Journal of Natural Products
Note
of sinapic acid rather than to the carboxy group of quinic acid
in 5-CQA. Concerning the latter linkage, it appears that
esterification of quinic acid to glucose or other saccharides is
extremely rare; a 6-quinylglucoside of the flavonol herbacetin
from Ephedra alata is the only example we found in the
natural products literature.
Standards and Reagents. 5-O-(E)-Caffeoylquinic acid (5-CQA),
-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), 1,1-
6
diphenyl-2-picrylhydrazyl (DPPH), and HPLC grade MeOH were
purchased from Sigma-Aldrich (St. Louis, MO, USA). Diammonium
2,2′-azinobis(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS) was from
12
TCIAce (Tokyo, Japan). Ultrapure H O was prepared using a Milli-
2
RO 12 Plus system (Millipore Corp., Bedford, MA, USA).
Plant Material. Seeds of Solanum viarum accession 311, collected
Antioxidant activity was quantified for viarumacids A and B
•
•+
using the DPPH and ABTS radical scavenging assays. Because
viarumacids A and B both include two 5-CQA moieties linked by
glucose, 5-CQA was included in the assays as a reference com-
pound in addition to Trolox. The TEAC values ± SD from the
by one of the authors (R.S.M.) in China, were obtained from The New
1
3
York Botanical Garden DNA bank. Seeds were germinated indoors,
and two plants were transferred to an outdoor bed at the Calder
Research Center in Armonk, New York, where they were grown in
the summer of 2009. On the basis of morphological features, the
plants were confirmed to be S. viarum by Dr. Michael Nee, curator of
Solanaceae at The New York Botanical Garden. For further confirmation,
DNA was isolated and the nuclear ribosomal internal transcribed spacer
•
DPPH assay were 1.707 ± 0.013, 2.471 ± 0.052, and 2.158 ±
0
.043, respectively, for 5-CQA, viarumacid A, and viarumacid B. The
•+
•
ABTS assay (Table 3) gave similar results to the DPPH assay.
1
3
was sequenced according to Meyer et al., which showed Meyer 311
GenBank ID JQ638805) to be a 100% match to S. viarum accession
AY561275 used in the Solanum phylogenetic study by Levin et al.
Table 3. ABTS• Results Reported in Trolox Equivalent
+
(
a
Antioxidant Capacity (TEAC)
1
4
(
2006). Two fruit from each plant of Meyer accession 311 were
5-CQA
viarumacid A (1)
viarumacid B (2)
harvested at the mature unripe stage, i.e., full size but mottled green
and white in color rather than yellow. The four fruit were pooled as a
time (min) TEAC
SD
TEAC
SD
TEAC
SD
0
5
1
1
2
2
3
3
4
1.44
1.67
1.68
1.60
1.72
1.65
1.71
1.72
1.73
±0.17
±0.12
±0.09
±0.05
±0.10
±0.07
±0.06
±0.05
±0.04
1.51
2.71
3.01
3.18
3.64
3.73
3.71
3.81
3.89
±0.26
±0.15
±0.22
±0.34
±0.17
±0.37
±0.16
±0.16
±0.16
−0.31
1.61
1.85
2.10
2.26
2.49
2.42
2.50
2.56
±0.25
±0.21
±0.21
±0.18
±0.22
±0.28
±0.18
±0.19
±0.20
single sample, frozen in liquid N , and stored at −80 °C. The frozen
2
tissue was lyophilized and powdered with a mortar and pestle prior to
extraction.
0
5
0
5
0
5
0
Fruit Tissue Extraction and Isolation of Compounds. Two 1.0 g
samples of the lyophilized, powdered tissue were each extracted twice
with 30 mL of MeOH−H O, 4:1, for 30 min using a magnetic stir bar
2
in a 50 mL screw-cap tube that was sealed after flushing with N . After
2
centrifugation for 5 min at 1500g, the supernatant was vacuum filtered
through a sintered glass funnel fitted with a glass fiber disk. The com-
bined extracts were reduced to 20 mL by N evaporation at 40 °C, and
2
a
Technical replicates: 7.
two 10 mL portions were fractionated on 500 mg Strata X polymeric
solid-phase extraction (SPE) tubes (Phenomenex, Torrance, CA,
USA) with a step gradient of 25%, 40%, and 60% aqueous MeOH
In both cases, viarumacid A was a better scavenger than
viarumacid B. Activity observed for viarumacid A in the ABTS
assay was roughly twice that of 5-CQA, suggesting that the caffeic
acid moieties contribute most or all of the activity (Table 3).
The roughly 1.5-fold greater values for viarumacid A compared
with B suggest that the malonyl group in B results in reduced
activity. Another factor that distinguished the viarumacids from
1
1
•
+
(10 mL each). Analysis of aliquots by C -HPLC-DAD showed the
18
4
0% MeOH SPE fraction was enriched in compounds 1 and 2. This
fraction was processed to yield two 1.2 mL samples in 10% aqueous
MeOH plus 0.02% H PO , which were injected in 80 μL portions onto
3
4
a Phenomenex Luna C18(2) column (5 μm particle size, 250 mm
long, 4.6 mm i.d.) for isolation of compounds 1 and 2 using an HP
1
100 Series HPLC system (Agilent Technologies) and a binary gra-
5
-CQA was the rate at which maximum activity was attained
dient consisting of 0.02% H PO in H O and MeOH as previously
3
4
2
•+
4
after initiation of incubation with the ABTS radical: 5-CQA
activity showed an immediate sharp increase and plateaued at
5
described by Ma et al.
DPPH Radical (DPPH ) Scavenging Activity. DPPH scaveng-
•
•
1
1
ing activity was quantified according to Ma et al. Five 1:2 serial
dilutions of each sample made from a stock solution of 250 μg/mL
were used to calculate the standard curve. Two technical replicates were
performed. In addition to Trolox, 5-CQA was included as a reference
standard. Activity was reported as Trolox equivalent antioxidant capacity
min, whereas activities of viarumacids A and B increased
asymptotically, rising rapidly from 0 to 5 min and reaching a
plateau by 20 to 25 min (Table 3). In summary, these two new
compounds that include two 5-CQA moieties as part of their
structure have antioxidant potentials exceeding those of 5-CQA
and Trolox. Moreover, the protracted radical scavenging activity of
the viarumacids compared with 5-CQA merits investigation into
whether these new 5-CQA derivatives exert different biological
activities potentially useful for the natural products industry.
(
TEAC, μM Trolox/μM compound) after incubation for 30 min.
•+
•+
ABTS Radical (ABTS ) Scavenging Activity. ABTS scaveng-
11
ing activity was quantified as described in Ma et al. Six 1:2 serial dilu-
tions were made from a stock solution of 500 μg/mL, and these were
used to calculate the standard curve. Seven technical replicates were
performed. Activity was reported in TEAC values at 5 min intervals
from 0 to 40 min.
EXPERIMENTAL SECTION
■
20
Viarumacid A (1): C H O , colorless oil; [α] −45.0 (c 0.020,
General Experimental Procedures. Optical rotations were
determined on an AUTOPOL III polarimeter (Rudolph Research
Analytical, Hackettstown, NJ, USA) equipped with a sodium lamp
38 44 22
D
−1
MeOH); IR νmax cm 3305, 2945, 2823, 1727, 1599, 1450, 1409,
106, 1025, and 604; UV (MeOH), λmax (log ε) 324 (4.26), 296
1
(
−
4.06); HRESIMS (negative) m/z 851.2282 ([M − H] calcd for
(
589 nm) and a 10 cm microcell. UV and IR spectra were obtained on
1
a UV-2450 spectrometer (Shimadzu, Japan) and a Nicolet iS10 spec-
C H O22, 851.2246), (positive) see Table 1; H NMR (methanol-d ,
300 MHz) and C NMR (methanol-d
3
8
4
3
4
1
3
trometer (Thermo Scientific, Waltham, MA, USA), respectively. A
Bruker Avance 300 NMR spectrometer, equipped with bbi (for H
, 75 MHz) data, see Table 2.
4
1
20
Viarumacid B (2): C38 22, colorless oil; [α]
H
44
O
−54.2 (c 0.048,
D
13
−1
and 2D) and bbo (for C) probes, was operated at 300.1312 MHz for
H and at 75.4753 MHz for 13C NMR experiments. HRESIMS was
performed using an LCT Premier XE TOF mass spectrometer
Waters, Milford, MA, USA) equipped with an ESI interface and con-
trolled by Masslynx V4.1 software.
MeOH); IR νmax cm 3313, 2945, 2823, 1726, 1598, 1446, 1409,
1
1115, 1017, and 620; UV (MeOH), λmax (log ε) 325 (4.19), 296 (4.00);
+
HRESIMS (negative) m/z 937.2253 ([M + H] calcd for C41
H
O ,
25
45
1
(
937.2250), positive please see Table 1; H NMR (methanol-d , 300
4
MHz) and 13C NMR (methanol-d , 75 MHz) data, see Table 2.
4
2
249
dx.doi.org/10.1021/np300553t | J. Nat. Prod. 2012, 75, 2246−2250