First total synthesis of artepillin C established by o,o′-diprenylation of p-halophenols in water

Article abstract of DOI:10.1021/jo0056904

We have demonstrated that prenylation of phalophenols was dependent on the solvent effect and succeeded in o,o′-diprenylation of p-halophenols in water. Following the Mizoroki-Heck coupling of the diprenyl-piodophenol 3c with methyl acrylate and then hydr

Full text of DOI:10.1021/jo0056904

J . Org. Chem. 2002, 67, 2355-2357
2355
is not an ideal functional group for the further modifica-
tions required for our synthesis.
F ir st Tota l Syn th esis of Ar tep illin C
Esta blish ed by o,o′-Dip r en yla tion of
p-Ha lop h en ols in Wa ter
We here present the first total synthesis of artepillin
C. Our synthetic strategy was based on an initial o,o′-
diprenylation reaction of p-halophenols. The advantage
of this approach is that fewer side reactions occur in the
reaction of p-halophenols than p-hydroxycinnamic acid,
and that product isolation and purification are greatly
simplified. Our short and efficient synthesis of artepillin
C was completed by Mizoroki-Heck reaction of the
isolated 2,6-di(3-methyl-2-butenyl)-4-halophenols (o,o′-
diprenyl-p-halophenols) with methyl acrylate followed by
hydrolysis of the ester.
Concerning the o-prenylation of phenol derivatives,
Fatope et al. reported that prenylation of 4-(p-hydroxy-
phenyl)butyric acid in organic solvents such as tetrahy-
drofuran or toluene gave 3-prenyl-4-(p-hydroxyphenyl)-
butyric acid and no o,o′-diprenyl compound was ob-
tained.6 Bates et al. also reported that prenylation of
p-halophenols in toluene gave o-prenyl-p-halophenols,
and no o,o′-diprenyl compounds were obtained.7 Lewis
et al. achieved initial success of the o,o′-dialkylation of
phenol with ethylene. However, their procedure, use of
an o-metalated ruthenium phosphite complex under a
pressure of 6.5 bar, is not a convenient laboratory
operation.8
Yoshihiro Uto, Akihiko Hirata, Tomoya Fujita,
Syunsuke Takubo, Hideko Nagasawa, and Hitoshi Hori*
Department of Biological Science & Technology,
Faculty of Engineering, The University of Tokushima,
Minamijosanjima-cho 2-1, Tokushima, 770-8506 J apan
hori@bio.tokushima-u.ac.jp
Received October 20, 2000
Abstr a ct: We have demonstrated that prenylation of p-
halophenols was dependent on the solvent effect and suc-
ceeded in o,o′-diprenylation of p-halophenols in water.
Following the Mizoroki-Heck coupling of the diprenyl-p-
iodophenol 3c with methyl acrylate and then hydrolysis, we
first synthesized artepillin C [3-{4-hydroxy-3,5-di(3-methyl-
2-butenyl)phenyl}-2(E)-propenoic acid] (1), which is a bio-
logically active constituent of propolis. These reactions may
be applicable to the synthesis of various useful natural
products such as 2,4,6-trisubstituted phenol derivatives.
In the work we present here, we reveal that o,o′-
diprenylation of p-halophenols can be readily achieved
at room temperature in alkaline solution (Scheme 1).
First, we focused on the effect of the nature of the halogen
on reactivity and regioselectivity of C- and O-prenylation
of p-halophenols. The results are summarized in Table
1. The yield of o,o′-diprenyl-p-iodophenol 3c (24%, entry
3) was the highest, followed by that of chlorophenol 3a
(16%, entry 1) and bromophenol 3b (15%, entry 2). Ratios
of C- to O-prenylation of 2a , 2b, and 2c were 1.3, 1.5,
and 1.6, respectively. Next, we tried to obtain the o,o′-
diprenyl-p-halophenols selectively with 2c as a starting
material. When 2c was treated with 1.0 equiv of prenyl
bromide for 1 h, 3c was obtained selectively (27%) and
byproducts 4c and 5c were obtained in very low yield
(2 and 3%, respectively), while byproduct 6c (7%) was
newly obtained (entry 4). The conversion yield of 2c to
3c was 47% (2c was recovered in 42%).
To understand the differences in reactivity between
p-halophenols and p-hydroxycinnamic acid in the o,o′-
diprenylation reaction, we calculated their molecular
orbital energies using the AM1 method on the MOPAC97
program (Figure 1). The highest occupied molecular
orbital (HOMO) energy of 2a ′-c′ (from -3.040 to -3.246
eV) was close to the lowest unoccupied molecular orbital
(LUMO) energy of prenyl bromide (0.841 eV). Moreover,
2a ′-c′ showed high C2 values at their ortho (from 0.213
to 0.216) and para positions (from 0.307 to 0.319). Thus,
p-halophenols suffered electrophilic attack by prenyl
bromide at their ortho positions. However, the order of
reactivity of o,o′-diprenylation was opposite that pre-
dicted by the nucleophilicities of ortho positions obtained
Propolis, obtained from beehives, consists of a mixture
of compounds that includes various 2,4,6-trisubstituted
phenols. Because of its well-known biological effects such
as antibacterial, antiviral, antiinflammatory, antioxida-
tive, and immunomodulatory effects, propolis has been
used as a popular medicine. Artepillin C [3-{4-hydroxy-
3,5-di(3-methyl-2-butenyl)phenyl}-2(E)-propenoic acid]
(1) is an antimicrobial 2,4,6-trisubstituted phenol isolated
as a major constituent (>5%) from Brazilian propolis.1
The recent report that artepillin C has important biologi-
cal activities such as antitumor,2 apoptosis-inducing,3
immunomodulating,4 and antioxidative activities5 shows
that artepillin C may be one of the important active
principles of Brazilian propolis. Indeed, the importance
of artepillin C has been increasingly recognized and it
has recently been designated a criterion of the quality of
Brazilian propolis. These biological properties have made
clear the urgent need for an efficient total synthesis of
artepillin C. However, to the best of our knowledge, the
total synthesis of artepillin C has never been accom-
plished. Our attempts to achieve a direct synthesis of
artepillin C by o,o′-diprenylation of p-hydroxycinnamic
acid were thwarted in part because of difficulty in
isolation of pure artepillin C from the reaction mixture.
This was due to the large amount of prenyl bromide and
byproducts in the reaction mixture. Moreover, the R,â-
unsaturated carboxyl group of p-hydroxycinnamic acid
(1) Aga, H.; Shibuya, T.; Sugimoto, T.; Nakajima, S.; Kurimoto, M.
Biosci. Biotech. Biochem. 1994, 58, 945.
(2) Kimoto, T.; Arai, S.; Aga, M.; Hanaya, T.; Kohguchi, M.; Nomura,
Y.; Kurimoto, M. Gan to Kagaku Ryoho 1996, 23, 1855.
(3) Matsuno, T.; J ung, S. K.; Matsumoto, Y.; Saito, M.; Morikawa,
J . Anticancer Res. 1997, 17, 3565.
(4) Kimoto, T.; Arai, S.; Kohguchi, M.; Aga, M.; Nomura, Y.; Micallef,
M. J .; Kurimoto, M.; Mito, K. Cancer. Detect. Prev. 1998, 22, 506.
(5) Hayashi, K.; Komura, S.; Isaji, N.; Ohishi, N.; Yagi, K. Chem.
Pharm. Bull. 1999, 47, 1521.
(6) Fatope, M. O.; Okogun, J . I. J . Chem. Soc., Perkin Trans. 1 1982,
1, 1601.
(7) Bates, R. W.; Gabel, C. J . Tetrahedron Lett. 1993, 34, 3547.
(8) Lewis, L. N.; Smith, J . F. J . Am. Chem. Soc. 1986, 108, 2728.
10.1021/jo0056904 CCC: $22.00 © 2002 American Chemical Society
Published on Web 03/13/2002

2356 J . Org. Chem., Vol. 67, No. 7, 2002
Notes
Ta ble 1. P r od u ct Distr ibu tion s Obta in ed in th e P r en yla tion of p-Ha lop h en ols 2a -c
productsa (isolated yield/%)
entry
2
prenyl bromide (equiv)
time (h)
3
4
5
6
recovered 2
1
2
3
4
2a
2b
2c
2c
2.5
2.5
2.5
1.0
5
5
5
1
3a (16, 23b)
3b (15, 23b)
3c (24, 29b)
3c (27, 47b)
4a (20)
4b (16)
4c (23)
4c (2)
5a (35)
5b (26)
5c (35)
5c (3)
32
36
17
42
6c (7)
a
b
Assignment of structures is based on NMR data (see Experimental Section). Conversion yield of 2 to 3.
Sch em e 2. Syn th esis of Ar tep illin C 1 by
Mizor ok i-Heck Rea ction
F igu r e 1. HOMO energies and the square of atomic orbital
coefficients (C2) calculated for phenolates 2a ′-c′ and p-
hydroxycinnamate.
above, o,o′-diprenylated p-hydroxycinnamic acid was
obtained in low yield.
Sch em e 1. P r en yla tion of p-Ha lop h en ols 2a -c
Next, we examined the Mizoroki-Heck coupling of o,o′-
diprenyl-p-halophenols 3a -c with methyl acrylate. Mi-
zoroki-Heck coupling of aromatic rings bearing an
electron-donating substituent such as the hydroxyl group
with an alkene is generally inefficient because of slow
alkene coupling and decomposition of the palladium
catalyst.9 Therefore, we first acetylated the hydroxyl
group of 3a -c to reduce the electron density of the ring,
but we did not obtain any coupling products. This result
may be due to the steric hindrance of two o-prenyl groups
against the hydroxyl group.
We finally were able to effect Mizoroki-Heck coupling
of o,o′-diprenyl-p-halophenols 3c with methyl acrylate
without protection of its hydroxyl group by using 5 mol
% palladium diacetate, 10 mol % tri-o-tolylphosphine, and
an excess of triethylamine to produce artepillin C methyl
ester 7 in 31% yield (Scheme 2). In the case of 3a and
3b, only starting materials were recovered under similar
conditions. Our results are consistent with those of Bates
et al., who reported that, without protection of the
hydroxyl group, the Mizoroki-Heck coupling proceeded
only with p-iodophenol derivatives.7 Finally, artepillin C
methyl ester 7 was hydrolyzed in aqueous/methanolic
potassium hydroxide to give artepillin C 1 in 78% yield.
In summary, the total synthesis of artepillin C was
achieved by o,o′-diprenylation of p-iodophenol in water
followed by Mizoroki-Heck coupling with methyl acry-
late. In future work, we will apply this strategy to the
synthesis of other biologically active 2,4,6-trisubsutituted
phenol derivatives.
from calculated HOMO energies. This finding suggested
that the halogen group was influenced by the protic
solvent. However, our calculation did not consider the
solvent effect. In the regioselectivity of C- or O-prenyla-
tion, a difference in the C2 values of hydroxyl anion
(0.193) and ortho positions (0.214) of 2a ′ agreed closely
with the observed C- to O-prenylation ratio.
Exp er im en ta l Section
All commercial materials were used without purification.
Toluene was distilled under nitrogen from CaH2. Analytical thin-
layer chromatography was performed using silica 60-F254 coated
0.25 mm plates. Column chromatography was performed using
the indicated solvent on silica gel (63-200 µm). 1H NMR (400
Mz) spectra were recorded using diluted solutions of each
compound in CDCl3 as the solvent and tetramethylsilane as the
internal standard. All melting points were determined by a micro
melting point apparatus and are uncorrected. The semiempirical
molecular orbital (MO) calculation was applied by the AM1
Additionally, the HOMO energy of p-hydroxycinnama-
te (-0.373 eV) is higher than those of phenolates 2a ′-
c′. This finding predicted that p-hydroxycinnamic acid
actually would have a higher reactivity toward prenyl
bromide than p-halophenols. However, as discussed
(9) Ziegler, C. B.; Heck, R. F. J . Org. Chem. 1978, 43, 2941.

Notes
J . Org. Chem., Vol. 67, No. 7, 2002 2357
methods of Stewart and the program MOPAC97. The orbital
energy (eV) and C2 value of the highest occupied molecular
orbital (HOMO) were calculated. The o-prenyl-p-iodophenol 6c
is a known compound.7
4-Ch lor o-2,6-d i(3-m eth yl-2-bu ten yl)p h en ol (3a ): yellow
oil;
1H NMR (400 MHz, CDCl3) δ 6.94 (s, 2H, H3,H5-Ar), 5.31
(s, 1H, OH), 5.27 (t, J ) 7.1 Hz, 2H, Ar(CH2CHdC(CH3)2)2), 3.29
(d, J ) 7.1 Hz, 4H, Ar(CH2CHdC(CH3)2)2), 1.77 (s, 6H, C(CH3)2),
1.75 (s, 6H, C(CH3)2); EI-MS m/z 264 (M+). Anal. Calcd for
P r en yla tion of 4-Ha lop h en ols 2a -c. Gen er a l P r oced u r e.
The corresponding p-substituted phenol (1.0 mmol) was dissolved
in 10% NaOH (2.4 mL per mol of p-substituted phenol), and
prenyl bromide (2.5 or 1.0 mol per mol of p-substituted phenol)
was added dropwise. After the reaction mixture was stirred at
room temperature for the indicated period of time, the mixture
was cooled to 0-5 °C in an ice-bath and then acidified with 2 M
CH3COOH. The mixture was extracted with Et2O and washed
with H2O. The Et2O layer was dried (MgSO4) and evaporated
under reduced pressure. Residues were purified by column
chromatography on silica gel (hexanes/Et2O, 20:1). After pre-
liminary examination of the eluates by TLC, the following major
products (in order of elution) were obtained.
C
16H21ClO: C, 72.57; H, 7.99. Found: C, 72.31; H, 7.88.
4-Br om o-2,6-d i(3-m eth yl-2-bu ten yl)p h en ol (3b): yellow
1
oil; H NMR (400 MHz, CDCl3) δ 7.08 (s, 2H, H3,H5-Ar), 5.34
(s, 1H, OH), 5.27 (t, J ) 7.3 Hz, 2H, Ar(CH2CHdC(CH3)2)2), 3.29
(d, J ) 7.3 Hz, 4H, Ar(CH2CHdC(CH3)2)2), 1.77 (s, 6H, C(CH3)2),
1.75 (s, 6H, C(CH3)2); EI-MS m/z 308 (M+). Anal. Calcd for
C16H21BrO: C, 62.14; H, 6.84. Found: C, 62.35; H, 6.87.
4-Iod o-2,6-d i(3-m eth yl-2-bu ten yl)p h en ol (3c): yellow oil;
1H NMR (400 MHz, CDCl3) δ 7.26 (s, 2H, H3,H5-Ar), 5.36 (s,
1H, OH), 5.26 (t, J ) 6.8 Hz, 2H, Ar(CH2CHdC(CH3)2)2), 3.27
(d, J ) 6.8 Hz, 4H, Ar(CH2CHdC(CH3)2)2), 1.77 (s, 6H, C(CH3)2),
1.75 (s, 6H, C(CH3)2); FAB-HRMS (glycerol matrix) m/z calcd
for C16H21IO [(M + H)+] 356.0637, found 356.0654. Anal. Calcd
for C16H21IO: C, 53.94; H, 5.94. Found: C, 53.79; H, 5.69.
4-Ch lor o-2,6-d i(3-m eth yl-2-bu ten yl)p h en yl (3-Meth yl-2-
1
bu ten yl) Eth er (4a ): colorless oil; H NMR (400 MHz, CDCl3)
δ 6.97 (s, 2H, H3,H5-Ar), 5.56 (t, J ) 7.1 Hz, 1H, OCH2CHd
C(CH3)2), 5.25 (t, J ) 7.3 Hz, 2H, Ar(CH2CHdC(CH3)2)2), 4.27
(d, J ) 7.1 Hz, 2H, OCH2CHdC(CH3)2), 3.34 (d, J ) 7.3 Hz, 4H,
Ar(CH2CHdC(CH3)2)2), 1.79 (s, 3H, CH3), 1.75 (s, 6H, C(CH3)2),
1.71 (s, 6H, C(CH3)2), 1.68 (s, 3H, CH3); EI-MS m/z 332 (M+).
Anal. Calcd for C21H29ClO: C, 75.76; H, 8.78. Found: C, 75.96;
H, 8.68.
4-Br om o-2,6-d i(3-m eth yl-2-bu ten yl)p h en yl (3-Meth yl-2-
bu ten yl) Eth er (4b): colorless oil; 1H NMR (400 MHz, CDCl3)
δ 7.12 (s, 2H, H3,H5-Ar), 5.56 (t, J ) 7.1 Hz, 1H, OCH2CHd
C(CH3)2), 5.25 (t, J ) 7.1 Hz, 2H, Ar(CH2CHdC(CH3)2)2), 4.27
(d, J ) 7.1 Hz, 2H, OCH2CHdC(CH3)2), 3.34 (d, J ) 7.1 Hz, 4H,
Ar(CH2CHdC(CH3)2)2), 1.79 (s, 3H, CH3), 1.75 (s, 6H, C(CH3)2),
1.71 (s, 6H, C(CH3)2), 1.68 (s, 3H, CH3); EI-MS m/z 376 (M+).
Anal. Calcd for C21H29BrO: C, 66.84; H, 7.75. Found: C, 67.03;
H, 7.57.
3-{4-H yd r oxy-3,5-d i(3-m et h yl-2-b u t en yl)p h en yl}-2(E)-
p r op en oic Acid (1). A solution of 3c (1.67 g, 4.69 mmol), methyl
acrylate (2.10 mL, 23.3 mmol), triethylamine (1.30 mL, 9.33
mmol), tri-o-tolylphosphine (143 mg, 0.470 mmol), and palladium
diacetate (52 mg, 0.232 mmol) in dry toluene (7 mL) was heated
at 100 °C for 20 h. The mixture was cooled to room temperature,
diluted with EtOAc/Et2O, and filtered through Celite. The
filtrate was washed with saturated NH4Cl solution. The aqueous
layer was reextracted with EtOAc/Et2O. The combined organic
layers were dried (MgSO4) and evaporated under reduced
pressure. Residues were purified by column chromatography on
silica gel (hexanes/EtOAc, 10:1) to give 521 mg (1.46 mmol, 31%)
of product 7 as a brown oil. KOH (15 mL of a 10% aqueous
solution) was added to a solution of the methyl ester 7 (274 mg,
0.769 mmol) in MeOH (15 mL). The mixture was heated under
reflux for 1 h, cooled to 0-5 °C, and acidified with 1 N HCl. The
MeOH was evaporated under reduced pressure, and the aqueous
residue was extracted with EtOAc. Extracts were washed with
saturated NH4Cl and brine, dried (MgSO4), and evaporated
under reduced pressure. The residue was purified by column
chromatography on silica gel (CHCl3/MeOH, 20:1) to give 181
mg (0.60 mmol, 78%) of product 1 as a white solid: mp 98-99
°C; 1H NMR (400 MHz, CDCl3) δ 7.69 (d, 1H, J ) 15.6 Hz,
ArCHd), 7.20 (s, 2H, H3,H5-Ar), 6.29 (d, 1H, J ) 15.6 Hz,
CHCO2H), 5.31 (t, 2H, J ) 6.8 Hz, Ar(CH2CHdC(CH3)2)2), 3.35
(d, 4H, J ) 6.8 Hz, Ar(CH2CHdC(CH3)2)2), 1.79 (s, 6H, C(CH3)2),
1.78 (s, 6H, C(CH3)2); FAB-HRMS (glycerol matrix) m/z calcd
for C19H24O3 [(M + H)+] 300.1725, found 300.1682. Anal. Calcd
for C19H24O3: C, 75.97; H, 8.05. Found: C, 75.75; H, 7.89.
4-Iod o-2,6-d i(3-m et h yl-2-b u t en yl)p h en yl (3-Met h yl-2-
1
bu ten yl) Eth er (4c): colorless oil; H NMR (400 MHz, CDCl3)
δ 7.31 (s, 2H, H3,H5-Ar), 5.55 (t, J ) 7.1 Hz, 1H, OCH2CHd
C(CH3)2), 5.24 (t, J ) 7.1 Hz, 2H, Ar(CH2CHdC(CH3)2)2), 4.27
(d, J ) 7.1 Hz, 2H, OCH2CHdC(CH3)2), 3.32 (d, J ) 7.1 Hz, 4H,
Ar(CH2CHdC(CH3)2)2), 1.79 (s, 3H, CH3), 1.75 (s, 6H, C(CH3)2),
1.71 (s, 6H, C(CH3)2), 1.68 (s, 3H, CH3); EI-MS m/z 424 (M+).
Anal. Calcd for C21H29IO: C, 59.44; H, 6.89. Found: C, 59.68;
H, 6.72.
1-Ch lor o-4-[(3-m eth yl-2-bu ten yl)oxy]ben zen e (5a ): color-
1
less oil; H NMR (400 MHz, CDCl3) δ 7.23 (d, J ) 9.3 Hz, 2H,
H3,H5-Ar), 6.84 (d, J ) 9.3 Hz, 2H, H2,H6-Ar), 5.49 (t, J ) 6.8
Hz, 1H, OCH2CHdC(CH3)2), 4.49 (d, J ) 6.8 Hz, 2H, OCH2CHd
C(CH3)2), 1.81 (s, 3H, CH3), 1.75 (s, 3H, CH3); EI-MS m/z 196
(M+). Anal. Calcd for C11H13ClO: C, 67.18; H, 6.66. Found: C,
67.22; H, 6.63.
Ack n ow led gm en t. We express our gratitude to
Yoshinori Nakagawa and Masaaki Hayami, Hayash-
ibara Biochemical Laboratories, for supplying artepillin
C isolated from Brazilian propolis. We thank M. Naka-
mura, E. Okayama, and K. Yamashita of our faculty for
excellent NMR spectral, elemental analysis, and mass
spectral techniques, respectively. Additional thanks are
given to the staff of the Center for Cooperative Research.
We thank Dr. Kenneth L. Kirk for his critical and
valuable comments.
1-Br om o-4-[(3-m eth yl-2-bu ten yl)oxy]ben zen e (5b): color-
less oil; H NMR (400 MHz, CDCl3) δ 7.36 (d, J ) 8.8 Hz, 2H,
H3,H5-Ar), 6.79 (d, J ) 8.8 Hz, 2H, H2,H6-Ar), 5.46 (t, J ) 6.6
Hz, 1H, OCH2CHdC(CH3)2), 4.47 (d, J ) 6.6 Hz, 2H, OCH2CHd
C(CH3)2), 1.79 (s, 3H, CH3), 1.73 (s, 3H, CH3); EI-MS m/z 240
(M+). Anal. Calcd for C11H13BrO: C, 54.79; H, 5.43. Found: C,
55.02; H, 5.42.
1-Iod o-4-[(3-m eth yl-2-bu ten yl)oxy]ben zen e (5c): colorless
oil; 1H NMR (400 MHz, CDCl3) δ 7.54 (d, J ) 8.8 Hz, 2H, H3,H5-
Ar), 6.69 (d, J ) 8.8 Hz, 2H, H2,H6-Ar), 5.46 (t, J ) 6.8 Hz, 1H,
OCH2CHdC(CH3)2), 4.47 (d, J ) 6.8 Hz, 2H, OCH2CHdC(CH3)2),
1.79 (s, 3H, CH3), 1.73 (s, 3H, CH3); EI-MS m/z 288 (M+). Anal.
Calcd for C11H13IO: C, 45.85; H, 4.55. Found: C, 46.05; H, 4.49.
1
J O0056904