Received: March 17, 2015 | Accepted: April 18, 2015 | Web Released: April 25, 2015
CL-150237
Enzymatic Allylation of Catechols
Yixin Zhang,1,3 Wujun Liu,1 Muhammad Sohail,1 Xueying Wang,1,3 Yuxue Liu,1,3 and Zongbao K. Zhao*1,2
1Division of Biotechnology, Dalian Institute of Chemical Physics, Dalian 116023, P. R. China
2State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Dalian 116023, P. R. China
3University of Chinese Academy of Sciences, 19A Yuquanlu, Beijing 100049, P. R. China
(E-mail: zhaozb@dicp.ac.cn)
Enzymatic allylation of catechols was realized via catechol
O-methyltransferase (COMT) using an allylated S-adenosyl-
L-methionine (allyl-SAM) analog, with relatively good chemo-
and regioselectivities. This new reaction offered an alternative
procedure for allylation of catechols, which can be expanded as
a biocatalytic allylation method in organic synthesis.
Catechol O-MTases (COMTs) are primarily responsible for the
methylation of one hydroxy group of a catechol structure.
COMTs especially human-soluble COMT (hsCOMT) are widely
studied due to their importance as drug targets.15 Large-scale
production of active hsCOMT was once reported on fermenta-
tion conditions for pharmaceutical trials.16 Microorganisms have
been engineered to overproduce COMT for the production of
highly valuable metabolites.17,18 The enormous biocatalytic
potential of COMTs, such as their accessible source, good
chemo- and regioselectivities, and wide substrate scope,19,20
encourage us to expand their catalytic capacities to allylation
of different catechol derivatives. In this paper, we report the
allylation of catechols by hsCOMT using allyl-SAM as the allyl
donor.
We overexpressed hsCOMT in engineered Escherichia coli
cells and purified the recombinant protein.21 Allyl-SAM was
synthesized in gram-scale from readily available chemicals
according to a reported procedure.5 Our initial experiment was
performed at 37 °C for 15 h in a 250-μL reaction mixture of
hsCOMT (1.92 mg mL¹1, 78.7 ¯M), allyl-SAM (400 ¯M), and
catechol 1a (800 ¯M), and the reaction was followed by TLC
analysis.22 The corresponding allylated product was formed over
time. When the reaction was performed on a larger scale with
crude lysate-expressed hsCOMT, the product was isolated in
48.3% yield (Table 1). Under the standard reaction conditions,
allylation was also observed for compounds 1b-1i and the
products were successfully isolated regardless the substituents
on the aromatic nucleus. The yield can be improved as an
increased conversion of 1a was detected by adding more
enzymes. Although hsCOMT showed a wide substrate scope,
a catechol-containing structure remained essential. Thus, no
allylation products were detected in case of salicylic acid,
phenylenediamine, and phenol.
Allylation is a valuable reaction in organic synthesis, as it
introduces an allyl group that can be further modified by various
transformation protocols. Although several synthetic methods
have been developed for allylation on C-, O-, N-, or S-centers,1-3
no natural counterpart has been found in the biological system.
Recently, the cofactor specificity of methyltransferases (MTases)
have been relaxed to include S-adenosyl-L-methionine (SAM)
analogs, leading to alkylation of biomacromolecules including
protein, DNA, and RNA.4-7 In particular, an allylated SAM
(allyl-SAM, Scheme 1) analog has been used as a SAM
surrogate by some wild-type protein MTases for protein label-
ing.5 As allyl-SAM and SAM share high structural similarity, it
is intriguing to explore allyl-SAM as an allyl donor by a wider
spectrum of MTases for enzymatic allylation of small molecules.
However, only a few examples have been reported on this
subject. Two C-MTases, NovO and CouO, found in the
biosynthesis pathway of the antibiotics coumermycin A1 and
novobiocin, respectively, mediate Friedel-Crafts allylation on
coumarin intermediate using allyl-SAM as the cofactor.8 In
another example, a sugar O-MTase, RebM, catalyzes the O-
allylation of a rebeccamycin congener in an enzyme-coupled
system while allyl-SAM was produced in situ from ATP and
S-allylhomocysteine.9 In these examples, the scope of the allyl
acceptor, i.e., the substrate of the corresponding MTase, was
not explored, most probably because it was difficult to access
structurally diversified acceptors.
The allylation reaction showed good chemoselectivity. In
protocatechuic acid (1d) and dopamine (1e) (Table 1, Entries 4
and 5), only one of the two hydroxys was allylated, and no
allylation was found on the carboxyl group or the amino group.
Thus, only mono-allylated products were obtained regardless the
presence of excessive allyl-SAM. This provided an attractive
route for the preparation of mono-allylated catechol derivatives
for other applications. These results indicated that enzymatic
allylation was advantageous over regular chemical synthesis as
tedious and costly protection-deprotection procedures may be
omitted.
Regioselectively alkylated catechols are important entities
for the preparation of bioactive natural products and drug
intermediates.10,11 Chemical synthesis of allylated catechols are
associated with several disadvantages such as low selectivity
and tedious protection-deprotection procedures.12-14 Therefore,
enzymatic allylation of catechols would be of particular interest.
CO2H
N
CO2H
N
H2N
H2N
NH2
N
NH2
N
It is known that meta-methylation is preferred by COMTs.19
For protocatechuic aldehyde 1c and 1d, our results indicated that
hsCOMT afforded the meta-methylation product and the para-
product in the ratios of 2.1:1 and 3.0:1, respectively, which
were consistent with literature results.23,24 To determine whether
hsCOMT followed a similar regioselectivity for allylation, we
S+
S+
N
O
N
O
N
N
OH OH
SAM
OH OH
Allyl-SAM
Scheme 1. Chemical structures of SAM and allyl-SAM.
© 2015 The Chemical Society of Japan | 949
Zhang, Yixin
