Synthesis of deuterium-labeled cinnamic acids: Understanding the volatile benzenoid pathway in the flowers of the Japanese loquat Eriobotrya japonica

Article abstract of DOI:10.1002/jlcr.3933

Cinnamic acids are widely distributed in plants, including crops for human use, and exhibit a variety of activities that are beneficial to human health. They also occupy a pivotal position in the biosynthesis of phenylpropanoids such as lignins, anthocyanins, flavonoids, and coumarins. In this context, deuterium-labeled cinnamic acids have been used as tracers and internal standards in food and medicinal chemistry as well as plant biochemistry. Therefore, a concise synthesis of deuterium-labeled cinnamic acids would be highly desirable. In this study, we synthesized deuterium-labeled cinnamic acids using readily available deuterium sources. We also investigated a hydrogen–deuterium exchange reaction in an ethanol-d₁/Et₃N system. This method can introduce deuterium atoms at the ortho and para positions of the phenolic hydroxy groups as well as at the C-2 position of alkyl cinnamates and is applicable to various phenolic compounds. Using the synthesized labeled compounds, we demonstrated that the benzenoid volatiles, such as 4-methoxybenzaldehyde, in the scent of the flowers of the Japanese loquat Eriobotrya japonica are biosynthesized from phenylalanine via cinnamic and 4-coumaric acids. This study provides easy access to a variety of deuterium-labeled (poly)phenols, as well as to useful tools for studies of the metabolism of cinnamic acids in living systems.

Full text of DOI:10.1002/jlcr.3933

Received: 21 May 2021
DOI: 10.1002/jlcr.3933
Revised: 6 July 2021
Accepted: 7 July 2021
R E S E A R C H A R T I C L E
Synthesis of deuterium-labeled cinnamic acids:
Understanding the volatile benzenoid pathway in the
flowers of the Japanese loquat Eriobotrya japonica
1
2
2
Bunta Watanabe
| Shiori Nishitani | Takao Koeduka
1
Institute for Chemical Research, Kyoto
University, Kyoto, Japan
Cinnamic acids are widely distributed in plants, including crops for human
use, and exhibit a variety of activities that are beneficial to human health. They
also occupy a pivotal position in the biosynthesis of phenylpropanoids such as
lignins, anthocyanins, flavonoids, and coumarins. In this context, deuterium-
labeled cinnamic acids have been used as tracers and internal standards in
food and medicinal chemistry as well as plant biochemistry. Therefore, a con-
cise synthesis of deuterium-labeled cinnamic acids would be highly desirable.
In this study, we synthesized deuterium-labeled cinnamic acids using readily
available deuterium sources. We also investigated a hydrogen–deuterium
2
Graduate School of Sciences and
Technology for Innovation, Yamaguchi
University, Yamaguchi, Japan
Correspondence
Bunta Watanabe, Institute for Chemical
Research, Kyoto University, Gokasho, Uji,
Kyoto 611-0011, Japan.
Email: bunta@scl.kyoto-u.ac.jp
Funding information
exchange reaction in an ethanol-d /Et N system. This method can introduce
Japan Society for the Promotion of
Science, Grant/Award Number: 20K05840
1
3
deuterium atoms at the ortho and para positions of the phenolic hydroxy
groups as well as at the C-2 position of alkyl cinnamates and is applicable to
various phenolic compounds. Using the synthesized labeled compounds, we
demonstrated that the benzenoid volatiles, such as 4-methoxybenzaldehyde, in
the scent of the flowers of the Japanese loquat Eriobotrya japonica are bio-
synthesized from phenylalanine via cinnamic and 4-coumaric acids. This study
provides easy access to a variety of deuterium-labeled (poly)phenols, as well as
to useful tools for studies of the metabolism of cinnamic acids in living
systems.
(
to T.K.)
K E Y W O R D S
benzenoid volatiles, biosynthesis, cinnamic acids, Eriobotrya japonica, hydrogen–
deuterium exchange
1
4
1
| INTRODUCTION
antihypercholesterolemia, antiobesity, and antidepres-
6
sant activity. The simplest member of cinnamic acids
Cinnamic acids, such as cinnamic acid (1a), ferulic acid
2a), sinapic acid (3a), and 4-coumaric acid (4a), are
derivatives of acrylic acid with substituted or
unsubstituted phenyl group at the C-3 position
(1a) is biosynthesized from phenylalanine via the enzy-
matic reaction of phenylalanine ammonia lyase and sub-
sequently derivatized to cinnamoyl CoA, other cinnamic
acids and their CoA esters, cinnamaldehydes, and
cinnamyl alcohols. These cinnamic acid derivatives serve
as biosynthetic intermediates of diverse end-products of
the phenylpropanoid pathway such as lignins, anthocya-
(
a
(
Figure 1). They are widely distributed in plants that
1
–3
include crops for human use
and exhibit
a
variety of beneficial effects on human health
1,3,4
1–4
1,3,5
7,8
such as antioxidant,
anticancer,
antimicrobial,
nins, flavonoids, and coumarins.
J Label Compd Radiopharm. 2021;64:403–416.
wileyonlinelibrary.com/journal/jlcr
© 2021 John Wiley & Sons, Ltd.
403

404
WATANABE ET AL.
2
2
| EXPERIMENTAL
.1 | General information
Melting points (mp) were measured using an AS ONE
ATM-01 mp apparatus and are uncorrected. Nuclear
magnetic resonance (NMR) spectra were obtained using
a JEOL JNM-ECA600 or a Bruker AVANCEIII 600 spec-
FIGURE 1 Representative cinnamic acids (1a–4a) and their
deuterium-labeled counterparts (1b–4b)
1
2
trometer (600 MHz for H; 92 MHz for H; 151 MHz for
13
C). Chemical shifts are reported in parts per million rel-
To investigate the complicated biosynthesis of phen-
ylpropanoids, stable-isotope-labeled cinnamic acids and
their derivatives have frequently been used. For example,
ative to the internal standards (tetramethylsilane
1
(0.00 ppm) for H; CHCl3 (7.26 ppm) and CH SOCH
3
3
3
2
(2.55 ppm) for H; CDCl (77.00 ppm) and CD SOCD
3
3
13
using
deuterium-labeled
cinnamic
acids,
(39.70 ppm) for
C). High-resolution mass spectra
cinnamaldehydes, and cinnamyl alcohols, Sakakibara
and coworkers have quantitated their endogenous coun-
terparts in the seeds of Carthamus tinctorius and demon-
strated that ferulic acid serves as a biosynthetic precursor
(HRMS) were recorded using a JEOL JMS-700 or a
Bruker timsTOF spectrometer. Electron ionization (EI),
fast atom bombardment (FAB) using 3-nitrobenzyl alco-
hol (NBA) as a matrix, or electrospray ionization (ESI)
were used as ionization methods. Flash column chroma-
tography on silica gel was carried out using a Biotage
Isolera One chromatograph with SNAP Ultra cartridges
(silica gel, 25 μm).
9
of lignans in the plant. Deuterium-labeled cinnamic
acids have also been used as tracers in the biosynthetic
investigation of coumarins in the root of the cassava
10
plant (Manihot esculenta Crantz). Moreover, using
deuterium-labeled coniferyl alcohol, Lairez and
coworkers have analyzed the polymerization of coniferyl
alcohol in a pectin solution as a model system for the first
Reagents were used as received from common com-
mercial suppliers unless otherwise stated. 5 (98+% D)
was purchased from FUJIFILM Wako Pure Chemical
11
step of lignification.
Corporation (Osaka, Japan). CD I (99.9% D) and 17
3
In addition to plant biochemistry, stable-isotope-
labeled cinnamic acids have also been used in the areas
of food and medicinal chemistry. For example, they have
been used as internal standards for the determination of
the concentration of cinnamic acids in milk to establish a
(98.9% D) were purchased from C/D/N Isotopes Inc.
(Quebec, Canada). CD OD (min 99.8% D) was purchased
3
from Merck KGaA (Darmstadt, Germany). Ethanol-d1
(EtOD, 99% D) was purchased from Cambridge Isotope
Laboratories, Inc. (MA, USA). In the hydrogen–
deuterium exchange reaction of phenols in the EtOD/
Et N system, Et N was dried over 3A molecular sieves
12
method that can determine the type of cattle forage.
Piber and coworkers have used deuterium-labeled ferulic
acid as a tracer to detect chemical reactions of ferulic acid
3
3
prior to use, whereas in the Mizoroki–Heck reaction,
13
16
17
18,19
20
in flours during bread baking. In addition, using
deuterium-labeled cinnamic acids as an authentic sample
and an internal standard, Sarkissian and coworkers have
quantitated cinnamic acids as one of metabolites of phe-
nylalanine in the plasma and urine of phenylketonuric
Et N was used as received. 22a, 24a, 25a,
and 33a
31a,
3
21,22
were prepared according to literature proce-
dures. A pressure tube (Ace pressure tube equipped with
front seal plug, Z181099) was purchased from Sigma-
Aldrich, Co. (MO, USA). The flowers of E. japonica were
collected from the farm at the Yoshida campus of
Yamaguchi University, Japan in the winter 2019–2020.
14
model mice. The development of efficient synthetic
routes to stable-isotope-labeled cinnamic acids and their
derivatives is thus strongly desirable.
Herein, we report the synthesis of deuterium-labeled
cinnamic acids 1b–4b (Figure 1). We also investigated a
hydrogen–deuterium exchange reaction in an ethanol-d1/
2
2.2 | Ethyl (E)-3-[ H ]phenylacrylate
5
2
(6b), ethyl (Z)-3-[ H ]phenylacrylate (7),
5
2
Et N system that can introduce deuterium atoms at the
and 1-ethoxy-3-[ H ]phenyl-
3
5
ortho and para positions of the phenolic hydroxy group
as well as at the C-2 position of alkyl cinnamates. More-
over, using the synthesized labeled compounds, we dem-
onstrated that 4-methoxybenzaldehyde and methyl
1-trimethylsiloxyallene (8)
2.2.1 | Preparation of the Grignard reagent
4
-methoxybenzoate, the major benzenoid volatiles in the
A suspension of magnesium turnings (1.62 g, 66.6 mmol)
in anhydrous THF (10 ml) in the presence of a catalytic
amount of iodine was treated with approximately 1/20th
15
flower scent of the Japanese loquat Eriobotrya japonica,
are biosynthesized from phenylalanine via 1a and 4a.

WATANABE ET AL.
405
13
of a solution of 5 (10.4 g, 64.2 mmol) in anhydrous THF
J = 7.1 Hz), 6.82 (1H, s). C NMR (151 MHz, CDCl , δ):
3
(
30 ml) at room temperature under a nitrogen atmo-
ꢀ1.68 (3C), 14.13, 60.42, 127.73 (2C, t, J = 23.8 Hz),
sphere. The mixture was heated until the solvent began
to boil gently to initiate the formation of the Grignard
reagent. Then, the mixture was treated dropwise with the
rest of the solution of 5 under cooling by an ice bath.
After completion of the addition, the mixture was stirred
for 30 min at room temperature and then left to stand for
127.78 (2C, t, J = 24.6 Hz), 127.89 (t, J = 25.3 Hz),
+
136.34, 138.61, 141.24, 171.99. HRMS-EI (m/z): [M]
calcd for C H D O Si, 253.1546; found, 253.1541.
14
15 5 2
2
2.3 | (E)-3-[ H ]Phenylacrylic acid (1b)
5
2
9 h at room temperature under a nitrogen atmosphere.
The black supernatant (35 ml, ca. 56 mmol) was used in
the next step.
A mixture of 6 (500 mg, 2.76 mmol), EtOH (2.0 ml), and
an aqueous NaOH solution (10% w/v, 2.0 ml) was stirred
for 1 h at room temperature, and 2 M HCl (5.0 ml) and
water (5.0 ml) were then added to the mixture. The pre-
cipitate was collected using suction filtration, washed
successively with water and hexane, and dried in vacuo
to give 1b (330 mg, 78%) as a colorless solid. mp 134–
2.2.2 | Reaction of the Grignard reagent with
ethyl propiolate
ꢁ
23
ꢁ
1
A mixture of CuI (2.90 g, 15.2 mmol) and LiCl (1.30 g,
135 C (lit. mp 133 C). H NMR (600 MHz, CDCl , δ):
3
3
1
0.7 mmol) in anhydrous THF (200 ml) was stirred for
5 min at room temperature under a nitrogen atmo-
6.46 (1H, d, J = 16.0 Hz), 7.81 (1H, d, J = 16.0 Hz). The
2
carboxylic acid proton was not observed. H NMR
1
3
sphere, and then ethyl propiolate (3.0 ml, 30 mmol) and
(92 MHz, CHCl , δ): 7.46 (br s), 7.62 (br s).
C
3
Me SiCl (5.0 ml, 39 mmol) were successively added to
NMR (151 MHz, CDCl , δ): 117.32, 127.94 (2C, t,
J = 23.8 Hz), 128.44 (2C, t, J = 23.8 Hz), 130.22 (t,
3
3
ꢁ
ꢁ
the mixture at ꢀ78 C. After 10 min of stirring at ꢀ78 C,
the Grignard reagent solution prepared as described in
the previous paragraph was added dropwise to the mix-
ture over 1 h, and the reaction mixture was stirred for an
additional hour at the same temperature. Then, the reac-
tion was quenched by the dropwise addition of 2 M HCl
J = 24.6 Hz), 133.86, 147.04, 172.59. HRMS-EI (m/z):
+
[M] calcd for C H D O , 153.0838; found, 153.0837.
9
3 5 2
2
2.4 | 4-Hydroxy-3-[ H ]
3
(
30 ml), and the mixture was allowed to warm to room
temperature. The mixture was diluted with water
150 ml), and the separated aqueous layer was extracted
methoxybenzaldehyde (10)
(
NaH (3.20 g, 60% dispersion in mineral oil, 80 mmol) was
added portionwise to an ice-cold solution of 9 (5.00 g,
36.2 mmol) in anhydrous DMSO (90 ml). After 30 min of
with EtOAc (3 ꢂ 50 ml). The combined organic layers
were washed successively with water (2 ꢂ 100 ml) and
brine (50 ml), dried over anhydrous MgSO and filtered,
stirring at room temperature, a solution of CD I (5.40 g,
4
3
before the filtrate was concentrated in vacuo. The residue
was purified by flash column chromatography on silica
gel (hexane/EtOAc = 98/2 ! 95/5, v/v). 8 (538 mg, 7.2%)
eluted first followed by 7 (1.25 g, 23%), while 6b (3.18 g,
37.3 mmol) dissolved in anhydrous DMSO (10 ml) was
added dropwise to the mixture. The reaction mixture
was stirred for 11 h at room temperature and quenched
ꢁ
by the dropwise addition of 2 M HCl (30 ml) at 0 C. The
5
9%) eluted last. 6b–8 were obtained as colorless oils. 6b:
mixture was diluted with water (200 ml) and extracted
with toluene (3 ꢂ 75 ml). The combined organic layers
were washed successively with water (2 ꢂ 50 ml) and
1
H NMR (600 MHz, CDCl , δ): 1.34 (3H, t, J = 7.1 Hz),
4
3
.27 (2H, q, J = 7.1 Hz), 6.44 (1H, d, J = 16.2 Hz), 7.69
13
(1H, d, J = 16.2 Hz). C NMR (151 MHz, CDCl , δ):
brine (50 ml), dried over anhydrous MgSO and filtered,
3
4
1
4.30, 60.48, 118.25, 127.60 (2C, d, J = 23.8 Hz), 128.34
before the filtrate was concentrated in vacuo. The com-
bined aqueous layers were extracted again with CHCl3
(3 ꢂ 75 ml) because the toluene extraction was not suffi-
(2C, d, J = 24.6 Hz), 129.68 (d, J = 23.8 Hz), 134.29,
+
1
44.52, 167.00. HRMS-EI (m/z): [M]
calcd for
1
C H D O , 181.1151; found, 181.1149. 7: H NMR
cient. The CHCl layers were combined and washed with
11
7
5
2
3
(
600 MHz, CDCl , δ): 1.24 (3H, t, J = 7.1 Hz), 4.17 (2H,
brine (50 ml), dried over anhydrous MgSO and filtered,
3
4
q, J = 7.1 Hz), 5.95 (1H, d, J = 12.5 Hz), 6.95 (1H, d,
before the filtrate was concentrated in vacuo. The crude
products thus obtained were combined and purified by
flash column chromatography on silica gel (hexane/
EtOAc = 80/20 ! 60/40, v/v) to give 10 (4.33 g, 77%) as
13
J = 12.5 Hz). C NMR (151 MHz, CDCl , δ): 14.06,
3
6
0.26, 119.84, 127.44 (2C, t, J = 24.6 Hz), 128.44 (t,
J = 24.6 Hz), 129.25 (2C, t, J = 24.6 Hz), 134.66, 142.89,
+
ꢁ
11,24
ꢁ
1
1
1
66.21. HRMS-EI (m/z): [M] calcd for C H D O ,
a pale-yellow solid. mp 82 C (lit.
mp 81–83 C). H
11
7 5 2
1
81.1151; found, 181.1151. 8: H NMR (600 MHz, CDCl ,
NMR (600 MHz, CDCl , δ): 6.25 (1H, s), 7.05 (1H, d,
3
3
13
δ): 0.24 (9H, s), 1.22 (3H, t, J = 7.1 Hz), 4.20 (2H, q,
J = 7.9 Hz), 7.42–7.44 (2H, m), 9.83 (1H, s). C NMR

406
WATANABE ET AL.
(
1
151 MHz, CDCl , δ): 55.22 (septet, J = 22.2 Hz), 108.77,
J = 15.8 Hz), 6.80 (1H, d, J = 8.1 Hz), 7.09 (1H, dd,
J = 8.1, 1.8 Hz), 7.28 (1H, d, J = 1.8 Hz), 7.50 (1H, d,
3
14.39, 127.49, 129.72, 147.15, 151.75, 190.98. The spectra
24
2
were in good accordance with those reported.
J = 15.8 Hz), 9.55 (1H, s), 12.14 (1H, br s). H NMR
13
(
92 MHz, CH SOCH , δ): 3.77 (br s). C NMR (151 MHz,
3
3
CD SOCD , δ): 55.07 (septet, J = 21.7 Hz), 111.33,
3
3
2
2
.5 | Ethyl (E)-3-(4-hydroxy-3-[ H ]
115.72, 115.83, 123.01, 125.99, 144.74, 148.12, 149.28,
3
+
methoxyphenyl)acrylate (11b) and its (Z)-
isomer (12)
168.21. HRMS-EI (m/z): [M] calcd for C H D O ,
10
7 3 4
197.0767; found, 197.0766.
A mixture of 10 (3.75 g, 24.2 mmol) and (car-
2
bethoxymethylene)triphenylphosphorane
(9.00 g,
2.7 | (E)-3-(4-Hydroxy-3-[ H ]
3
2
3
5.8 mmol) in anhydrous toluene (100 ml) was stirred for
h at 90 C. The reaction mixture was cooled to room
methoxyphenyl)prop-2-enol (13)
ꢁ
temperature before the addition of silica gel (100 ml).
The solvent was evaporated, and the residue was sub-
jected to flash column chromatography on silica gel (hex-
ane/EtOAc = 85/15 ! 50/50, v/v). 12 (838 mg, 15%)
eluted first, followed by 11b (4.60 g, 85%). Both 11b and
A solution of 11 (2.60 g, 11.5 mmol) in anhydrous CH Cl
2 2
(150 ml) was treated dropwise with a solution of
diisobutylaluminum hydride (1.0 M in toluene, 48 ml,
ꢁ
48 mmol) at ꢀ78 C under an argon atmosphere, and the
mixture was stirred for 2 h at room temperature.
The reaction was quenched by the dropwise addition of
an aqueous solution of potassium sodium tartrate
1
2 were obtained as oils at first and solidified during stor-
ꢁ
age in a refrigerator. 11b: Colorless solid. mp 55–58 C.
1
ꢁ
H NMR (600 MHz, CDCl , δ): 1.34 (3H, t, J = 7.2 Hz),
tetrahydrate (50% w/v, 100 ml) at ꢀ78 C. The mixture
3
4
.26 (2H, q, J = 7.2 Hz), 5.88 (1H, s), 6.29 (1H, d,
was stirred for a further 21 h at room temperature and
then slightly acidified (pH = 6) with 2 M HCl. The sepa-
rated aqueous layer was extracted with CHCl3
(2 ꢂ 50 ml), and the combined organic layers were
J = 16.2 Hz), 6.92 (1H, d, J = 8.2 Hz), 7.03 (1H, d,
J = 1.7 Hz), 7.07 (1H, dd, J = 8.2, 1.7 Hz), 7.61 (1H, d,
13
J = 16.2 Hz). C NMR (151 MHz, CDCl , δ): 14.27, 55.03
3
(septet, J = 22.2 Hz), 60.31, 109.30, 114.71, 115.48,
washed with brine (100 ml), dried over anhydrous MgSO
4
1
22.91, 126.91, 144.66, 146.75, 147.91, 167.30. HRMS-EI
and filtered, before the filtrate was concentrated in vacuo.
The residue was purified by flash column chromatogra-
phy on silica gel (hexane/EtOAc = 50/50 ! 20/80, v/v)
+
(
m/z): [M] calcd for C H D O , 225.1080; found,
12 11 3 4
ꢁ
1
2
25.1080. 12: Colorless solid. mp 37–39 C. H NMR
ꢁ
(600 MHz, CDCl , δ): 1.29 (3H, t, J = 7.1 Hz), 4.19 (2H,
to give 13 (1.95 g, 92%) as a colorless solid. mp 74–75 C.
3
1
q, J = 7.1 Hz), 5.81 (1H, d, J = 13.1 Hz), 5.85 (1H, s), 6.79
H NMR (600 MHz, CDCl , δ): 1.84 (1H, br s), 4.29 (2H,
3
(
1H, d, J = 13.1 Hz), 6.88 (1H, d, J = 8.1 Hz), 7.11 (1H,
br d, J = 5.5 Hz), 5.83 (1H, s), 6.20 (1H, dt, J = 15.8,
5.5 Hz), 6.51 (1H, br d, J = 15.8 Hz), 6.85 (1H, d,
J = 8.2 Hz), 6.87 (1H, dd, J = 8.2, 1.5 Hz), 6.89 (1H, d,
13
dd, J = 8.1, 2.1 Hz), 7.77 (1H, d, J = 2.1 Hz). C NMR
151 MHz, CDCl , δ): 14.17, 55.08 (septet, J = 21.9 Hz),
(
6
1
3
13
0.12, 112.72, 113.79, 116.73, 125.55, 127.13, 143.65,
J = 1.5 Hz). C NMR (151 MHz, CDCl , δ): 54.99 (septet,
3
+
45.86, 147.00, 166.50. HRMS-EI (m/z): [M] calcd for
J = 22.2 Hz), 63.75, 108.30, 114.44, 120.20, 126.05, 129.17,
+
C H D O , 225.1080; found, 225.1081.
131.30, 145.51, 146.60. HRMS-FAB-NBA (m/z): [M]
12
11 3 4
calcd for C H D O , 183.0975; found, 183.0972.
10
9 3 3
2
2
.6 | (E)-3-(4-Hydroxy-3-[ H ]
3
2
methoxyphenyl)acrylic acid (2b)
2.8 | (E)-3-(4-Hydroxy-3-[ H ]
3
methoxyphenyl)prop-2-enal (14)
A mixture of 11 (660 mg, 2.93 mmol), EtOH (4.0 ml), and
an aqueous NaOH solution (10% w/v, 4.0 ml) was stirred
for 2 h at room temperature. Additional aqueous NaOH
solution (10% w/v, 8.0 ml) was added to the mixture, and
the reaction mixture was stirred for a further 5 h at room
temperature. The mixture was acidified (pH = 1) with
DDQ (1.00 g, 4.41 mmol) was added portionwise to an ice-
cold solution of 13 (734 mg, 4.01 mmol) in anhydrous
1,4-dioxane (40 ml), and the mixture was stirred for
ꢁ
30 min at 0 C. After a further 2.5 h of stirring at room tem-
perature, the insoluble materials were filtered off, and the
filtrate was concentrated in vacuo. The residue was puri-
fied by flash column chromatography on silica gel (hex-
ane/EtOAc = 65/35 ! 50/50, v/v) to give a yellow oil,
which was dissolved in EtOAc (100 ml). The organic layer
was washed successively with a saturated aqueous
2
M HCl before being poured onto water (50 ml). The
precipitate was collected using suction filtration, washed
successively with water and hexane, and dried in vacuo
to give 2b (393 mg, 68%) as a pale-yellow solid. mp 172–
ꢁ
1
1
73 C. H NMR (600 MHz, CD SOCD , δ): 6.37 (1H, d,
3
3

WATANABE ET AL.
407
solution of NaHCO and brine (40 ml each), dried over
7.51 (1H, d, J = 15.8 Hz), 8.92 (1H, br s), 12.16 (1H, br s).
3
2
13
anhydrous MgSO and filtered, before the filtrate was con-
H NMR (92 MHz, CH SOCH , δ): 3.76 (br s). C NMR
4
3
3
centrated in vacuo to give 14 (397 mg, 55%) as a pale-
(151 MHz, CD SOCD , δ): 55.50 (2C, septet, J = 21.9 Hz),
3
3
ꢁ
1
yellow solid. mp 80–82 C. H NMR (600 MHz, CDCl , δ):
106.27 (2C), 116.30, 124.86, 138.29, 145.10, 148.28 (2C),
3
+
6
.28 (1H, br s), 6.60 (1H, dd, J = 15.8, 7.7 Hz), 6.96 (1H, d,
168.24. HRMS-EI (m/z): [M] calcd for C H D O ,
11
6 6 5
J = 8.2 Hz), 7.06 (1H, d, J = 2.1 Hz), 7.12 (1H, dd, J = 8.2,
230.1061; found, 230.1061.
2
.1 Hz), 7.41 (1H, d, J = 15.8 Hz), 9.64 (1H, d, J = 7.7 Hz).
1
3
C NMR (151 MHz, CDCl , δ): 55.12 (septet, J = 22.2 Hz),
3
2
1
1
09.47, 114.93, 123.99, 126.25, 126.54, 146.95, 148.97,
53.23, 193.70. HRMS-FAB-NBA (m/z): [M + H] calcd
2.11 | 4-Bromo[2,3,5,6- H ]phenol (18)
4
+
for C H D O , 182.0896; found, 182.0891.
Tetrabutylammonium tribromide (25.0 g, 51.8 mmol) was
added portionwise to a solution of 17 (5.10 g, 50.9 mmol)
in MeOH (200 ml) and CH Cl (300 ml) at room tempera-
ture. The mixture was stirred for a further 30 min at room
temperature and then concentrated in vacuo. Water
10
8 3 3
2
2
2
2
.9 | 4-Hydroxy-3,5-di[ H ]
3
methoxybenzaldehyde (16)
(
150 ml) was added to the residue, and the aqueous layer
CD OD (8.5 ml) was removed from a solution of CD ONa
was extracted with toluene (1 ꢂ 100 ml, 2 ꢂ 50 ml). The
combined organic layers were washed successively with
water (100 ml) and brine (50 ml), dried over anhydrous
3
3
(
(
prepared from sodium (1.15 g, 50.0 mmol) and CD OD
25 ml)) under reflux with stirring, and the solution was
3
cooled to room temperature. A mixture of 15 (1.77 g,
.32 mmol) and CuCl (680 mg, 5.06 mmol) in anhydrous
MgSO and filtered, before the filtrate was concentrated in
4
6
vacuo. The residue was purified by flash column chroma-
tography on silica gel (hexane/EtOAc = 90/10 ! 60/40,
v/v) to give 18 (6.29 g, 70%) as a pale-yellow oil, which
2
DMF (12.5 ml) was added to the solution. The reaction
ꢁ
mixture was stirred for 1 h at 110 C with the removal of
ꢁ
13
CD OD (13 ml) and cooled to room temperature. The
solidified during storage in a refrigerator. mp 61–65 C.
C
3
mixture was diluted with water (50 ml), acidified
NMR (151 MHz, CDCl , δ): 112.64, 116.81 (2C, t,
3
ꢁ
(
pH = 1) at 0 C with concentrated HCl, and extracted
J = 24.6 Hz), 132.06 (2C, t, J = 25.3 Hz), 154.46.
with EtOAc (3 ꢂ 50 ml). The combined organic layers
were washed with brine (40 ml), dried over anhydrous
2
MgSO and filtered, before the filtrate was concentrated
2.12 | Ethyl (E)-3-(4-hydroxy[2,3,5,6- H ]
phenyl)[2- H]acrylate (19 )
4
4
2
0
in vacuo. The residue was purified by flash
column chromatography on silica gel (hexane/
EtOAc = 60/40 ! 30/70, v/v) to give 16 (1.03 g, 87%) as
2.12.1 | Mizoroki–Heck reaction
ꢁ
1
a colorless solid. mp 112–113 C. H NMR (600 MHz,
13
CDCl , δ): 6.05 (1H, s), 7.15 (2H, s), 9.82 (1H, s).
C
Under a nitrogen atmosphere, a mixture of 18 (5.92 g,
33.4 mmol), tri(o-tolyl)phosphine (2.00 g, 6.57 mmol),
3
NMR (151 MHz, CDCl3, δ): 55.56 (2C, septet,
J = 22.2 Hz), 106.59 (2C), 128.25, 140.79, 147.28 (2C),
ethyl acrylate (7.00 g, 69.9 mmol), Pd (OAc) (250 mg,
2
+
1
90.72. HRMS-FAB-NBA (m/z): [M + H] calcd for
1.11 mmol), and Et N (35 ml) was stirred for 23 h under
3
C H D O , 189.1034; found, 189.1034.
reflux. The reaction mixture was cooled to room tempera-
ture and concentrated in vacuo. The residue was
suspended in CH Cl , and AcOH (10 ml) was added to
9
5 6 4
2
2
2
2.10 | (E)-3-(4-Hydroxy-3,5-di[ H ]
the suspension. This mixture was subjected to
flash column chromatography on silica gel (hexane/
EtOAc = 80/20 ! 15/85, v/v) to give 19 as a pale-yellow
3
methoxyphenyl)acrylic acid (3b)
1
A mixture of 16 (988 mg, 5.25 mmol), malonic acid
solid (6.53 g, quant). From its H NMR spectrum, the
(
1.20 g, 11.5 mmol), and aniline (64.7 mg, 0.695 mmol) in
deuterium content at the ortho position of the phenolic
hydroxy group and the C-2 position of 19 was estimated
to be 55% and 18%, respectively.
ꢁ
anhydrous pyridine (2.7 ml) was stirred for 20 h at 60 C
and then cooled to room temperature. Crushed ice (5.6 g)
and concentrated HCl (3.3 ml) were added successively
to the reaction mixture. After 30 min of stirring at room
temperature, the precipitate was collected using suction
filtration and dried in vacuo to give 3b (994 mg, 82%) as a
2.12.2 | Re-incorporation of deuterium
ꢁ
1
pale-yellow solid. mp 194–195 C. H NMR (600 MHz,
CD SOCD , δ): 6.43 (1H, d, J = 15.8 Hz), 6.99 (2H, s),
19 (400 mg, 2.05 mmol) was transferred to a pressure
tube, the interior of which was flushed with argon.
3
3

408
WATANABE ET AL.
EtOD (12 ml, 0.21 mol) and Et N (0.60 ml, 4.3 mmol)
2.14 | General procedure for the
3
were added to the tube, and the upper space of the
tube was flushed with argon again before the tube was
plugged. The plugged tube was heated at 100 C for
hydrogen–deuterium exchange reaction in
the EtOD/Et N system
3
ꢁ
7
2 h with stirring. The mixture was cooled to room
In the cases of 11a, 12, 20a, and 22a–31a, which contain
one phenolic hydroxy group, the substrate (1.35–
1.44 mmol) was transferred to a pressure tube, the inte-
rior of which was flushed with argon. EtOD (8.0 ml,
temperature, concentrated in vacuo, and purified by
flash column chromatography on silica gel (hexane/
EtOAc = 80/20 ! 20/80, v/v) to give a colorless solid
(
(
390 mg, 97%). Part of the product from the first cycle
384 mg, 1.95 mmol) was treated and purified as
0.14 mol, ca. 100 equiv.) and Et N (0.40–0.42 ml, 2.9–
3
3.0 mmol, ca. 2.1 equiv.) was added to the tube, and the
upper space of the tube was flushed with argon again
before the tube was plugged. The plugged tube was
described above to give a colorless solid (382 mg, 99%).
Part of the product from the second cycle (373 mg,
1
and Et N (0.55 ml, 3.9 mmol) at 100 C for 72 h and
purified chromatographically to give 19 as a colorless
solid (368 mg, 99%). mp 75–77 C. H NMR (600 MHz,
CDCl , δ): 1.34 (3H, t, J = 7.0 Hz), 4.26 (2H, q,
ꢁ
.89 mmol) was treated with EtOD (11 ml, 0.19 mol)
heated at 100 C for 72 h with stirring. The mixture
ꢁ
was cooled to room temperature, concentrated in vacuo,
and purified by flash column chromatography on silica
gel. The deuterium content in the product was deter-
3
0
ꢁ
1
1
mined using H NMR. The same reaction conditions
3
13
J = 7.0 Hz), 5.66 (1H, br s), 7.63 (1H, s). C NMR
151 MHz, CDCl3, δ):14.20, 60.80, 114.52 (t,
J = 23.8 Hz), 115.57 (2C, t, J = 23.8 Hz), 126.34,
were applied to 6a and 37. In the cases of 21a, 33a, 34a,
and 36a, which contain two phenolic hydroxy groups,
0.670–0.700 mmol of the substrate was treated with
(
1
29.62 (2C, t, J = 23.8 Hz), 145.06, 158.39, 168.45.
approximately 3 equiv. of Et N (0.28–0.30 ml) and
3
ꢀ
1
HRMS-ESI (m/z): [M ꢀ H] calcd for C H D O ,
200 equiv. of EtOD (8.0 ml). The same conditions were
applied to 4a. In the cases of 32a and 35a, which contain
three phenolic hydroxy groups, 0.469–0.470 mmol of the
11
6 5 3
1
96.1028; found, 196.1029. H NMR analysis of the
products of each operation revealed that the deuterium
content at the ortho position of the phenolic hydroxy
group was 84% (first cycle), 94% (second cycle), and
9
9
substrate was treated with approximately 4 equiv. of Et N
3
(0.26 ml) and 300 equiv. of EtOD (8.0 ml). For detailed
conditions, see the Supporting Information.
8% (third cycle) and that at the C-2 position was
1% (first cycle), 98% (second cycle), and 98% (third
cycle).
2
.15 | Method for feeding the flower
0
petals with 1b and 4b and GC-MS analysis
2
2
.13 | (E)-3-(4-Hydroxy[2,3,5,6- H ]
of the benzenoid volatiles
4
2
phenyl)[2- H]acrylic acid (4b')
Petals of opened flowers (100 mg fresh weight) were
harvested and immersed in a 12-well cell culture plate
0
A solution of 19 (192 mg, 0.973 mmol) in EtOH
(
1.5 ml) was treated dropwise with an aqueous NaOH
(well size: 23 mm i.d. and 18 mm in height), in which 1b
0
solution (10% w/v, 4.5 ml) at room temperature. After
or 4b dissolved in an aqueous NaHCO solution (5%
3
3
h of stirring at room temperature, the mixture was
w/v) had been incorporated. The feed concentration was
50 mM and feeding was performed via vacuum infiltra-
tion with a vacuum desiccator and pump. After 5 h of
incubation, the flower petals were homogenized in liquid
nitrogen with a mortar and pestle, and the volatile com-
pounds were extracted using 3 ml of EtOAc containing
isobutyl benzene (5 μg) as an internal standard. The
ꢁ
acidified (pH = 1) with 2 M HCl at 0 C. The precipi-
tate was collected using suction filtration, washed with
ice-cold water, and dried in vacuo to give 4b (150 mg,
0
0
9
1%) as a colorless solid. The mp of 4b was difficult
to measure, as it sublimed gradually from approxi-
mately 180–219 C. H NMR (600 MHz, CD SOCD , δ):
7
OH) were not clearly observed. H NMR (92 MHz,
CH SOCH , δ): 6.31 (br s), 6.82 (br s), 7.54 (br s).
NMR
ꢁ
1
3
3
.49 (1H, br s). Active protons (CO H and phenolic
resulting volatiles were analyzed using
a
gas
2
2
chromatography–mass spectrometry (GC-MS) instrument
(QP-2010 Plus, Shimadzu Corporation, Kyoto, Japan)
equipped with a DB-5ms column (30 m ꢂ 0.25 mm
i.d. ꢂ 0.4 μm film thickness, Agilent Technologies, Inc.,
CA, USA). The column temperature was programmed as
13
C
3
3
(151 MHz,
CD SOCD ,
δ):
115.28
(t,
3
3
J = 23.8 Hz), 115.60 (2C, t, J = 23.8 Hz), 125.31,
1
29.93 (2C, t, J = 23.1 Hz), 144.28, 159.74, 168.20.
ꢀ
ꢁ
ꢀ1
ꢁ
ꢀ1
ꢁ
HRMS-ESI (m/z): [M ꢀ H]
calcd for C H D O ,
follows: 40 C for 2 min, 10 C min
ramp to 170 C,
9
2 5 3
ꢁ
ꢁ
1
68.0715; found, 168.0714.
20 C min ramp to 240 C, 10 min hold. Helium was

WATANABE ET AL.
409
used as the carrier gas. The MS was operated in the elec-
tron ionization mode at 70 eV. The ion source and inter-
face temperature were set to 240 C.
after chromatographic purification on silica gel. Such
intermediates were not isolated in the original report.
Basic hydrolysis of 6b furnished 1b in 78% yield.
3
2
ꢁ
3
3
| RESULTS AND DISCUSSION
.1 | Synthesis of deuterium-labeled
cinnamic acids
For the synthesis of cinnamic acids, Knoevenagel con-
densations between benzaldehydes and malonic acid
9
–13,25–27
have frequently been used.
Knoevenagel conden-
sations involving monoalkyl malonates have also been
9,28
employed for the synthesis of alkyl cinnamates,
in
addition to Wittig reactions between benzaldehydes and
2
9,30
stable ylides,
and Mizoroki–Heck reactions between
31
aryl halides and alkyl acrylates. In this study, we
selected previously established reaction conditions con-
sidering both the availability of deuterium sources and
that of non-deuterated starting materials.
The synthesis of 1b is illustrated in Scheme 1,
whereby bromobenzene-d (5) was used as the deuterium
5
source. In Schemes 1–4, “D” represents a deuterium con-
tent greater than 98%, as the deuterium content of the
SCHEME 2 Synthesis of 2b, 13, and 14
starting materials is ≥98%. We used a Me SiCl-mediated
3
catalytic carbocupration, considering that the reaction
between PhMgBr and ethyl propiolate under the reported
conditions furnishes ethyl cinnamate in high yield with
32
E-selectivity (91%, E/Z = 95/5). In our experiment, the
desired product 6b and its Z-isomer 7 were obtained in
5
9% and 23% yield, respectively. Allene 8 (7.2%), the
intermediate in the carbocupration, was also obtained
SCHEME 3 Synthesis of 3b
0
SCHEME 1 Synthesis of 1b
SCHEME 4 Synthesis of deuterium-labeled 4-coumaric acid 4b

410
WATANABE ET AL.
The synthesis of 2b is illustrated in Scheme 2,
whereby CD I was used as the deuterium source. As
18 and 19. In both cases, Et N is considered to promote
the exchange reaction.
3
3
11,33
described in the literature,
the 3-hydroxy group of
Therefore, the re-incorporation of the deuterium
3
,4-dihydroxybenzaldehyde (9) was regioselectively
atoms on 19 was carried out using ethanol-d (EtOD,
1
alkylated to produce 10 in 77% yield. We used a Wittig
reaction for the construction of the cinnamate
skeleton, because the reaction between 4-hydroxy-
ca. 100 equiv.) as the deuterium source. EtOD has been
used for the deuteration of the methyl ketone moiety of a
steroidal compound in the presence of catalytic amount
37
3
-methoxybenzaldehyde and (carbethoxymethylene)
of PhLi. The reaction was carried out in the presence of
ꢁ
triphenylphosphorane has been reported to provide ethyl
Et N (ca. 2.1 equiv.) at 100 C using a pressure tube. The
3
29
ferulate in 95% yield. In our reaction, the Wittig reac-
tion between 10 and the stable ylide provided 11b and its
Z-isomer 12 in 85% and 15% yield, respectively. Basic
hydrolysis of 11b furnished 2b in 68% yield. Employing
progress of the reaction was monitored after 24, 48, and
1
72 h using H NMR. The deuterium content at the ortho
position of the phenolic hydroxy group reached its maxi-
mum value after 72 h, whereas that at the C-2 position
reached its maximum after 24 h. The reaction did not
proceed at room temperature, and only the hydrogen
atoms at the C-2 position were slowly exchanged by deu-
30
the reported procedure, 11b was also reduced with
diisobutylaluminum hydride to give 13 in 92% yield,
which was oxidized with DDQ to afford 14 in 55% yield
ꢁ
(
Scheme 2).
The synthesis of 3b is illustrated in Scheme 3,
whereby CD OD was used as the deuterium source.
terium atoms at 70 C. The use of CD CO D instead of
3
2
Et N did not affect the deuterium content of 19.
3
After 72 h of reaction, the ratios of deuterium atoms
at the ortho position of the phenolic hydroxy group and
the C-2 position were 84% and 91%, respectively. When
the reaction product was subjected to the second opera-
tion, the deuterium content at the former and latter posi-
tions increased to 94% and 98%, and that at the former
also reached 98% after the third operation. The yield of
each reaction was nearly quantitative (97%, 99%, 99%).
3
34
The procedure of Rao and Stuber was applied to
,5-dibromo-4-hydroxybenzaldehyde (15) to produce 16
in 87% yield. Lan and coworkers have synthesized
6 from 4-hydroxy-3,5-diiodobenzaldehyde in 76% yield
3
1
35
under similar conditions. We used a Knoevenagel
condensation in the next step, because the reaction
between 4-hydroxy-3,5-dimethoxybenzaldehyde and
malonic acid in the presence of aniline has been
0
The thus-obtained 19 was hydrolyzed under basic condi-
27
0
reported to afford sinapic acid in 86% yield. Using
tions to give 4b in 91% yield. No loss of deuterium atoms
these reaction conditions, 16 was converted to 3b in
was observed in the basic hydrolysis. Recently, Chen and
coworkers prepared 4b using a Mizoroki–Heck reaction
between 18 and acrylic acid in an aqueous K CO solu-
tion. They reported neither the loss of deuterium atoms
at the ortho position of the phenolic hydroxy group nor
the incorporation of deuterium at the C-2 position of 4b.
8
2% yield.
Our attempt to synthesize 4b is illustrated in
2
3
38
Scheme 4, whereby phenol-d (17) was used as the deute-
6
rium source. We used a Mizoroki–Heck reaction for the
construction of the cinnamate skeleton, because the reac-
tion between 4-bromophenol and methyl acrylate in the
presence of Pd(OAc) and P(o-tol) in refluxing Et N has
2
3
3
been reported to produce methyl 4-coumarate in 90%
3.2 | Hydrogen–deuterium exchange
31
yield.
Therefore, 17 was first brominated with
reaction of phenolic compounds in the
36
tetrabutylammonium tribromide to provide 18 in 70%
yield and then subjected to a Mizoroki–Heck reaction
with ethyl acrylate using the reported conditions. This
EtOD/Et N system
3
31
To assess the scope and limitations of the hydrogen–
deuterium exchange reaction, we investigated a series of
alkyl cinnamates and other phenolic compounds as sub-
strates. In this study, approximately 100 equiv. of EtOD
reaction quantitatively provided deuterium-labeled ethyl
4
-coumarate 19; however, approximately half the deute-
rium atoms at the ortho position of the phenolic hydroxy
group proved to be substituted with hydrogen atoms
based on the H NMR analysis. Interestingly, ca. 20% of
and 2.1 equiv. of Et N per equiv. of substrates bearing
3
1
one phenolic hydroxy group were employed; the ratio of
reagents per substrate was increased in the cases of
polyphenolic compounds (for detailed conditions, see
Section 2 and the Supporting Information). The results
the hydrogen atoms at the C-2 position of 19 were
exchanged for deuterium atoms. Because the Mizoroki–
Heck reaction was not carried out under strict anhydrous
conditions, hydrogen atoms from water in the reaction
system may have been exchanged with the deuterium
atoms of 18 and 19. Another possibility is the migration
of the hydrogen atoms on the phenolic hydroxy groups of
ꢁ
after a single operation at 100 C for 72 h are summarized
in Figure 2.
Except in the case of free carboxylic acid 4a, which
decomposed under the reaction conditions, hydrogen–

WATANABE ET AL.
411
3
FIGURE 2 Deuterium incorporation ratios of phenolic compounds in the EtOD/Et N system. The numbers in italics in each structure
represent the percentage of deuterium atom incorporation at the designated positions. The number in parentheses represents the yield of the
reaction after chromatographic purification
deuterium exchange occurred at the ortho positions of
the phenolic hydroxy groups of all the compounds tested,
although the meta positions were totally inactive.
Hydrogen–deuterium exchange also occurred at the C-2
position of alkyl cinnamates 20a, 21a, 11a, 22a, and 23a,
as well as the para positions of the phenolic hydroxy
groups of 21a, 23a, and 24a. Although the C-2 position of
nor deuterium contents were reported, and a systematic
investigation into this type of hydrogen–deuterium
exchange reaction was not carried out.
In the reaction, the Z-double bond of 12 was
completely isomerized to the E configuration to give
0
0
11b , whereas the E/Z ratio of isoeugenol (30a) did not
change during the reaction. A small amount (ca. 5%–8%)
of the olefinic hydrogen atoms were exchanged for deute-
rium atoms in the cases of 30a and 31a. In contrast, no
deuteration was observed at the olefinic position of res-
veratrol (32a). In addition, as seen in the cases of
geranylhydroquinone (33a) and biochanin A (34a), the
olefinic positions without conjugation with the anions of
the phenolic hydroxy groups are also inactive. Kirby and
Ogunkoya have reported the deuteration of the ortho
position of the hydroxy group of 30a under basic condi-
2
4a was inactive, the corresponding position was deuter-
ated in 25a–27a. Considering these results, it is obvious
that the deuterated positions can resonate with the anion
of the phenolic hydroxy groups. As seen in the cases of
2
8a and 29a, the hydrogen–deuterium exchange reaction
proceeds regardless of the presence or absence of an
electron-withdrawing group on a benzene ring. Roughley
and Whiting have reported the deuteration of 11a and
(
E)-4-(4-hydroxy-3-methoxyphenyl)but-3-en-2-one,
in
ꢁ
39
which a methoxy group is introduced to the benzene ring
tions (0.5 equiv. of Et N or NaOD in D O at 100 C), but
3
2
of 25a, in D O (ca. 50 equiv.)/1,4-dioxane in the presence
detailed results were not described.
2
ꢁ
25
of Et N (1 equiv.) at 100 C for 3 days. They observed
As seen in the cases of 25a and naringenin (35a), the
hydrogen atoms of methyl or methylene groups adjacent
to a ketone carbonyl group were exchanged with deute-
rium atoms, but the less-acidic hydrogen atoms at the
α-position of the ester carbonyl group of 29a were
3
the incorporation of the deuterium atoms at the ortho
positions of their phenolic hydroxy groups as well as the
C-2 position of 11a and the corresponding position of
the latter compound. However, neither chemical yield

412
WATANABE ET AL.
inactive. Curcumin (36a), which has active hydrogen
atoms between the two carbonyl groups, was unstable,
probably due to aldol and/or Michael addition reactions,
and the yield of the deuterium-labeled product 36b was
seen in the case of 19. In addition, not only alkyl
cinnamates, but also a variety of phenolic compounds,
are suitable substrates of the reaction. Furthermore,
EtOD is a readily available, inexpensive, and easy-
to-handle reagent, and can dissolve a wide range of
1
0%, even though the hydrogen–deuterium exchange
proceeded efficiently. In the cases of 6a and 37, in which
a free phenolic hydroxy group is not present, the
hydrogen–deuterium exchange did not proceed. Despite
the presence of a large excess of EtOD, the extent of the
ester exchange reaction (methyl to ethyl) was relatively
small (8%–20%), as seen in the cases of 20b, 21b, and 29b.
substrates compared with D O. Therefore, the hydrogen–
2
deuterium exchange reaction of phenolic compounds in
the EtOD/Et N system should provide a useful alterna-
3
tive to previously reported methods.
D O has been frequently used in combination with
3.3 | Tracking the biosynthesis of
benzenoid volatiles in the flowers of
E. japonica
2
acid catalysts for the deuteration of benzene
12,26,35,40
rings.
Recently, Sajiki and coworkers have devel-
oped a hydrogen–deuterium exchange reaction using
D O as a deuterium source in the presence of platinum-
With the desired deuterium-labeled cinnamic acids in
hand, we investigated the biosynthetic pathway of benze-
noid volatiles in the flowers of the Japanese loquat
E. japonica. Kuwahara and coworkers have reported that
(2-nitroethyl)benzene, 4-methoxybenzaldehyde (38a),
and methyl 4-methoxybenzoate (39a) are the major com-
2
41–43
group catalysts.
This method can deuterate not only
arenes but also saturated aliphatic compounds. In con-
trast to these methods, to our best of knowledge, only a
few examples of base-catalyzed hydrogen–deuterium
exchange on the benzene ring of phenolic compounds
25,39
15
have been reported.
However, Zhan and co-workers
ponents of the flower scent of E. japonica. In their
have recently developed a method for the deuteration of
study, 4-methoxybenzyl alcohol (40a) was detected as a
minor component, and the benzenoids 38a–40a were
phenols in D O in the presence of NaOH (0.5 equiv.)
2
44
45
under microwave irradiation. Under these basic condi-
tions, as seen in this study, the ortho and/or para posi-
tions of phenolic hydroxy groups are regioselectively
deuterated. As summarized in Figure 2, the yield of the
hydrogen–deuterium exchange reaction in the EtOD/
speculated to be biosynthesized from tyrosine. In their
proposed pathway, the involvement of cinnamic acids 1a
and 4a was not considered. To assess this point, we
0
administered deuterium-labeled 1b and 4b to the flowers
of E. japonica and analyzed the profile of the benzenoid
volatiles after 5 h of incubation. The results using 1b are
shown in Figure 3.
Et N system is satisfactory (72%–98%), and the deuterium
3
content can be increased by repeating the reaction, as
FIGURE 3 GC-MS analysis of the
benzenoid volatiles in the flowers of Eriobotrya
japonica after 5 h incubation with 1b. (A) Total
ion chromatogram of EtOAc extracts of the
loquat flowers. Deuterium-labeled volatiles are
indicated by asterisks. (B) MS profiles obtained
from the peaks corresponding to 38 (left), 40
(
middle), and 39 (right)

WATANABE ET AL.
413
As shown in Figure 3, the conversion of the adminis-
4a. Next, the acrylic side chain of 4a would be shortened
47
tered 1b to deuterium-labeled 38b–40b was observed. In
our experiment, not only 38a and 39a but also 40a, were
detected as the major endogenous components in the
by 4-hydroxybenzaldehyde synthase
to produce
4-hydroxybenzaldehyde (41). The catalytic activity of the
enzyme has been observed in tissue cultures of
47
scent of E. japonica flowers. Similarly, the formation of
the vanilla orchid Vanilla planifolia. The subsequent
methylation of the phenolic hydroxy group of 41 would
be catalyzed by an O-methyltransferase that we have pre-
0
3
(
8b–40b was observed after the administration of 4b
Supporting Information). Because 1a is biosynthesized
4
8
from phenylalanine, and the activity of tyrosine ammonia
viously characterized from the flowers of E. japonica.
lyase, which converts tyrosine to 4a, is generally limited
The successive reduction of the formyl group of 38a pro-
vides 40a (Scheme 5). Because both benzaldehyde 38b
and benzyl alcohol 40b, which bear a methoxy group
on the benzene ring, were detected as the metabolites of
administered 1b and 4b , the phenolic hydroxy group
of 41 is believed to be methylated prior to the reduction
of the formyl group.
46
in Poaceae (monocot grass family), our results indicate
that the biosynthetic precursor of benzenoids 38a–40a in
the flowers of E. japonica is phenylalanine but not tyro-
sine. Accordingly, we propose a putative biosynthetic
route for benzenoids 38a–40a in the flowers of
E. japonica as illustrated in Scheme 5.
0
As shown in Scheme 5, the first two steps of the bio-
synthesis of benzenoids 38a–40a in the flowers of
E. japonica are believed to be mediated by the well-
known enzymes phenylalanine ammonia lyase and
cinnamate 4-hydroxylase. The former converts phenylala-
nine to 1a, whereas the latter subsequently oxidizes 1a to
In the route leading to 39a, the formyl group of 41
would be oxidized to yield 4-hydroxybenzoic acid (42)
(Scheme 5). Although 4-hydroxybenzaldehyde dehydro-
genase, which catalyzes this step, has not been identified
in plants, enzymatic activities that transform 4a to 42 via
41 have been reported in cell-free extracts from the tuber
49
of the potato Solanum tuberosum as well as cell cultures
50
of Lithospermum erythrorhizon and the carrot Daucus
51,52
carota.
The authors did not note whether the trans-
formation was catalyzed by a single enzyme or not. How-
ever, the above-mentioned 4-hydroxybenzaldehyde
47
synthase does not produce 42, and cell-free extracts
generally contain a number of enzymes. Therefore, the
transformation of 4a to 42 via 41 in the flowers of
E. japonica would be mediated by two distinct enzymes,
that is, 4-hydroxybenzaldehyde synthase, which converts
4a to 41, and 4-hydroxybenzaldehyde dehydrogenase,
which oxidizes 41 to 42. It is reported that cell-free
extracts from L. erythrorhizon and D. carota recognize 41
as the best substrate in the benzaldehyde oxidation
50,53
step.
The subsequent methylation of the phenolic hydroxy
48
group of 42 by the O-methyltransferase
provides
4-methoxybenzoic acid (43), followed by the esterification
of the carboxy group of 43, which affords 39a (Scheme 5).
Recently, we characterized 4-methoxybenzoic acid car-
boxyl methyltransferase from E. japonica, which strongly
5
4
prefers 43 as a substrate and gives 39a as a product.
Therefore, the methylation of the phenolic hydroxy group
of 42 would occur in advance of the esterification of the
carboxylic group, as depicted in Scheme 5.
SCHEME 5 Putative biosynthetic route of benzenoids 38a–40a
in the flowers of Eriobotrya japonica. The deuterium-free structures
of the administered cinnamic acids and their metabolites are
enclosed in rounded rectangles and shadowed rounded rectangles,
respectively. The solid arrows represent reactions catalyzed by
identified enzymes in the plants. The dashed arrows represent the
reactions mediated by unidentified enzymes
4
| CONCLUSIONS
In this study, deuterium-labeled cinnamic acids 1b–3b
and 4b were synthesized using readily available deute-
rium sources. We also developed a hydrogen–deuterium
0

414
WATANABE ET AL.
exchange reaction in an EtOD/Et N system, which can
introduce deuterium atoms at the ortho and the para
positions of the phenolic hydroxy groups of (poly)phenols
as well as at the C-2 position of alkyl cinnamates. Using
7. Dixon RA, Achnine L, Kota P, Liu C-J, Reddy MSS, Wang L.
The phenylpropanoid pathway and plant defence—a genomics
perspective. Mol Plant Pathol. 2002;3(5):371-390.
3
8
. Vogt T. Phenylpropanoid biosynthesis. Mol Plant. 2010;3(1):
-20.
2
0
1
3
b and 4b , we demonstrated that the benzenoid volatiles
8a–40a in the scent of E. japonica flowers would be bio-
9
. Sakakibara N, Nakatsubo T, Suzuki S, Shibata D, Shimada M,
Umezawa T. Metabolic analysis of the cinnamate/monolignol
pathway in Carthamus tinctorius seeds by a stable-isotope-
dilution method. Org Biomol Chem. 2007;5(5):802-815.
synthesized from phenylalanine via 1a and 4a.
ACKNOWLEDGMENTS
10. Bayoumi SAL, Rowan MG, Blagbrough IS, Beeching JR. Bio-
synthesis of scopoletin and scopolin in cassava roots during
post-harvest physiological deterioration: the E-Z-isomerisation
stage. Phytochemistry. 2008;69(17):2928-2936.
This work was supported in part by a Grant-in-Aid for
Scientific Research from the Japan Society for the Promo-
tion of Science (JSPS; 20K05840 to T.K.). Some of the
experimental measurements were carried out using
the JEOL JNM-ECA600 and the Bruker AVANCEIII
1
1. Lairez D, Cathala B, Monties B, Bedos-Belval F, Duran H,
Gorrichon L. Aggregation during coniferyl alcohol polymeriza-
tion in pectin solution: a biomimetic approach of the first steps
of lignification. Biomacromolecules. 2005;6(2):763-774.
6
00 NMR spectrometers as well as the JEOL JMS-700 and
the Bruker timsTOF mass spectrometers at the Joint
Usage/Research Center (JURC) at the Institute for Chem-
ical Research (ICR) of Kyoto University. The authors are
greatly indebted to Ms. Akiko Fujihashi and Ms. Kyoko
Ohmine (Kyoto University) for performing instrumental
analyses, and to Dr. Tomohiko Tsuge (Kyoto University)
for providing a picture of E. japonica flowers.
1
2. Waterstraat M, Hildebrand A, Rosler M, Bunzel M. Develop-
ment of a QuEChERS-based stable-isotope dilution LC-MS/MS
method to quantitate ferulic acid and its main microbial and
hepatic metabolites in milk. J Agric Food Chem. 2016;64(45):
8667-8677.
3. Piber M, Koehler P. Identification of dehydro-ferulic acid-
tyrosine in rye and wheat: evidence for a covalent cross-link
between arabinoxylans and proteins. J Agric Food Chem. 2005;
1
5
3(13):5276-5284.
CONFLICT OF INTEREST
The authors declare the absence of any potential conflicts
of interest.
1
4. Sarkissian CN, Scriver CR, Mamer OA. Quantitation of phenyl-
alanine and its trans-cinnamic, benzoic and hippuric acid
metabolites in biological fluids in a single GC-MS analysis.
J Mass Spectrom. 2007;42(6):811-817.
DATA AVAILABILITY STATEMENT
The data that supports the findings of this study are avail-
able in the supplementary material of this article
15. Kuwahara Y, Ichiki Y, Morita M, Asano Y. (2-Nitroethyl)ben-
zene: a major flower scent from the Japanese loquat Eriobotrya
japonica [Rosales: Rosaceae]. Biosci Biotechnol Biochem. 2014;
78(8):1320-1323.
1
6. Gaspar A, Martins M, Silva P, et al. Dietary phenolic acids and
derivatives. Evaluation of the antioxidant activity of sinapic
acid and its alkyl esters. J Agric Food Chem. 2010;58(21):
ORCID
Bunta Watanabe https://orcid.org/0000-0003-3645-
5
712
11273-11280.
1
7. Zhang B, Lv C, Li W, et al. Ethyl cinnamate derivatives as
promising high-efficient acaricides against Psoroptes cuniculi:
synthesis, bioactivity and structure–activity relationship. Chem
Pharm Bull. 2015;63(4):255-262.
REFERENCES
1
. El-Seedi HR, El-Said AMA, Khalifa SAM, et al. Biosynthesis,
natural sources, dietary intake, pharmacokinetic properties,
and biological activities of hydroxycinnamic acids. J Agric Food
Chem. 2012;60(44):10877-10895.
1
8. Paul S, Gupta M. A simple and efficient method for selective
single aldol condensation between arylaldehydes and acetone.
Synth Commun. 2005;35(2):213-222.
2
. Abotaleb M, Liskova A, Kubatka P, Büsselberg D. Therapeutic
potential of plant phenolic acids in the treatment of cancer.
Biomolecules. 2020;10(2):221-221.
. Pei K, Ou J, Huang J, Ou S. p-Coumaric acid and its conju-
gates: dietary sources, pharmacokinetic properties and biologi-
cal activities. J Sci Food Agric. 2016;96(9):2952-2962.
1
9. Pizzirani D, Roberti M, Grimaudo S, et al. Identification of
biphenyl-based hybrid molecules able to decrease the intracel-
lular level of Bcl-2 protein in Bcl-2 overexpressing leukemia
cells. J Med Chem. 2009;52(21):6936-6940.
3
20. Bernal FA, Gerhards M, Kaiser M, Wünsch B, Schmidt TJ.
(±)-trans-2-Phenyl-2,3-dihydrobenzofurans as leishmanicidal
agents: synthesis, in vitro evaluation and SAR analysis. Eur J
Med Chem. 2020;205:112493.
21. Baeza E, Catal ꢀa n K, Peña-Cortés H, Espinoza L, Villena J,
Carrasco H. Synthesis of geranylhydroquinone derivatives
with potencial cytotoxic activity. Quim Nova. 2012;35(3):
523-526.
4
5
. Coman V, Vodnar DC. Hydroxycinnamic acids and human
health: recent advances. J Sci Food Agric. 2020;100(2):483-499.
. Guzman JD. Natural cinnamic acids, synthetic derivatives and
hybrids with antimicrobial activity. Molecules. 2014;19(12):
1
9292-19349.
6
. Diniz LRL, de Santana Souza MT, Barboza JN, de Almeida RN,
de Sousa DP. Antidepressant potential of cinnamic acids:
mechanisms of action and perspectives in drug development.
Molecules. 2019;24(24):4469.
22. Oshikiri H, Watanabe B, Yamamoto H, Yazaki K, Takanashi K.
Two BAHD acyltransferases catalyze the last step in the

WATANABE ET AL.
415
shikonin/alkannin biosynthetic pathway. Plant Physiol. 2020;
84(2):753-761.
3. Baracchi A, Chimichi S, De Sio F, et al. Preparation of O-
bilabelled carboxylic acids: cinnahic and phenylpropiolic acids
via styrylisoxazoles. J Label Compd Radiopharm. 1986;23(5):
isotope deuterium-labeling coupled with RRLC-MS/MS
method. Molecules. 2019;24(23):4287.
39. Kirby GW, Ogunkoya L. Deuterium and tritium exchange reac-
tions of phenols and the synthesis of labelled
1
1
8
2
3,4-dihydroxyphenylalanines.
6914-6922.
J
Chem Soc. 1965;1965:
4
87-493.
2
2
2
4. Kim SW, Park JH, Jung SJ, et al. Synthesis and elucidation of
deuterated vanillylamine hydrochloride and capsaicin. J Label
Compd Radiopharm. 2009;52(13):563-565.
40. Pająk M, Pałka K, Winnicka E, Ka nꢀ ska M. The chemo-
enzymatic synthesis of labeled L-amino acids and some of
their derivatives.
643-666.
J Radioanal Nucl Chem. 2018;317(2):
5. Roughley PJ, Whiting DA. Experiments in the biosynthesis
of curcumin.
379-2388.
J
Chem Soc, Perkin Trans 1. 1973;1973:
41. Ito N, Esaki H, Maesawa T, Imamiya E, Maegawa T,
Sajiki H. Efficient and selective Pt/C-catalyzed H–D exchange
reaction of aromatic rings. Bull Chem Soc Jpn. 2008;81(2):
278-286.
42. Sawama Y, Monguchi Y, Sajiki H. Efficient H–D exchange
reactions using heterogeneous platinum-group metal on
2
6. Angelini G, Speranza M, Felici M, Luna M. Preparation of
substituted cinnamic acids labelled with deuterium and tritium
at the ring positions. J Radioanal Nucl Chem. 1984;87(5):
2
83-289.
2
2
7. Lancefield CS, Westwood NJ. The synthesis and analysis of
carbon–H
2
2
–D O system. Synlett. 2012;23(7):959-972.
advanced lignin model polymers. Green Chem. 2015;17(11):
43. Sawama Y, Park K, Yamada T, Sajiki H. New gateways to the
platinum group metal-catalyzed direct deuterium-labeling
method utilizing hydrogen as a catalyst activator. Chem Pharm
Bull. 2018;66(1):21-28.
44. Zhan M, Xu R, Tian Y, et al. A simple and cost-effective
method for the regioselective deuteration of phenols. Eur J Org
Chem. 2015;2015(15):3370-3373.
45. Kuwahara Y, Asano Y. Generation of (2-nitroethyl)benzene
and related benzenoids from L-phenylalanine; flower scents of
the Japanese loquat Eriobotrya japonica [Rosales: Rosaceae].
Biosci Biotechnol Biochem. 2018;82(11):1855-1866.
46. Barros J, Dixon RA. Plant phenylalanine/tyrosine ammonia-
lyases. Trends Plant Sci. 2020;25(1):66-79.
4
980-4990.
8. Umezawa T, Davin LB, Lewis NG. Formation of lignans
ꢀ)-secoisolariciresinol and (ꢀ)-matairesinol with Forsythia
intermedia cell-free extracts. Biol Chem. 1991;266(16):
0210-10217.
(
J
1
2
9. Ganesh KR, Rao KR, Pratap TV, et al. First stereo selective syn-
thesis of 5-O-feruloyl-2-deoxy-D-ribono-γ-lactone. Tetrahedron
Lett. 2016;57(42):4734-4736.
0. Konr ꢀa dov ꢀa D, Kozubíkov ꢀa H, Doležal K, Pospíšil J. Micro-
wave-assisted synthesis of phenylpropanoids and coumarins:
total synthesis of osthol. Eur J Org Chem. 2017;2017(35):
3
5
204-5213.
3
3
1. Quitschau M, Schuhmann T, Piel J, von Zezschwitz P,
Grond S. The new metabolite (S)-cinnamoylphosphoramide
from Streptomyces sp. and its total synthesis. Eur J Org Chem.
47. Podstolski A, Havkin-Frenkel D, Malinowski J, Blount JW,
Kourteva G, Dixon RA. Unusual 4-hydroxybenzaldehyde
synthase activity from tissue cultures of the vanilla orchid
Vanilla planifolia. Phytochemistry. 2002;61(6):611-620.
48. Koeduka T, Kajiyama M, Furuta T, Suzuki H, Tsuge T,
Matsui K. Characterization of an O-methyltransferase
specific to guaiacol-type benzenoids from the flowers of
loquat (Eriobotrya japonica). J Biosci Bioeng. 2016;122(6):
679-684.
2
008;2008(30):5117-5124.
2. Jennings MP, Sawant KB. TMSCl -mediated catalytic car-
bocupration of alkynoates: an unprecedented and remarkable
effect of catalyst loading on highly selective stereochemical
induction via a TMS-allenoate intermediate. Eur J Org Chem.
2
004;2004(15):3201-3204.
3
3. Kessar SV, Gupta YP, Mohammad T, Goyal M,
49. French CJ, Vance CP, Towers GHN. Conversion of p-coumaric
acid to p-hydroxybenzoic acid by cell free extracts of potato
tubers and Polyporus hispidus. Phytochemistry. 1976;15(4):
564-566.
50. Yazaki K, Heide L, Tabata M. Formation of p-hydroxybenzoic
acid from p-coumaric acid by cell free extract of Lithospermum
erythrorhizon cell cultures. Phytochemistry. 1991;30(7):
2233-2236.
51. Schnitzler J-P, Madlung J, Rose A, Seitz HU. Biosynthesis of p-
hydroxybenzoic acid in elicitor-treated carrot cell cultures.
Planta. 1992;188(4):594-600.
52. Sircar D, Mitra A. Evidence for p-hydroxybenzoate formation
involving enzymatic phenylpropanoid side-chain cleavage in
hairy roots of Daucus carota. J Plant Physiol. 2008;165(4):
407-414.
Sawal KK. Regioselective mono-O-alkylation of same
pyrocatechoxide dianions.
983;1983(7):400-401.
4. Rao DV, Stuber FA. An efficient synthesis of
J Chem Soc, Chem Commun.
1
3
3
3
3
1
,4,5-trimethoxybenzaldehyde from vanillin. Synthesis. 1983;
983(4):308-308.
5. Lan W, Rencoret J, Lu F, et al. Tricin-lignins: occurrence and
quantitation of tricin in relation to phylogeny. Plant J. 2016;
8
8(6):1046-1057.
6. Kajigaeshi S, Kakinami T, Okamoto T, Nakamura H,
Fujikawa M. Halogenation using quaternary ammonium poly-
halides. IV. Selective bromination of phenols by use of tetra-
alkylammonium tribromides. Bull Chem Soc Jpn. 1987;60(11):
4
187-4189.
3
3
7. Miyano M. Regioselective synthesis using the deuterium iso-
tope effect. J Org Chem. 1981;46(9):1854-1857.
53. Sircar D, Mukherjee C, Beuerle T, Beerhues L, Mitra A. Char-
acterization of p-hydroxybenzaldehyde dehydrogenase, the
final enzyme of p-hydroxybenzoic acid biosynthesis in hairy
roots of Daucus carota. Acta Physiol Plant. 2011;33(5):
2019-2024.
8. Chen T, Wu H, He Y, et al. Simultaneously quantitative analy-
sis of naringin and its major human gut microbial metabolites
0
naringenin and 3-(4 -hydroxyphenyl) propanoic acid via stable

416
WATANABE ET AL.
5
4. Koeduka T, Kajiyama M, Suzuki H, Furuta T, Tsuge T,
Matsui K. Benzenoid biosynthesis in the flowers of Eriobotrya
japonica: molecular cloning and functional characterization of
p-methoxybenzoic acid carboxyl methyltransferase. Planta.
How to cite this article: Watanabe B,
Nishitani S, Koeduka T. Synthesis of deuterium-
labeled cinnamic acids: Understanding the volatile
benzenoid pathway in the flowers of the Japanese
loquat Eriobotrya japonica. J Label Compd
Radiopharm. 2021;64:403–416. https://doi.org/10.
1002/jlcr.3933
2
016;244(3):725-736.
SUPPORTING INFORMATION
Additional supporting information may be found online
in the Supporting Information section at the end of this
article.