi, S.;
Matsumura, T.; Kaneda, Y.; Yorek, M. A.; Beebe, D.; Oates, P. J.;
Hammes, H. P.; Giardino, I.; Brownlee, M. Nature 2000, 404, 787−
790.
(4) Brownlee, M. Nature 2001, 414, 813−820.
(5) Takenaka, K.; Yamagishi, S.; Matsui, T.; Nakamura, K.; Imaizumi,
T. Curr. Neurovasc. Res. 2006, 3, 73−77.
(6) Li, C.; Ikeda, H.; Ohba, H. In Wu, Z. Y.; Raven, P. H.; Hong, D.
Y., Eds. Flora of China, Vol. 9; Science Press: Beijing, and Missouri
Botanical Garden Press: St. Louis, 2003; p 384.
(7) Bilia, A. R.; Morelli, I.; Hamburger, M.; Hostettmann, K.
Phytochemistry 1996, 43, 887−892.
(8) Rawat, M. S. M.; Prasad, D.; Joshi, R. K.; Pant, G. Phytochemistry
1999, 50, 321−324.
(9) Prasad, D. Fitoterapia 2000, 71, 245−253.
(10) Mabry, T. J.; Markham, K. R.; Thomas, M. B. The Systematic
Identification of Flavonoids; Springer: New York, 1970; p 41.
(11) Jacques, D.; Haslam, E.; Bedford, G. R.; Greatbanks, D. J. Chem.
Soc., Perkin Trans. 1 1974, 2663−2671.
(12) Lou, H.; Yamazaki, Y.; Sasaki, T.; Uchida, M.; Tanaka, H.; Oka,
S. Phytochemistry 1999, 51, 297−308.
G
dx.doi.org/10.1021/np400442b | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(13) Kolodziej, H.; Sakar, M. K.; Burger, J. F. W.; Engelshowe, R.;
Ferreira, D. Phytochemistry 1991, 30, 2041−2047.
(14) Botha, J. J.; Young, D. A.; Ferreira, D.; Roux, D. G. J. Chem. Soc.,
Perkin Trans. 1 1981, 1213−1219.
(15) Hikino, H.; Shimoyama, N.; Kasahara, Y.; Takahashi, M.;
Konno, C. Heterocycles 1982, 19, 1381−1384.
(16) Korver, O.; Wilkins, C. K. Tetrahedron 1971, 27, 5459−5465.
(17) Balde, A. M.; De Bruyne, T.; Pieters, L.; Kolodziej, H.; Berghe,
D. V.; Claeys, M.; Vlietinck, A. Phytochemistry 1995, 38, 719−723.
(18) Morimoto, S.; Nonaka, G. I.; Nishioka, I. Chem. Pharm. Bull.
1987, 35, 4717−4729.
(19) Calzada, F.; Cedillo-Rivera, R.; Bye, R.; Mata, R. Planta Med.
2001, 67, 677−680.
(20) Jang, D. S.; Yoo, N. H.; Kim, N. H.; Lee, Y. M.; Kim, C. S.; Kim,
J.; Kim, J. H.; Kim, J. S. Biol. Pharm. Bull. 2010, 33, 329−333.
(21) Brownlee, M.; Vlassara, H.; Kooney, A.; Ulrich, P.; Cerami, A.
Science 1986, 232, 1629−1632.
(22) Morimitsu, Y.; Yoshida, K.; Esaki, S.; Hirota, A. Biosci.
Biotechnol. Biochem. 1995, 59, 2018−2021.
(23) Lou, H.; Yuan, H.; Yamazaki, Y.; Sasaki, T.; Oka, S. Planta Med.
2001, 67, 345−349.
(24) Stitt, A. W.; Li, Y. M.; Gardiner, T. A.; Bucala, R.; Archer, D. B.;
Vlassara, H. Am. J. Pathol. 1997, 150, 523−531.
(25) Stitt, A.; Gardiner, T. A.; Alderson, N. L.; Canning, P.; Frizzell,
N.; Duffy, N.; Boyle, C.; Januszewski, A. S.; Chachich, M.; Baynes, J.
W.; Thorpe, S. R. Diabetes 2002, 51, 2826−2832.
(26) Gardiner, T. A.; Anderson, H. R.; Stitt, A. W. J. Pathol. 2003,
201, 328−333.
(27) Alvarez, Y.; Chen, K.; Reynolds, A. L.; Waghorne, N.;
O’Connor, J. J.; Kennedy, B. N. Dis. Model. Mech. 2010, 3, 236−245.
(28) Bagchi, D.; Garg, A.; Krohn, R. L.; Bagchi, M.; Tran, M. X.;
Stohs, S. J. Res. Commun. Mol. Pathol. Pharmacol. 1997, 95, 179−189.
(29) Bagchi, M.; Bagchi, D. J.; Balmoori, J.; Stohs, S. J. Gen.
Pharmacol. 1998, 30, 771−776.
(30) Shahat, A. A.; Cos, P.; De Bruyne, T.; Apers, S.; Hammouda, F.
M.; Ismail, S. I.; Azzam, S.; Claeys, M.; Goovaerts, E.; Pieters, L.;
Vanden Berghe, D.; Vlietinck, A. J. Planta Med. 2002, 68, 539−541.
H
dx.doi.org/10.1021/np400442b | J. Nat. Prod. XXXX, XXX, XXX−XXX