![HPLC chromatograms of 60% EtOH extract (ALEE) [ A ] and hot water extracts (ALWE) [ B ] on Capcell Pak UG120 (5 μm, 2.0 i.d. × 250 mm) (40 °C ). a UV 330 nm, b − f Positive ion ESI- MS chromatograms of total ion current profiles ( b : 1.0 a ) and selec- tive ion modes ( c − e ). [ A ] c m/z 315 (16.0); d m/z 299 (25.0); e m/z 285 (5.5); f m/z 247 (12.0) [ B ] c m/z 315 (6.0); d m/z 299 (7.0); e m/ z 285 (2.6); f m/z 247 (9.0).](https://www.researchgate.net/profile/Tetsuro-Ito/publication/249967998/figure/fig1/AS:298340467134472@1448141318305/HPLC-chromatograms-of-60-EtOH-extract-ALEE-A-and-hot-water-extracts-ALWE-B_Q320.jpg)
![Three-dimensional HPLC analysis of major chemical compounds in agarwood leaves. a time, 5.0 − 40.0 min; wavelength, 190 − 400 nm; intensity max, 180 mAU, b time, 10.0 − 40.0 min; wavelength, 240 − 400 nm; intensity max, 5 mAU, c UV spectra of key peaks [ 1 (benzophenone), 3 (xanthone), and 5 (flavone)].](https://www.researchgate.net/profile/Tetsuro-Ito/publication/249967998/figure/fig2/AS:298340467134473@1448141318327/Three-dimensional-HPLC-analysis-of-major-chemical-compounds-in-agarwood-leaves-a-time_Q320.jpg)
Content uploaded by Tetsuro Ito
Author content
All content in this area was uploaded by Tetsuro Ito
Content may be subject to copyright.
Food Sci. Technol. Res., 18 (2), 259–262, 2012
Identication of Phenolic Compounds in Aquilaria crassna Leaves Via Liquid
Chromatography-Electrospray Ionization Mass Spectroscopy
Tetsuro iTo
1
, Mamoru KaKino
2
, Shigemi TazaWa
3
, Masayoshi oyama
1
, Hiroe maruyama
3
, Yoko araKi
3
,
Hideaki
Hara
2
and Munekazu iinuma
1*
1
Laboratory of Pharmacognosy, Gifu Pharmaceutical University, 1-25-4 Daigaku-nishi, Gifu 501-1196, Japan
2
Laboratory of Molecular Pharmacology, Gifu Pharmaceutical University, 1-25-4 Daigaku-nishi, Gifu 501-1196, Japan
3
Nagaragawa Research Center, Api Company Limited, 692-3, Yamasaki, Nagara, Gifu 502-0071, Japan
Received June 19, 2011; Accepted December 19, 2011
In this study, we extracted Aquilaria crassna with aqueous ethanol and water and analyzed the extracts
via liquid chromatography diode array detection and electrospray ionization mass spectrometry (LC–
ESI–MS) methods. Phenolics were separated using semi-micro HPLC and were identied as iriophenone
3,5-C-β-diglucoside (1), iriophenone 3-C-β-glucoside (2), mangiferin (3), iriophenone 2-O-α-rhamnoside
(4), genkwanin 5-O-β-primeveroside (5), genkwanin 5-O-β-glucoside (6), genkwanin 4′-methyl ether
5-O-β-primeveroside (7), and genkwanin (8) via a comparison with authentic samples. The collision-in-
duced dissociation (CID)-MS/MS spectra of these polyphenols and the unknown chromatographic peaks
were detected using hybrid ion trap time-of-ight (IT-TOF) mass spectrometry. The results of the present
study demonstrated that LC–ESI–MS can be useful for the specic quality control of extracts of extracted
A. crassna.
Keywords: Aquilaria crassna, agarwood leaves, LC–ESI–MS, quality control, polyphenols, laxative effect
*To whom correspondence should be addressed.
E-mail: iinuma@gifu-pu.ac.jp
Introduction
Aquilaria is a woody plant in the Thymelaeaceae family
native to Southeast Asia, and some members of the genus,
represented by A. crassna and A. sinensis, are known as
incense trees (Ng, Chang, and Kadir, 1997). The depletion
of wild trees due to indiscriminate cutting of agarwood has
resulted in these trees being listed and protected as an endan-
gered species. In Thailand, A. crassna has been systemati-
cally cultivated to ensure a constant production of resin and
processed for perfume. In addition, the leaves have been
used as tea. The polyphenols present in A. crassna leaves
could be an important source of bioactive components. The
major polyphenols in this material are glycosides of flavo-
noids, benzophenones, and xanthones. Researchers have
recently claried the laxative properties of agarwood leaves
(AL) and identied mangiferin (3) and 5-O-β-primeveroside
(5) as the active components (Hara et al., 2008; Kakino et
al., 2010). The functional food applications of AL require an
exact understanding of not only the chemical components,
but also the AL properties such as their chemical natures,
sizes, solubilities, and the degrees and positions of glycosyl-
ation, which inuence their pharmacokinetics and pharmaco-
dynamics in humans. In addition, it is necessary to identify
the AL phenolics present in these suppliable materials (A.
crassna and A. sinensis) and control the standard qualities of
the extracts prepared via standardized conditions. Due to the
O OHHO
HO
OHO
β-glucose
HO
OH
OR
1
OH
O
1 : R
1
= H, R
2
= R
3
= β-glucose
2
: R
1
= R
3
= H, R
2
= β-glucose
4
: R
1
= α-rhamnose, R
2
= R
3
= H
R
2
R
3
3
O
OR
2
MeO
OR
1
O
5 : R
1
= β-primeverose, R
2
= H
6
: R
1
= β-glucose, R
2
= H
7
: R
1
= β-primeverose, R
2
= CH
3
8 : R
1
= R
2
= H
phase gradient was used with the percentage of B in A vary-
ing as follows: initial concentration, 10% B; 30 min, 50% B;
and 40 min, 50% B.
IT-TOF MS analysis The diode array analysis was per-
formed on a SPD-M20A (Shimadzu; Kyoto, Japan) scanning
in the range 200 − 400 nm. CID-MS
n
experiments were per-
formed on a hybrid IT-TOF mass spectrometer with an ESI
interface (Shimadzu). The negative ESI conditions were as
follows: high voltage probe, −3.5 kV; nebulizing gas flow,
1.5 L/min; CDL temperature, 200℃; heat block temperature,
200℃; and drying gas pressure, 200 KPa. CID parameters
were chosen as 70% for the CID energy and 50% for the
collision gas parameter. We used N
2
gas for CID. The detec-
tor voltage of TOF was 1.6 kV. A solution of triuoroacetic
acid and sodium hydrate was used as the standard sample to
adjust the sensitivity and resolution and to perform the mass
number calibration (ion trap and TOF analyzer).
Results and Discussion
General The common solvents used for the extraction
of phenolic compounds from foods include water, metha-
lack of standard methods for sample preparation, extraction,
and analysis, there is no general consensus on a standard pro-
tocol for the quantitation of phenolic compounds in AL. Our
present study focused only on the identication and not the
quantitation of these phenolic compounds in AL in 60% etha-
nol extracts (ALEE) and hot water extracts (ALWE). The ob-
jective of our study was to identify the phenolic compounds
and generate characteristic chromatographic fingerprints of
ALEE and ALWE via liquid chromatography–electrospray
ionization mass spectrometry (LC−ESI−MS) with multi-
stage analysis (MS
n
). Our study provides useful information
necessary for the generation of standardized AL materials for
in vitro and in vivo studies and for the authentication of AL-
based food products.
Experimental Methods
General experimental procedures The instruments used
in this study were DGU-20A3, LC-20AP, CBM-20A, SPD-
M20A, and SIL-20A LC instruments for semi-micro HPLC
(Shimadzu, Kyoto, Japan)and a Shimadzu hybrid IT-TOF
mass spectrometer (Shimadzu).
Plant material A. crassna was collected in Pechaboon,
Thailand, in October 2009 and was identied by M. Iinuma,
one of the authors. A voucher specimen has been deposited
at the herbarium of API Co. (Gifu, Japan).
Extraction procedures The dried and chopped leaves of
A. crassna (50 g) were separately extracted with either 60%
(v/v) ethanol (1.0 L, 24 h × 1, room temperature) or water (1.0
L, 1 h × 1, 95℃), and then were ltered. The extracts were
concentrated in vacuo at 50℃ to yield alcohol and water AL
extracts [ALEE (11.3 g) and ALWE (11.1 g)], respectively.
Reagents for HPLC All solvents were HPLC grade and
purchased from Sigma Aldrich Co. (St. Louis, MO). EA,
catechin, epicatechin, and quercetin standards were also pur-
chased from Sigma Aldrich Co. Reagent grade formic acid
(98%) was purchased from Nacalai Tesque Inc. (Tokyo, Ja-
pan). Ultra-pure water was prepared using a Millipore Milli-
Q purication system (Bedford, MA).
Sample preparation for LC−MS ALEE and ALWE
(each 40.0 mg) were dissolved in 50% ethanol and water (20
mL each), respectively, and injected directly for HPLC−MS
analysis.
HPLC analysis We performed HPLC analysis using a
Shimadzu HPLC system. Chromatographic separation was
performed on a Capcell Pak UG120 (5 μm, 2.0 i.d. × 250
mm; Shiseido, Tokyo, Japan). Mobile phase A was water
containing 0.1% acetic acid, and mobile phase B was CH
3
CN
containing 0.1% acetic acid. The column temperature was
40℃. The HPLC ow rate was 0.2 mL/min. A sample solu-
tion of 1 μL was injected into the HPLC system. A mobile
Fig. 1. HPLC chromatograms of 60% EtOH extract (ALEE) [A]
and hot water extracts (ALWE) [B] on Capcell Pak UG120 (5 μm,
2.0 i.d. × 250 mm) (40℃). a UV 330 nm, b − f Positive ion ESI-
MS chromatograms of total ion current proles (b: 1.0
a
) and selec-
tive ion modes (c − e). [A] c m/z 315 (16.0); d m/z 299 (25.0); e m/z
285 (5.5); f m/z 247 (12.0) [B] c m/z 315 (6.0); d m/z 299 (7.0); e m/
z 285 (2.6); f m/z 247 (9.0).
a
Relative magnitudes are given in parentheses.
T. iTo et al.260
sults would give promising information for the determination
of the molecular composition, and the negative mode would
provide extensive information via CID fragmentations. The
product ion spectra of the pseudomolecular ions [M+H]+ and
[M−H]− as well as the selective ion chromatograms for the
aglycons, were obtained by conducting CID-MS/MS experi-
ments. The chromatograms of the MS total ion current (TIC)
and the selective ion monitoring (SIM) mode in positive
mode are shown in Fig. 1. The MS
n
data of 1 − 5 are summa-
rized in Table 1.
Identication of 1 – 8 by HPLC in ALEE and ALWE
HPLC profiles of ALEE and ALWE monitored at 330 nm
showed that peaks numbered 1 − 8 were completely sepa-
rated (Fig. 1). These peaks were identied as iriophenone
3,5-C-β-diglucoside (1), iriflophenone 3-C-β-glucoside (2),
nol, aqueous acetone, ethanol, and ethyl acetate (Naczk and
Shahidi, 2004). Because our primary aim is to use AL as a
food material in Japan, we performed the preliminary ex-
tractions with water and various concentrations of aqueous
ethanol at various temperatures. The results of the compre-
hensive HPLC analysis (data not shown) indicated that the
60% aqueous ethanol at room temperature and water at 95℃
extracts (ALEE and ALWE) were similar and yielded the
most peaks in the HPLC–diode array detection (DAD) chro-
matogram when recorded at 330 nm (Fig. 1, [A] a and [B] a).
The peaks showed absorbance wavelengths (200 − 400 nm)
typical of phenolics, including benzophenones (1, 2, and 4),
xanthone (3), and avones (5 − 8) (Fig. 2, a – c). However,
80% ethanol is also the accepted solvent for the most effi-
cient extraction to investigate one of the laxative principles
(5) (data not shown). The extraction scheme, facilitated by
high density ethanol, requires custom equipment and is not a
practical method that can be used in API Co. (Gifu, Japan).
We discuss the comparative study of ALEE and ALWE in
this paper.
To obtain good resolution of the peaks in a reasonably
short analysis time, we screened different mobile phases and
semi-micro column compositions. We found that one of the
suitable eluting solvent systems was acetonitrile and 0.1%
aqueous acetic acid, and the suitable column was a Capcell
Pak UG120 (5 μm, 2.0 i.d. × 250 mm; Shiseido; Tokyo,
Japan). The optimized LC conditions permitted good sepa-
ration of eight target polyphenols as well as other phenolic
compounds within 35 min.
The wavelength for the comparison of the chromato-
grams of ALEE and ALWE was 330 nm which is the λmax
of 5, and we performed a DAD analysis, scanning in the
range 190 − 400 nm. For the MS analysis, we utilized both
positive and negative ion modes of ESI, since dual mode re-
Fig. 2. Three-dimensional HPLC analysis of major chemical compounds in agarwood leaves.
a time, 5.0 − 40.0 min; wavelength, 190 − 400 nm; intensity max, 180 mAU, b time, 10.0 − 40.0 min; wavelength, 240 − 400 nm; intensity
max, 5 mAU, c UV spectra of key peaks [1 (benzophenone), 3 (xanthone), and 5 (avone)].
Table 1. MS
n
spectra in the positive and negative ion modes of
major polyphenol glycosides in AL.
Peak
a
Rt (min) Isolated m/z MS
n
type Major product ions (m/z)
1 6.9 571 [M+H]+ MS
2
535 (100)
b
, 433 (57)
569 [M−H]− MS
2
449 (21), 431 (13), 359 (57), 329 (100)
449 [M−H]− MS
3
329 (100)
359 [M−H]− MS
3
239 (100)
2 7.4 409 [M+H]+ MS
2
391 (100), 325 (64), 313 (64)
407 [M−H]− MS
2
287 (100)
3 9.3 423 [M+H]+ MS
2
369 (17), 357 (17), 351 (26), 327 (41),
303 (21), 299 (17), 273 (100)
421 [M−H]− MS
2
331 (49), 301 (100), 271 (23), 259 (37)
310 [M−H]− MS
3
273 (25), 272 (25), 258 (100)
4 11.8 393 [M+H]+ MS
2
339 (100), 247 (62)
339 [M+H]+ MS
3
247 (100)
391 [M−H]− MS
2
245 (100), 151 (15)
245 [M−H]− MS
3
151 (100)
5 16.0 579 [M+H]+ MS
2
285 (100)
577 [M−H]− MS
2
283 (100), 268 (24)
283 [M−H]− MS
3
268 (100)
a
Peak numbers are the same as in Fig. 1.
b
Relative abundances (%) are given in parentheses.
Phenolic Compounds in Agarwood Leaves 261
methoxyavone, MW 284) for 5 and 6, and 5-hydroxy-7,4′-
dimethoxyavone (MW 298) for 7. The SIM chromatograms
in Fig. 1 present those of [M+H]+ at m/z 299 (d), 285, (e)
and 247 (f) in addition to that of the noticeable ion at m/z
315 (c). It was evident that 4 − 7 produced product ion peaks
of their aglycons. In addition, we also detected some O-
glucosides hitherto unidentied in the SIM chromatograms,
exemplied by the aglycons with [M+H]+ at m/z 315 (c) and
247 (f). The composition of the SIM peak for [M+H]+ at m/
z 315 at 33.6 min (Fig. 1, [A] c) was calculated for C
17
H
15
O
6
with a pseudomolecular ion [M+H]+ at m/z 315.0878 (calcd.
315.0863), corresponding to dihydroxydimethoxyflavone
and adding a further ngerprint to AL.
Conclusions
In conclusion, we have identified the major phenolic
compounds present in agarwood leaves (AL) and established
their characteristic chromatographic profiles and composi-
tional similarities for 60% ALEE and ALWE. Among the
identified compounds, genkwanin 5-O-β-primeveroside (5)
and mangiferin (3) showed laxative activities via different
mechanisms.
Because it is meaningful to evaluate chemical composi-
tional similarities and biological properties of various extract
forms, these methods should aid in the standardization of AL
materials for in vitro and in vivo studies. The chromatograph-
ic ngerprinting of AL should also be useful for the authenti-
cation of AL-based food products. The information provided
by our study will aid in the evaluation of the importance of
AL consumption on human health.
References
Ng, L.T., Chang, Y.S. and Kadir, A.A. (1997) A review on agar
(gaharu) producing Aquilaria species. J. Tropic. Forest Prod., 2,
272-285.
Hara, H., Ise, Y., Morimoto, N., Shimazawa, M., Ichihashi, K.,
Ohyama, M. and Iinuma, M. (2008) Laxative effect of agarwood
leaves and its mechanism. Biosci. Biotechnol. Biochem., 72, 335-
345.
Kakino, M., Izuta, H., Ito, T., Tsuruma, K., Araki, Y., Shimazawa,
M., Oyama, M., Iinuma, M. and Hara, H. (2010) Agarwood in-
duced laxative effects via acetylcholine receptors on loperamide-
induced constipation in mice. Biosci. Biotechnol. Biochem., 74,
1550-1555.
Naczk, M. and Shahidi, F. (2004) Extraction and analysis of pheno-
lics in food. J. Chromat. B, 1054, 95-111.
mangiferin (3), iriflophenone 2-O-α-rhamnoside (4), genk-
wanin 5-O-β-primeveroside (5), genkwanin 5-O-β-glucoside
(6), genkwanin 4′-methyl ether 5-O-β-primeveroside (7), and
genkwanin (8) by comparison with standard samples. The
composition of ALEE (Fig. 1, [A]) and ALWE (Fig. 1, [B])
were similar except for the ratios of representative polyphe-
nols (1 − 8). Because our study focused on identication and
not quantitation, the differences in the ratio of each polyphe-
nol with the peak of 3 were discussed. For quality control of
the AL extracts, it is important to identify exact information
on the bioactive polyphenols bearing laxative properties (3
and 5). We found evidence for the existence of both com-
pounds in both extracts as well as the effective extraction of
5 with hot water, strongly indicating that both extracts are
substantial functional food materials with signicant laxative
effects. The other evident differences were the ratios of the
other polyphenols such as 1 (ALEE < ALWE), 2 (ALEE <
ALWE), 7 (ALEE < ALWE), and 8 (ALEE > ALWE). These
differences were acceptable when we considered the struc-
tures of 1 − 8 and the solvents used for the extraction. Pheno-
lic glycosides, represented as 1 − 7 in AL, are well known to
dissolve in hot water. In contrast, genin (8) was more soluble
in organic solvents, thus reecting the analytical results.
DAD analysis of major polyphenols in the ALEE
Three-dimensional HPLC analysis showed absorbance wave-
lengths typical of phenolic compounds. Figure 2 presents the
UV spectra of three majorities (Fig. 2 a, maximum mAU:
180) and the others (Fig. 2 b, maximum mAU: 5), as well as
UV spectra of 1, 3, and 5 (Fig. 2c). The results indicated that
the LC−DAD ngerprint of the AL extract existed as three
major peaks of two benzophenones (1 and 2) and xanthone (3)
and the other minor peaks of benzophenones (4) and avones
(5 and 8).
ESI-MS/IT-TOF analysis of ALEE and ALWE Figure
1 shows the chromatograms of TIC and SIM in the positive
mode (A, ALEE and B, ALWE). Not all compounds in AL
were detected in the DA detector, e.g., due to lack of conju-
gated systems in some compounds in the extract. However,
the TIC chromatograms of ALEE (Fig. 1, [A] b) and ALWE
(Fig. 1, [B] b) showed superimposable features with those
of the respective chromatograms at 330 nm (a) except for
the initial 5 min. The TIC chromatograms also supported the
compositional similarity of ALEE and ALWE.
The major phenolic compounds in AL existed as C-glu-
cosides (1 − 3) and O-glucosides (4 − 7), and the aglycons
of 4 − 7 were iriflophenone (1,3,4,4′-tetrahydroxy benzo-
phenone, MW 246) for 4, genkwanin (8: 5,4′-dihydroxy-7-
T. iTo et al.262
