DPPH (=2,2-diphenyl-1-picrylhydrazyl=2,2-diphenyl-1-(2,4,6-trinitrophenyl) hydrazyl) radical-scavenging reaction of protocatechuic acid esters (=3,4-dihydroxybenzoates) in alcohols: Formation of bis-alcohol adduct

Article abstract of DOI:10.1002/hlca.200690074

Protocatechuic acid esters (=3,4-dihydroxybenzoates) scavenge ca. 5 equiv. of radical in alcoholic solvents, whereas they consume only 2 equiv. of radical in nonalcoholic solvents. While the high radical-scavenging activity of protocatechuic acid esters in alcoholic solvents as compared to that in nonalcoholic solvents is due to a nucleophilic addition of an alcohol molecule at C(2) of an intermediate o-quinone structure, thus regenerating a catechol (=benzene-l,2-diol) structure, it is still unclear why protocatechuic acid esters scavenge more than 4 equiv. of radical (C(2) refers to the protocatechuic acid numbering). Therefore, to elucidate the oxidation mechanism beyond the formation of the C(2) alcohol adduct, 3,4-dihydroxy-2-methoxybenzoic acid methyl ester (4), the C(2) MeOH adduct, which is an oxidation product of methyl protocatechuate (1) in MeOH, was oxidized by the DPPH radical (=2,2-diphenyl-1-picrylhydrazyl) or o-chloranil (=3,4,5,6-tetrachlorocyclohexa- 3,5-diene-1,2-dione) in CD₃-OD/(D₆)acetone 3:1). The oxidation mixtures were directly analyzed by NMR. Oxidation with both the DPPH radical and o-chloranil produced a C(2),C(6) bis-methanol adduct (7), which could scavenge additional 2 equiv. of radical. Calculations of LUMO electron densities of o-quinones corroborated the regioselective nucleophilic addition of alcohol molecules with o-quinones. Our results strongly suggest that the regeneration of a catechol structure via a nucleophilic addition of an alcohol molecule with a o-quinone is a key reaction for the high radical-scavenging activity of protocatechuic acid esters in alcoholic solvents.

Full text of DOI:10.1002/hlca.200690074

Helvetica Chimica Acta – Vol. 89 (2006)
821
DPPH (=2,2-Diphenyl-1-picrylhydrazyl=2,2-Diphenyl-1-(2,4,6-
trinitrophenyl)hydrazyl) Radical-Scavenging Reaction of Protocatechuic Acid
Esters (=3,4-Dihydroxybenzoates) in Alcohols:
Formation of Bis-alcohol Adduct
by Shizuka Saito, Hong Gao, and Jun Kawabata*
Laboratory of Food Biochemistry, Division of Applied Bioscience, Graduate School of Agriculture,
Hokkaido University, Kita-ku, Sapporo 060-8589, Japan
(phone: +81-11-706-2496; fax: +81-11-706-2496; e-mail: junk@chem.agr.hokudai.ac.jp)
Protocatechuic acid esters (=3,4-dihydroxybenzoates) scavenge ca. 5 equiv. of radical in alcoholic
solvents, whereas they consume only 2 equiv. of radical in nonalcoholic solvents. While the high radi-
cal-scavenging activity of protocatechuic acid esters in alcoholic solvents as compared to that in nonal-
coholic solvents is due to a nucleophilic addition of an alcohol molecule at C(2) of an intermediate o-qui-
none structure, thus regenerating a catechol (=benzene-1,2-diol) structure, it is still unclear why proto-
catechuic acid esters scavenge more than 4 equiv. of radical (C(2) refers to the protocatechuic acid num-
bering). Therefore, to elucidate the oxidation mechanism beyond the formation of the C(2) alcohol
adduct, 3,4-dihydroxy-2-methoxybenzoic acid methyl ester (4), the C(2) MeOH adduct, which is an oxi-
dation product of methyl protocatechuate (1) in MeOH, was oxidized by the DPPH radical (=2,2-
diphenyl-1-picrylhydrazyl) or o-chloranil (=3,4,5,6-tetrachlorocyclohexa-3,5-diene-1,2-dione) in CD3-
OD/(D6)acetone 3 :1). The oxidation mixtures were directly analyzed by NMR. Oxidation with both
the DPPH radical and o-chloranil produced a C(2),C(6) bis-methanol adduct (7), which could scavenge
additional 2 equiv. of radical. Calculations of LUMO electron densities of o-quinones corroborated the
regioselective nucleophilic addition of alcohol molecules with o-quinones. Our results strongly suggest
that the regeneration of a catechol structure via a nucleophilic addition of an alcohol molecule with a
o-quinone is a key reaction for the high radical-scavenging activity of protocatechuic acid esters in alco-
holic solvents.
Introduction. – Hydroxybenzoic acids such as protocatechuic acid (=3,4-dihydroxy-
benzoic acid) and gallic acid (=3,4,5-trihydroxybenzoic acid) are ubiquitously found in
vegetables and fruits [1][2]. These compounds are known to exhibit potent antioxidant
activities [3–5], and could prevent carcinogenesis and cardiovascular diseases that are
associated with radicals [6–8]. Recently, numerous studies have been reported on the
radical-scavenging activity of phenolic compounds. It is well-known that the radical-
scavenging activity of phenolic acids largely depends on the number and arrangement
of phenolic OH groups in the molecule [9–11]. Among them, o- or p-dihydroxy com-
pounds (=ꢀo- or p-diphenolsꢁ) such as protocatechuic acid exhibit a high radical-scav-
enging activity, since they would be converted to the corresponding o- or p-quinone
derivatives, respectively [9]. In addition, the higher radical-scavenging activity of ben-
zenepolyols (=ꢀpolyphenolsꢁ) can be, in part, ascribed to the dimerization of o-qui-
nones or semiquinone radicals, since the resultant dimers could scavenge additional
radicals if they possess a catechol (=benzene-1,2-diol) structure [12]. However, more
ꢂ 2006 Verlag Helvetica Chimica Acta AG, Zürich

822
Helvetica Chimica Acta – Vol. 89 (2006)
detailed studies are needed to understand the radical-scavenging mechanism beyond
the formation of quinones.
We have recently reported the DPPH (=2,2-diphenyl-1-picrylhydrazyl=Ph2-
A
uates) in nonalcoholic and alcoholic solvents [13]. In nonalcoholic acetone or acetoni-
trile, protocatechuic acid and its esters scavenge 2 equiv. of radical to yield the corre-
sponding o-quinones. In contrast, in alcoholic solvents, protocatechuates rapidly
react with ca. 5 equiv. of radical with a concomitant conversion to the corresponding
o-quinone derivatives, 3-hemiacetals [14], and their alcohol adducts at C(2) [13]. It
was suggested that the higher radical-scavenging activity of protocatechuates in alco-
holic solvents as compared to that in nonalcoholic solvents is due to a regeneration
of a corresponding catechol structure via a nucleophilic addition of an alcohol molecule
at C(2) of an o-quinone intermediate (C(2) refers to the protocatechuic acid number-
ing) [13]. Moreover, we found that catechols possessing strong electron-withdrawing
substituents in p-position C(1) exhibit a high DPPH-radical-scavenging activity in alco-
holic solvents, since electron-withdrawing substituents enhance the electrophilicity of
o-quinones, and hence facilitate a nucleophilic addition of an alcohol molecule with
an o-quinone [15]. However, the reason why protocatechuates scavenge more than 4
equiv. of radical in alcoholic solvents remained to be clarified. Therefore, the aim of
this study is to elucidate the radical-scavenging mechanism beyond the formation of
the C(2) alcohol adduct of protocatechuate. We hypothesized that a second nucleo-
philic addition of an alcohol molecule occurs with an o-quinone intermediate generated
from the C(2) adduct. In the present study, protocatechuic acid esters and authentic bis-
alcohol adducts were oxidized by the DPPH radical or by o-chloranil, and the oxidation
mixtures were directly analyzed by NMR. In addition, to substantiate the regioselective
nucleophilic addition, the LUMO energy and electron densities of the o-quinone inter-
mediates were calculated by a semi-empirical method. Finally, we propose a scheme
with the entire radical-scavenging reaction pattern of protocatechuic acid esters in alco-
holic solvents.
Result and Discussion. – Methyl protocatechuate (1) scavenges ca. 5 equiv. of
DPPH radical in MeOH [13]. Previously, we reported that 1 scavenges 2 equiv. of rad-
ical to yield the corresponding o-quinone derivative 2 and its hemiacetal 3 [14], and
subsequent nucleophilic addition of a MeOH molecule at C(2) of 2 produces 4,
which scavenges additional 2 equiv. of radical to give o-quinone derivative 5 and its
hemiacetal 6 [13] (Scheme 1). However, taking into account that 1 scavenges more
than 4 equiv. of radical, 5 and 6 should undergo further oxidation.
Therefore, to elucidate the oxidation mechanism beyond the formation of the C(2)
adduct 4, the reaction mixture of 4 and the DPPH radical in CD3OD/(D6)Acetone 3 :1
A
was directly analyzed by 1H-NMR. (D6)Acetone was added as a cosolvent to enhance
1
the solubility of the DPPH radical. In the H-NMR spectrum of the reaction mixture
after 10 min, the signals of 4 had completely disappeared, and a new s at d 5.75 was
observed, besides the d of the o-quinone derivative 5 at d 6.19 (d, J=10.3 Hz, Hꢀ
C(5)) and 7.38 (d, J=10.3 Hz, HꢀC(6)), and the signals of the 3-hemiacetal 6 at d
5.83 (d, J=10.3 Hz, HꢀC(5)) and 7.53 (d, J=10.3 Hz, HꢀC(6)) (Fig. 1,a) [13]. Since
further 2D-NMR analyses of the reaction mixture was hampered by the low solubility

Helvetica Chimica Acta – Vol. 89 (2006)
823

824
Helvetica Chimica Acta – Vol. 89 (2006)
Fig. 1. 1H-NMR Spectra of the reaction mixture of a) 4 and b) 7 with the DPPHradical in CD OD/
3
(D6)acetone 3 :1, 10 min after being mixed. The intense signals in the range d 7.1–7.4 are due to
DPPH.
of the DPPH radical in MeOH, o-chloranil (=3,4,5,6-tetrachlorocyclohexa-3,5-diene-
1
1,2-dione) was used as an oxidizing reagent. The H-NMR spectrum of the reaction
mixture of 4 and o-chloranil in CD3OD/(D6)acetone 3 :1 after 10 min showed, besides
G
the signals of unreacted 4, the signals of the corresponding o-quinone derivative 5 and
the 3-hemiacetal 6, which were identical to those of the oxidation products of 4 by
DPPH (Fig. 2,a). This result indicates that the reaction of 1 with o-chloranil proceeds
similarly to that with DPPH. After 30 min, two new s at d 5.37 and 5.75 appeared the
intensities of which gradually increased as those of the signals of 4, 5, and 6 decreased
(Fig. 2,b). The in situ HMBC analysis of the reaction mixture 4/o-chloranil showed the
correlation of d(H) 5.37 with the acetal C-atom at d(C) 91.5, and of d(H) 5.75 with the
carbonyl C-atom at d(C) 178.2. In addition, the 1H-NMR spectrum of the mixture 1/o-
chloranil, after 6 h, also showed the two s at d 5.37 and d 5.75, indicating that oxidation
of 1 also gives the same product as 4. We assumed that these s are the signals of bis-
methanol adducts.
To confirm the formation of the bis-methanol adduct, 2,6-dimethoxy- and 2,5-dime-
thoxyprotocatechuic acid methyl esters (7 and 10, resp.) were prepared according to the
procedure of Scheme 2, and the chemical shifts of their oxidation products were com-
pared with those of the unknown products obtained from 1 and 4.
The reaction mixture of 7 and DPPH in CD3OD/(D6)acetone 3 :1 was directly ana-
G
lyzed by NMR. In the 1H-NMR spectrum after 10 min, the signals of 7 disappeared and
two new s at d 5.37 and 5.75 were observed (Fig. 1,b). The HMBC plot of the reaction
mixture 7/DPPH was similar to that of the mixture 4/o-chloranil, and d(H) 5.75 showed
a cross peak with the quinone carbonyl C-atom at d(C) 178.2, and d(H) 5.37 with the
acetal C-atom at d(C) 91.5. Thus, the signals at d 5.75 and d 5.37 were assigned to
HꢀC(5) of the o-quinone derivative 8 and its 3-hemiacetal 9, respectively (Fig. 3). In
addition, the signals of 8 and 9 remained unchanged for more than 5 h, indicating
that 8 and 9 are more stable than the parents 2 and 3, which disappeared within 1 h
[13]. Hence, further nucleophilic addition of an alcohol molecule with 8 seems to be

Helvetica Chimica Acta – Vol. 89 (2006)
825
Fig. 2. 1H-NMR Spectra of the reaction mixture of 4 with o-chloranil in CD3OD/(D6)acetone 3 :1,
after a) 10 min and b) 60 min
Scheme 2. Synthesis of Compounds 7 and 10

826
Helvetica Chimica Acta – Vol. 89 (2006)
unlikely to occur. Moreover, oxidation of 7 with o-chloranil in CD3OD/(D6)acetone
G
3 :1 also formed the corresponding o-quinone derivative 8 and its 3-hemiacetal 9.
Fig. 3. Compounds 7, 10, 13, and their oxidation products
1
In the H-NMR spectrum of the reaction mixture of 10 and DPPH in CD3
A
(D6)acetone 3 :1, two s at d 5.34 and 6.22 were observed. The HMBC spectrum showed
correlations between d(H) 6.22 and the quinone carbonyl C-atom at d(C) 174.5, and
between d(H) 5.34 and the acetal C-atom at d(C) 92.5. Thus, d 6.22 and 5.34 were
assigned to HꢀC(6) of the o-quinone derivative 11 and its 4-hemiacetal 12, respectively
(Fig. 3). Oxidation of 10 with o-chloranil in CD3OD/(D6)acetone 3 :1 also produced 11
G
and 12. The absence of the s at d 5.34 and 6.22 in the 1H-NMR spectra of the reaction
mixtures of 1 or 4 with DPPH (or o-chloranil) in CD3OD/(D6)acetone 3 :1 strongly
AHCTREUNG
indicates that C(6) of 5 is the preferred position for the second nucleophilic attack of
a MeOH molecule.
We previously reported that oxidation of 1 in the presence of cysteine yields a C(2)
adduct, and subsequent oxidation of the latter leads to the corresponding o-quinone
derivative, which undergoes a second nucleophilic attack by cysteine at C(5) to produce
a C(2),C(5) bis-adduct instead of a C(2),C(6) bis-adduct [16]. Cheynier et al. also
reported that the quinone derivative of caffeoyltartaric acid undergoes a nucleophilic
attack by glutathione at C(2) of its aryl moiety, and further oxidation of the resultant
C(2) adduct in the presence of glutathione yields the C(2),C(5) bis-adduct [17]. To
understand the regioselectivities of the nucleophilic attacks, LUMO energies and elec-
tron densities of the o-quinone derivatives and their hemiacetals were calculated by a
semi-empirical method (Table). The result demonstrates that the LUMO electron den-
sities at C(3) of the o-quinone derivatives 2, 5, and 8 are higher than those at C(4), indi-
cating that an alcohol molecule preferentially attacks the C(3), rather than the C(4) car-
bonyl group, to form 3-hemiacetals 3, 6, and 9, respectively. In addition, the LUMO
energies of the o-quinone derivatives 2, 5, and 8 are much lower than those of the cor-
responding 3-hemiacetals 3, 6, and 9, suggesting that the o-quinone derivatives exclu-
sively undergo a nucleophilic attack by an alcohol molecule. In the case of compound
5, C(6) has a higher LUMO electron density as compared to C(5). This result confirms
that an alcohol molecule regioselectively attacks C(6). However, the presence of a MeS
group at C(2) produces a modification of the LUMO parameters. The LUMO electron
density at C(5) of 13 is, in fact, higher than at C(6), and thus C(5) is the preferred center
for a second nucleophilic attack. These results support the regioselective nucleophilic
additions of our experiments.

Helvetica Chimica Acta – Vol. 89 (2006)
827
Table. LUMO Energy and Electron Density at Each C-Atom of o-Quinone Derivatives 2, 5, 8, and 13 and
of Their 3-Hemiacetals 3, 6, and 9
2
3
5
6
8
9
13
LUMO ꢀ2.086 ꢀ1.336 ꢀ1.978 ꢀ1.171 ꢀ1.706 ꢀ1.368 ꢀ1.913
Electron density at C(1)
0.30
0.41
0.20
0.14
0.21
0.16
0.25
0.63
0.02
0.17
0.29
0.29
0.20
0.38
0.29
0.16
0.14
0.17
0.11
0.57
0.02
0.26
0.21
0.37
0.21
0.40
0.24
0.16
0.13
0.23
0.23
0.55
0.01
0.16
0.32
0.35
0.25
0.47
0.20
0.12
0.17
0.15
C(2)
C(3)
C(4)
C(5)
C(6)
The time course of the DPPH radical-scavenging activity of 1 and its oxidation
products 4 and 7 in MeOH is shown in Fig. 4. The DPPH-radical-scavenging equiva-
lence is expressed as a value relative to that of DL-a-tocopherol taken as 2.0. The result
shows that compounds 1, 4, and 7 rapidly reacted with the DPPH radical and reached a
plateau within 20 min. The relative radical-scavenging equivalences of each compound
after 30 min were 5.0 for 1, 3.1 for 4, and 1.9 for 7. Considering that 1 needs to scavenge
6 equiv. of radical to produce 8, a second MeOH addition with 5 to yield 7 might be lim-
ited. Furthermore, since 7 scavenged only 2 equiv. of radical, the nucleophilic addition
of an alcohol molecule with 8 would be unlikely to occur. This was supported by the
calculated values of the LUMO energies of the o-quinone derivatives which increased
in the order of 2<5<8 (Table). In addition, steric hindrance due to bis-methanol addi-
tion may also explain the observed lack of reactivity of 8 toward a nucleophilic attack.
&
*
*
Fig. 4. Time course of DPPH-radical-scavenging activity of 1 ( ), 4 ( ), and 7 ( ) in methanol. The
equivalence is expressed as the values relative to that of DL-a-tocopherol taken as 2.0.

828
Helvetica Chimica Acta – Vol. 89 (2006)
A plausible radical-scavenging mechanism of 1 in MeOH is shown in Scheme 1.
First, 1 scavenges two radicals and is converted to the corresponding o-quinone deriv-
ative 2. The latter undergoes a nucleophilic attack by a MeOH molecule at C(3) to yield
the 3-hemiacetal 3 [14], which is equilibrated with 2 in the reaction solution. Then, a
regeneration of a catechol structure occurs via a nucleophilic addition of a MeOH mol-
ecule at C(2) of 2 to give 4, which scavenges another two radicals to yield the corre-
sponding o-quinone derivative 5 and its 3-hemiacetal 6. The o-quinone derivative 5
undergoes a second nucleophilic addition of a MeOH molecule at C(6), leading to 7,
i.e., to a regeneration of a catechol structure which can consume additional 2 equiv.
of radical to give o-quinone derivative 8 and its 3-hemiacetal 9. Hence, the formation
of a bis-alcohol adduct contributes to the high radical-scavenging activity of 1 in alco-
holic solvents.
Conclusions. – In this study, we showed that oxidation of methyl protocatechuate
(1) in MeOH leads to a C(2),C(6) bis-methanol adduct, besides a C(2) adduct. Our
results strongly suggest that the regeneration of a catechol structure via a nucleophilic
addition of an alcohol molecule with an o-quinone derivative is a key reaction for the
high radical-scavenging activity of protocatechuic acid esters in alcoholic solvents. Fur-
ther study is needed to examine whether the oxidation mechanism described above also
occurs in a biological aqueous system.
We are grateful to Mr. Kenji Watanabe and Dr. Eri Fukushi, of the GC-MS and NMR Laboratory of
our school, for measuring mass spectra. This work was supported by a research fellowship for young sci-
entists from the Japan Society for the Promotion of Science (to S. S.).
Experimental Part
General. Protocatechuic acid was obtained from Sigma Chemical Co., and o-chloranil (=3,4,5,6-tet-
rachlorocyclohexa-3,5-diene-1,2-dione) from Aldrich Chemical Co. Methyl 2,4,6-trihydroxybenzoate was
purchased from Alfa Aesar. Methyl protocatechuate (1) and methyl 3,4-dihydroxy-2-methoxybenzoate
(4) were prepared by the methods described previously [13]. The 2,2-diphenyl-1-picrylhydrazyl
(DPPH) radical and other reagents were purchased from Wako Pure Chemical Industries. All solvents
used were of reagent grade.
Prep. and anal. TLC: silica gel plates Merck 60 F254 (0.5 and 0.25 mm thickness, resp.). Column chro-
matography (CC): silica gel, Wakogel C-300 (Wako Pure Chemical Industries). M.p.: hot-stage apparatus;
uncorrected. NMR Spectra: Bruker-AMX-500 spectrometer (1H, 500 MHz; 13C, 125 MHz); chemical
shifts d in ppm rel. to the residual signals of CDCl3 (d(H) 7.24, d(C) 77.0), CD3OD (d(H) 3.30, d(C)
G
49.0) and (D6)acetone (d(H) 2.04, d(C) 29.8). EI-MS and FD-MS: Jeol-JMS-AX-500 and Jeol-JMS-
SX102A instruments, resp.; in m/z (rel.%).
4-(Benzyloxy)-2,6-dimethoxybenzoic Acid Methyl Ester (7a). A mixture of methyl 2,4,6-trihydroxy-
benzoate (1.84 g, 10 mmol), benzyl bromide (1.2 ml, 10 mmol, 1.0 equiv.), and K2CO3 (1.4 g, 10 mmol, 1.0
equiv.) in acetone (50 ml) was refluxed for 3 h. After cooling to r.t., the mixture was filtered and the filter-
ate evaporated. The resulting benzyl ether was dissolved in acetone (50 ml), and MeI (1.2 ml, 20 mmol,
2.0 equiv.) and K2CO3 (2.8 g, 20 mmol, 2.0 equiv.) were added. After refluxing for 6 h, the mixture was
filtered, the filterate evaporated, and the crude product subjected to CC silica gel, hexane/AcOEt
1
2 :1: 7a (2.1 g, 70%). H-NMR (CD3
G
HꢀC(3), HꢀC(5)); 7.31–7.49 (m, PhCH2). HR-FD-MS: 302.1152 (M+, C17
4-Hydroxy-2,6-dimethoxybenzoic Acid Methyl Ester (7b). Compound 7a (1.8 g, 6.0 mmol) was
deprotected by hydrogenation at 1 atm. in the presence of a catalytic amount of 10% Pd/C in acetone

Helvetica Chimica Acta – Vol. 89 (2006)
829
(50 ml). The mixture was filtered through Celite, and the filtrate was evaporated: 7b (1.1 g, 87%). White
powder. 1H-NMR (CD3
OD): 3.71 (s, COOMe); 3.72 (s, 2 MeO); 6.13 (s, HꢀC(3), HꢀC(5)). HR-EI-MS:
212.0667 (M+, C10 Oþ5 ; calc. 212.0685).
H12
AHCTREUNG
A
ACHTREUNG
3,4-Dihydroxy-2,6-dimethoxybenzoic Acid Methyl Ester (7). According to the modified method of
Shaw et al. [18]: To a stirred mixture of 7b (1.0 g, 4.7 mmol) and NaOH (2.0 g, 50 mmol) in H2O (50
ml) was added dropwise a soln. of potassium persulfate (2.7 g, 10 mmol) in H2O (50 ml) at r.t. After stir-
ring for 24 h, the soln. was acidified to pH 4 with conc. HCl soln. The mixture was filtered, and the filtrate
was washed with Et2O to remove unreacted 7b. To the aq. phase was added conc. HCl soln. (10 ml), and
U
the mixture was refluxed for 2 h. After cooling to r.t., the mixture was extracted with AcOEt, the org.
layer dried (Na2SO4) and evaporated and the residue subjected to prep. TLC (CHCl3/MeOH/
1
HCOOH 100 :4 :0.1): 7 (0.15 g, 14%). Rf =0.34. Yellow crystalline solid. H-NMR ((D6)acetone): 3.68
(s, 1 Me); 3.76 (s, 1 Me); 3.78 (s, 1 Me); 6.35 (s, HꢀC(5)). 13C-NMR ((D6)acetone): 51.9; 56.6; 61.5;
96.8; 110.9; 132.3; 147.3; 148.9; 151.0; 166.8. HR-FD-MS: 228.0620 (M+, C10
N
N
3,4-Dihydroxy-5-methoxybenzoic Acid Methyl Ester (10a). According to the method of Chang et al.
[19]: To a soln. of sodium tetraborate decahydrate (38 g, 0.10 mol, 2.5 equiv.) in H2O (500 ml) was added
methyl gallate (=methyl 3,4,5-trihydroxybenzoate; 7.5 g, 41 mmol) at r.t. After stirring for 1 h, dimethyl
sulfate (15 ml, 0.16 mmol, 3.9 equiv.) and 6.5M aq. NaOH (25 ml) were added dropwise. After stirring for
an additional 12 h, the mixture was acidified to pH 2 with conc. H2SO4 soln. The mixture was poured into
H2O and extracted with AcOEt. The org. layer was evaporated and the residue was purified by CC (silica
gel, hexane/AcOEt 1:1): 10a (5.9 g, 73%). White powder. 1H-NMR ((D6)acetone): 3.80 (s, Me); 3.86 (s,
Me); 7.15 (d, J=2.0, 1 H); 7.21 (d, J=2.0, 1 H). HR-EI-MS: 198.0529 (M+, C9
N
N
2-Bromo-3,4-dihydroxy-5-methoxybenzoic Acid Methyl Ester (10b). According to the modified
method of Lai et al. [20]: To a soln. of 10a (3.0 g, 15 mmol) in AcOH/MeCN 3 :1 (20 ml) was added drop-
wise N-bromosuccinimide (2.7 g, 15 mmol, 1.0 equiv.) in AcOH/MeCN 3 :1 (24 ml), and the mixture was
stirred at r.t. for 4 h. The mixture was evaporated, the residue suspended in H2O and extracted with
AcOEt, the org. layer evaporated, and the crude product subjected to CC (silica gel, hexane/AcOEt
1
1:1): 10b (3.8 g, 91%). H-NMR ((D6)acetone): 3.82 (s, COOMe); 3.86 (s, MeOꢀC(5)); 7.06 (s, Hꢀ
C(6)). 13C-NMR ((D6)acetone): 52.2 (Me); 56.7 (Me); 102.8 (C(2)); 107.0 (C(6)); 123.7 (C(1)); 138.5
1
(C(4)); 144.4 (C(3)); 147.3 (C(5)); 167.0 (C=O). H,13C-HMBC: HꢀC(6) $ C(2), C(4), C=O; MeOꢀ
C(5) $ C(5); COOMe $ C=O. HR-EI-MS: 275.9651 (M+, C9
N
N
3,4-Bis(benzyloxy)-2-bromo-5-methoxybenzoic Acid Methyl Ester (10c). A mixture of 10b (3.8 g, 14
mmol), benzyl bromide (3.3 ml, 28 mmol, 2.0 equiv.), and potassium carbonate (3.9 g, 28 mmol, 2.0
equiv.) in acetone (50 ml) was refluxed for 3 h. After cooling to r.t., the mixture was filtered, the filtrate
evaporated and the crude product purified by CC (silica gel, hexane/AcOEt 2 :1): 10c (5.4 g, 86%). 1H-
NMR ((D6)acetone): 3.87 (s, 1 Me); 3.93 (s, 1 Me); 5.04 (s, 1 PhCH2); 5.15 (s, 1 PhCH2); 7.26 (s, HꢀC(6));
7.32–7.52 (m, 2 PhCH2). HR-FD-MS: 456.0563 (M+, C23
N
N
3,4-Bis(benzyloxy)-2-hydroxy-5-methoxybenzoic Acid (10d). According to the modified method of
Cakmak et al. [21]: To a stirred soln. of 10c (5.5 g, 12 mmol) in DMF (40 ml) was added a suspension
of NaOMe (6.5 g, 120 mmol, 10 equiv.) and CuI (1.1 g, 6.0 mmol, 0.50 equiv.) in DMF (80 ml). The mix-
ture was stirred for 2 d under N2 at 908. After cooling to r.t., the mixture was poured into H2O and
extracted with AcOEt. The org. layer was evaporated, and the crude product subjected to CC (silica
gel, hexane/AcOEt 1:1): 10d (1.9 g, 42%). 1H-NMR ((D6)acetone): 3.82 (s, MeOꢀC(5)); 5.07 (s,
PhCH2); 5.09 (s, PhCH2); 7.28–7.51 (m, 2 PhCH2, HꢀC(6)). HR-FD-MS: 380.1236 (M+, C22
N
G
calc. 380.1260).
3,4-Bis(benzyloxy)-2,5-dimethoxybenzoic Acid Methyl Ester (10e). To a soln. of 10d (1.0 g, 2.6 mmol)
in acetone (20 ml) were added MeI (0.32 ml, 5.2 mmol, 2.0 equiv.) and K2CO3 (0.72 g, 5.2 mmol, 2.0
equiv.). The mixture was refluxed for 5 h and then filtered. The filtrate was evaporated and the residue
subjected to CC (silica gel, CHCl3/MeOH 100 :3): 10e (0.96 g, 90%). 1H-NMR (CDCl3): 3.84 (s, 1 Me);
3.87 (s, 1 Me); 3.90 (s, 1 Me); 5.01 (s, PhCH2); 5.08 (s, PhCH2); 7.13 (s, HꢀC(6)); 7.28–7.44 (m, 2 PhCH2).
HR-FD-MS: 408.1581 (M+, C24
N
N
3,4-Dihydroxy-2,5-dimethoxybenzoic Acid Methyl Ester (10). Compound 10e (0.50 g, 1.2 mmol) in
MeOH (15 ml) was deprotected by hydrogenation at 1 atm. in the presence of a catalytic amount of
10% Pd/C. The mixture was filtered through Celite, the filtrate evaporated, and the crude product puri-

830
Helvetica Chimica Acta – Vol. 89 (2006)
1
fied by CC (silica gel, CHCl3/MeOH 10 :1): 10 (0.26 g, 93%). Pale yellow powder. M.p. 126–1288. H-
NMR (CD3 OD):
A
ACHTREUNG
52.3 (COOMe); 56.7 (Me); 62.0 (Me); 105.5 (C(6)); 114.6 (C(1)); 140.5 (C(3)); 141.3 (C(4)); 145.4, 145.6
1
(C(2), C(5)); 167.9 (C=O). H,13C-HMBC: HꢀC(6) $ C(2), C(4), C=O; MeOꢀC(2) $ C(2); MeOꢀ
C(5) $ C(5); COOMe $ C=O. HR-EI-MS: 228.0669 (M+, C10
U
G
CD3OD/(D6)acetone 3 :1 (0.4 ml). (D6)acetone was added as a cosolvent to enhance the solubility of
A
DPPH. The mixture was immediately transferred to a NMR tube and mixed vigorously. 1H-NMR spectra
were recorded 10 min after mixing.
Reaction of 4 with DPPH. 1H-NMR (CD3
ACHTREUNG
Reaction of 7 with DPPH. 1H-NMR (CD3
A
(s, 1 Me); 5.75 (s, HꢀC(5)); 1H,13C-HMBC: HꢀC(5) $ 123.5 (C(1)), 178.2 (C(3)); 9: 3.79 (s, 1 Me); 3.85
(s, 1 Me); 4.04 (s, 1 Me); 5.37 (s, HꢀC(5)); HMBC: HꢀC(5) $ 91.5 (C(3)), 109.6 (C(1)).
Reaction of 10 with DPPH. 1H-NMR (CD3
OD/(D6)acetone 3 :1): 11: 3.75 (s, 1 Me); 3.85 (s, 1 Me);
U
3.87 (s, 1 Me); 6.22 (s, HꢀC(6)); HMBC: HꢀC(6) $ 150.2, 151.9, 167.1, 174.5; 12: 3.72 (s, 1 Me); 3.76 (s,
1 Me); 3.84 (s, 1 Me); 5.34 (s, HꢀC(6)); HMBC: HꢀC(6) $ 92.5, 145.5, 161.0, 167.4.
NMR Measurements of the Reaction Mixtures of Catechol Derivatives 1, 4, 7, or 10 with o-Chloranil.
To a catechol derivative (20 mmol) was added o-chloranil (4.9 mg, 20 mmol, 1.0 equiv.) in CD3OD/
U
(D6)acetone 3 :1 (0.4 ml). The mixture was immediately transferred to a NMR tube and mixed vigo-
1
rously. H-NMR spectra of 1 and 4 were recorded 10, 30, 60, 120, 180, and 360 min after mixing, and
those of 7 and 10 10 min after mixing.
1
Reaction of 1 and o-Chloranil. H-NMR (CD3
N
6.93 (d, J=2.0, HꢀC(2)); 7.52 (dd, J=10.3, 2.0, HꢀC(6)); 3: 6.11 (d, J=10.3, HꢀC(5)); 7.21 (d,
J=2.0, HꢀC(2)); 5: 6.18 (d, J=10.3, HꢀC(5)); 6: 5.83 (d, J=10.3, HꢀC(5)); 8: 5.75 (s, HꢀC(5)); 9:
5.37 (s, HꢀC(5)).
1
Reaction of 4 and o-Chloranil. H-NMR (CD3
7.38 (d, J=10.3, HꢀC(6)); 6: 5.83 (d, J=10.3, HꢀC(5)); 7.53 (d, J=10.3, HꢀC(6)); 8: 5.75 (s, Hꢀ
C(5)). 9: 5.37 (s, HꢀC(5)).
Reaction of 7 and o-Chloranil. 1H-NMR (CD3
OD/(D6)acetone 3 :1): 8: 3.85 (s, 1 Me); 3.90 (s, 1 Me);
A
3.98 (s, 1 Me); 5.75 (s, HꢀC(5)); 9: 3.79 (s, 3 H, 1 Me); 3.84 (s, 1 Me); 4.04 (s, 1 Me); 5.37 (s, HꢀC(5)).
Reaction of 10 and o-Chloranil. 1H-NMR (CD3
AHCTREUNG
3.87 (s, Me); 6.22 (s, HꢀC(6)); 12: 3.72 (s, Me); 3.76 (s, Me); 3.84 (s, Me); 5.34 (s, HꢀC(6)).
Molecular-Orbital Calculations. The electron densities and energies of LUMOs were calculated by
the AM1 method with the MOPAC 2000 program included in the Chem3D package (CambridgeSoft Co).
REFERENCES
[1] R. J. Robbins, J. Agric. Food Chem. 2003, 51, 2866.
[2] P. Mattila, J. Kumpulainen, J. Agric. Food Chem. 2002, 50, 3660.
[3] F. Natella, M. Nardini, M. D. Felice, C. Scaccini, J. Agric. Food Chem. 1999, 47, 1453.
[4] Z. Sroka, W. Cisowski, Food Chem. Toxicol. 2003, 41, 753.
[5] L. Kong, Z.-L. Sun, L.-F. Wang, H.-Y. Zhang, S.-D. Yao, Helv. Chim. Acta 2004, 87, 511.
[6] H. Nakamura, A. Nishikawa, F. Furukawa, K. Kasahara, M. Miyauchi, H.-Y. Son, M. Hirose, Anti-
cancer Res. 2000, 20, 3423.
[7] T.-H. Tseng, J.-D. Hsu, M.-H. Lo, C.-Y. Chu, F.-P. Chou, C.-L. Huang, C.-J. Wang, Cancer Lett. 1998,
126, 199.
[8] T. Tanaka, T. Kojima, T. Kawamori, H. Mori, Cancer 1995, 75, 1433.
[9] J. H. Chen, C.-T. Ho, J. Agric. Food Chem. 1997, 45, 2374.
[10] C. A. Rice-Evans, N. J. Miller, G. Paganga, Free Radical Biol. Med. 1996, 20, 933.
´
[11] B. Tyrakowska, A. E. M. F. Soffers, H. Szymusiak, S. Boeren, M. G. Boersma, K. Lemanska, J. Ver-
voort, I. M. C. M. Rietjens, Free Radical Biol. Med. 1999, 27, 1427.

Helvetica Chimica Acta – Vol. 89 (2006)
831
[12] H. Hotta, S. Nagano, M. Ueda, Y. Tsujino, J. Koyama, T. Osakai, Biochim. Biophys. Acta 2002, 1572,
123.
[13] S. Saito, Y. Okamoto, J. Kawabata, Biosci. Biotechnol. Biochem. 2004, 68, 1221.
[14] S. Saito, Y. Okamoto, J. Kawabata, T. Kasai, Biosci. Biotechnol. Biochem. 2003, 67, 1578.
[15] S. Saito, J. Kawabata, Tetrahedron 2005, 61, 8101.
[16] S. Saito, J. Kawabata, J. Agric. Food Chem. 2004, 52, 8163.
[17] M. Salgues, V. Cheynier, Z. Gunata, R. Wylda, J. Food Sci. 1986, 51, 1191.
[18] S. C. Shaw, A. S. Jha, A. K. Gupta, J. Indian Chem. Soc. 1990, 67, 684.
[19] J. Chang, R. Chen, R. Guo, C. Dong, K. Zhao, Helv. Chim. Acta 2003, 86, 2239.
[20] C.-H. Lai, Y.-L. Shen, M.-N. Wang, N. S. K. Rao, C.-C. Liao, J. Org. Chem. 2002, 67, 6493.
[21] O. Cakmak, I. Demirtas, H. T. Balaydin, Tetrahedron 2002, 58, 5603.
Received November 29, 2005