Regioselective synthesis of piceatannol from resveratrol: Catalysis by two-component flavin-dependent monooxygenase HpaBC in whole cells

Article abstract of DOI:10.1016/j.tetlet.2014.03.076

Piceatannol, a valuable biologically active stilbene derivative, was efficiently synthesized from resveratrol. Whole-cell catalysis with HpaBC monooxygenase enabled the regioselective hydroxylation of resveratrol to produce 23 mM (5.2 g L^-1) of piceatannol.

Full text of DOI:10.1016/j.tetlet.2014.03.076

Tetrahedron Letters 55 (2014) 2853–2855
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Regioselective synthesis of piceatannol from resveratrol: catalysis
by two-component flavin-dependent monooxygenase HpaBC
in whole cells
⇑
Toshiki Furuya , Kuniki Kino
Department of Applied Chemistry, Faculty of Science and Engineering, Waseda University, 3-4-1 Ohkubo, Shinjuku-ku, Tokyo 169-8555, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Piceatannol, a valuable biologically active stilbene derivative, was efficiently synthesized from resvera-
trol. Whole-cell catalysis with HpaBC monooxygenase enabled the regioselective hydroxylation of resve-
ratrol to produce 23 mM (5.2 g L ) of piceatannol.
Received 14 February 2014
Revised 15 March 2014
Accepted 18 March 2014
Available online 26 March 2014
ꢀ1
Ó 2014 Elsevier Ltd. All rights reserved.
Keywords:
HpaBC
Hydroxylation
Monooxygenase
Piceatannol
0
0
Piceatannol (3,4,3 ,5 -tetrahydroxy-trans-stilbene) is a biologi-
cally active molecule that naturally occurs in several plants, includ-
ing grapes, Japanese knotweed, and passion fruit.1 This molecule
is a powerful antioxidant. It exhibits much higher antioxidant
regioselectively hydroxylate resveratrol might provide a new, effi-
cient, and cost-effective synthetic approach to piceatannol
(Scheme 1). This oxidation reaction cannot be performed by chem-
ical methods, and there have been only a few reports of biocata-
lysts (i.e., enzymes and microorganisms) that can oxidize
resveratrol to piceatannol. Human and bacterial cytochrome
P450s were shown on an analytical scale to be capable of this cat-
–3
activity in radical scavenging experiments than resveratrol
3,5,4 -trihydroxy-trans-stilbene), an extensively studied analog
of piceatannol. Furthermore, piceatannol can suppress prolifera-
tion of a variety of tumor cells, including leukemia, lymphoma, and
0
(
4
,5
1
0–12
alytic conversion, but with very low activity.
It was also re-
6
,7
melanoma.
In addition to having antioxidant and anticancer
ported that Streptomyces avermitilis converted 78% of a 100 lM
1
3
activities, recent reports indicate that piceatannol not only exerts
positive effects on human dermal cells through inhibition of mela-
nogenesis and synthesis of collagen, but also exhibits vasorelaxant
effects in rat thoracic aorta.8,9 Notably, these effects of piceatannol
are more pronounced than those of resveratrol. These beneficial
properties encourage the use of piceatannol in health and func-
tional foods, as well as in pharmaceuticals and cosmetics. Indeed,
an extract of passion fruit seed that contains concentrated picea-
tannol is marketed under the trade name Passienol (Morinaga &
Co., Ltd., Japan).
resveratrol solution to piceatannol.
In this study, we found that the two-component flavin-depen-
dent monooxygenase, HpaBC, from Pseudomonas aeruginosa has
catalytic activity for the regioselective hydroxylation of resveratrol
to piceatannol. Enzymes in the HpaBC family consist of
a
flavin-dependent monooxygenase (HpaB) and a NAD(P)H:flavin
oxidoreductase (HpaC), and catalyze hydroxylation using
molecular oxygen as an innocuous oxidant under ambient
1
4–16
conditions.
HpaBC monooxygenases are involved in the
4-hydroxyphenylacetate degradation pathway. These enzymes
reportedly catalyze the hydroxylation of monocyclic aromatic
Piceatannol has been found in several plants, but generally at
1
–3
very low levels,
extremely expensive. In contrast, resveratrol is
inexpensive compound common in many plants and at higher
making naturally extracted piceatannol
compounds, including 4-hydroxyphenylacetate,
L
-tyrosine, and
1
6–20
a
relatively
halophenols.
Recently, we reported that HpaBC from
21
P. aeruginosa can synthesize hydroxycinnamic acids. While the
present Letter was in preparation, a review article was published
that contained a short sentence regarding unpublished data; it
indicated that HpaBC from Escherichia coli exhibits oxidation
1
–3
concentrations.
Thus, oxidation catalysts that are able to
2
2,23
⇑
Corresponding author. Tel.: +81 3 5286 3211; fax: +81 3 3232 3889.
E-mail address: tfuruya@fuji.waseda.jp (T. Furuya).
activity for resveratrol.
However, no research papers describing
http://dx.doi.org/10.1016/j.tetlet.2014.03.076
040-4039/Ó 2014 Elsevier Ltd. All rights reserved.
0

2
854
T. Furuya, K. Kino / Tetrahedron Letters 55 (2014) 2853–2855
OH
OH
OH
mAU
2
HpaBC
HO
HO
150
125
100
75
1
OH
OH
5
2
0
5
0
Resveratrol
Piceatannol
Scheme 1. Regioselective hydroxylation of resveratrol to piceatannol by HpaBC.
0
5
10
15
Time (min)
piceatannol synthesis by any enzyme in the HpaBC family have been
published. Here, we examined in detail the catalytic activity of
HpaBC from P. aeruginosa toward resveratrol, and investigated the
application of the HpaBC whole-cell catalyst to the flask-scale
production of piceatannol.
Figure 1. HPLC chromatogram of the reaction of HpaBC whole-cell catalyst with
resveratrol. Peaks 1 and 2 were found to correspond to resveratrol and piceatannol,
respectively.
or Triton X-100 to the reaction mixture significantly enhanced
the productivity of piceatannol (Table 1, entries 6–9). In particular,
the whole cells produced 25 mM piceatannol in the presence of 1%
We explored the catalytic potential of HpaBC for resveratrol
hydroxylation by whole-cell assays using E. coli cells expressing
HpaBC (experimental details are shown in Supplementary data).
(
v/v) Tween 80, with a conversion yield of 83% (Table 1, entry 7).
2
1
Reaction conditions previously optimized for p-coumaric acid
were first employed (Table 1, entry 1). When resveratrol
10 mM) was reacted with the transformed E. coli cells (8 mg of
dry cell weight per milliliter) in a microtube containing 100 l of
This amount of product was 2.8 times higher than that in the
absence of any detergents.
(
Finally, we attempted to produce piceatannol on a flask scale
using the optimized reaction conditions. Resveratrol (30 mM)
l
the reaction mixture, HPLC analysis showed a major peak (reten-
tion time, 5.9 min) in addition to the substrate peak (6.3 min)
ꢀ1
was reacted with the HpaBC whole-cell catalyst (32 mg mL ) in
a 500-ml flask that contained 20 ml of the reaction mixture. The
time course of the conversion is shown in Figure 2. The initial rate
of resveratrol hydroxylation by the whole cells in the presence of
(Fig. 1). This new peak was not detected in the reaction mixture
containing E. coli cells carrying the empty vector without the hpaBC
genes (data not shown). The compound corresponding to this peak
was confirmed to be a monooxygenation product of resveratrol
1
% (v/v) Tween 80 was estimated as 3.2 lmol per g of cells per
min. This rate was approximately 1.5 times higher than that in
ꢀ
based on the determination of its mass value ([MꢀH] ,
1
13
m/z = 243.2). Furthermore, the H NMR and C NMR spectra of
this product were in agreement with those of piceatannol that had
been previously determined (see Supplementary data).2
4,25
Based
30
25
20
on these observations, this new product was identified as piceatan-
nol (Scheme 1). The whole cells regioselectively hydroxylated the
C-3 position of resveratrol without any detectable by-products.
Using these reaction conditions (Table 1, entry 1), 4.3 mM
piceatannol was produced from 10 mM resveratrol in 24 h.
0
We next examined reaction conditions to further enhance the
productivity of piceatannol (Table 1). As the concentrations of res-
veratrol substrate and the catalyst cells increased, the amount of
piceatannol product increased (Table 1, entries 1–5). When
1
5
0
5
0
1
ꢀ
1
3
0 mM resveratrol was reacted with 32 mg mL
formed E. coli cells, the production of piceatannol reached
.8 mM (Table 1, entry 5); however, the conversion yield de-
of the trans-
8
creased. We reasoned that the relatively large molecular size of
resveratrol might hamper the entrance of this substrate into
E. coli cells. Non-ionic detergents such as Tween 80 and Triton
X-100 reportedly improve the permeability of bacterial cell
membranes for substrate compounds without significantly
0
2
4
6
8
10
12
Time (h)
Figure 2. Flask-scale production of piceatannol by HpaBC whole-cell catalyst. The
whole-cell catalyst was reacted with resveratrol in the absence (white symbols) or
in the presence (black symbols) of Tween 80 (1%). The time courses of resveratrol
consumption (squares) and of piceatannol production (circles) are shown.
2
6,27
damaging the cells.
We found that the addition of Tween 80
Table 1
Production of piceatannol from resveratrol by a HpaBC whole-cell catalyst
Entry
Substrate (mM)
Catalysta (mg mLꢀ1
)
Detergent (% v/v)
Product (mM)
Conversionb (%)
1
2
3
4
5
6
7
8
9
10
10
20
20
30
30
30
30
30
8
0
0
0
0
4.3
6.2
7.2
8.1
8.8
13
25
9.9
22
43
62
36
41
29
43
83
33
73
16
16
32
32
32
32
32
32
0
0.1 (Tween 80)
1 (Tween 80)
0.1 (Triton X-100)
1 (Triton X-100)
a
Milligram of dry cell weight per milliliter.
Conversion (%) = product (mM)/substrate (mM) ꢁ 100.
b

T. Furuya, K. Kino / Tetrahedron Letters 55 (2014) 2853–2855
2855
the absence of the detergent. In the presence of Tween 80, the
production of piceatannol continued to increase and reached
4. Murias, M.; Jäger, W.; Handler, N.; Erker, T.; Horvath, Z.; Szekeres, T.; Nohl, H.;
Gille, L. Biochem. Pharmacol. 2005, 69, 903–912.
5.
Ovesná, Z.; Kozics, K.; Bader, Y.; Saiko, P.; Handler, N.; Erker, T.; Szekeres, T.
ꢀ
1
2
3 mM (5.2 g L ) in 12 h (Fig. 2). This production is the highest
Oncol. Rep. 2006, 16, 617–624.
level so far attained by any biocatalytic synthesis of piceatannol,
6. Niles, R. M.; Cook, C. P.; Meadows, G. G.; Fu, Y. M.; McLaughlin, J. L.; Rankin, G.
O. J. Nutr. 2006, 136, 2542–2546.
and is 300 times higher than that achieved by cells of S.
7
.
Tolomeo, M.; Grimaudo, S.; Di Cristina, A.; Roberti, M.; Pizzirani, D.; Meli, M.;
Dusonchet, L.; Gebbia, N.; Abbadessa, V.; Crosta, L.; Barucchello, R.; Grisolia, G.;
Invidiata, F.; Simoni, D. Int. J. Biochem. Cell Biol. 2005, 37, 1709–1726.
Matsui, Y.; Sugiyama, K.; Kamei, M.; Takahashi, T.; Suzuki, T.; Katagata, Y.; Ito,
T. J. Agric. Food Chem. 2010, 58, 11112–11118.
1
3
avermitilis.
In conclusion, we demonstrated that HpaBC from P. aeruginosa
is useful for the high-yield and regioselective synthesis of piceatan-
8
9
.
.
nol from resveratrol. Piceatannol can be also prepared by chemical
Sano, S.; Sugiyama, K.; Ito, T.; Katano, Y.; Ishihata, A. J. Agric. Food Chem. 2011,
59, 6209–6213.
synthesis through other routes,2
4,25
as well as by extraction from
1
1
1
0. Kim, D. H.; Ahn, T.; Jung, H. C.; Pan, J. G.; Yun, C. H. Drug Metab. Dispos. 2009, 37,
32–936.
1. Piver, B.; Fer, M.; Vitrac, X.; Merillon, J. M.; Dreano, Y.; Berthou, F.; Lucas, D.
Biochem. Pharmacol. 2004, 68, 773–782.
2. Potter, G. A.; Patterson, L. H.; Wanogho, E.; Perry, P. J.; Butler, P. C.; Ijaz, T.;
Ruparelia, K. C.; Lamb, J. H.; Farmer, P. B.; Stanley, L. A.; Burke, M. D. Br. J. Cancer
plant sources; however, our strategy provides an efficient and
green alternative for the production of piceatannol. Moreover, it
is noteworthy that HpaBC is able to hydroxylate resveratrol, a rel-
atively large molecule that includes two benzene rings, since en-
zymes in the HpaBC family had been believed to exhibit catalytic
activity only for monocyclic aromatic compounds. Thus, the find-
ings of this study open up possibilities for further exploitation of
the catalytic potential of HpaBC. For example, the approach pre-
sented here might be also applicable to the oxyfunctionalization
of other valuable polycyclic compounds, for example stilbenes
and flavonoids.
9
2002, 86, 774–778.
13. Lee, N.; Kim, E. J.; Kim, B. G. ACS Chem. Biol. 2012, 7, 1687–1692.
14. Chakraborty, S.; Ortiz-Maldonado, M.; Entsch, B.; Ballou, D. P. Biochemistry
2010, 49, 372–385.
1
1
5. Ellis, H. R. Arch. Biochem. Biophys. 2010, 497, 1–12.
6. Galán, B.; Díaz, E.; Prieto, M. A.; García, J. L. J. Bacteriol. 2000, 182, 627–636.
17. Coulombel, L.; Nolan, L. C.; Nikodinovic, J.; Doyle, E. M.; O’Connor, K. E. Appl.
Microbiol. Biotechnol. 2011, 89, 1867–1875.
1
1
8. Lee, J. Y.; Kun, L. Biotechnol. Lett. 1998, 20, 479–482.
9. Liebgott, P. P.; Amouric, A.; Comte, A.; Tholozan, J. L.; Lorquin, J. Res. Microbiol.
2
009, 160, 757–766.
Supplementary data
2
2
0. Lin, Y.; Yan, Y. Microb. Cell Fact. 2012, 11, 42.
1. Furuya, T.; Kino, K. Appl. Microbiol. Biotechnol. 2014, 98, 1145–1154.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2014.03.
22. Lin, Y.; Jain, R.; Yan, Y. Curr. Opin. Biotechnol. 2014, 26, 71–78.
23. The genome of E. coli strain MG1655, which was used as a template for the PCR
amplification of hpaBC in the Refs. 20 and 22, contains no hpaBC genes on the
basis of the published genome sequence, although that of E. coli strain W used
in the Refs. 16–18 contains these genes.
0
76.
References and notes
24. Murias, M.; Handler, N.; Erker, T.; Pleban, K.; Ecker, G.; Saiko, P.; Szekeres, T.;
Jäger, W. Bioorg. Med. Chem. 2004, 12, 5571–5578.
2
5. Sun, H. Y.; Xiao, C. F.; Cai, Y. C.; Chen, Y.; Wei, W.; Liu, X. K.; Lv, Z. L.; Zou, Y.
1
2
.
.
Piotrowska, H.; Kucinska, M.; Murias, M. Mutat. Res. 2012, 750, 60–82.
Benová, B.; Adam, M.; Onderková, K.; Královsk y´ , J.; Krajícek, M. J. Sep. Sci. 2008,
Chem. Pharm. Bull. 2010, 58, 1492–1496.
2
2
6. Ko, H.-J.; Bang, W.-G.; Kim, K. H.; Choi, I.-G. Biotechnol. Lett. 2012, 34, 677–682.
7. Mitsukura, K.; Kondo, Y.; Yoshida, T.; Nagasawa, T. Appl. Microbiol. Biotechnol.
31, 2404–2409.
3
.
Cantos, E.; Espín, J. C.; Fernández, M. J.; Oliva, J.; Tomás-Barberán, F. A. J. Agric.
Food Chem. 2003, 51, 1208–1214.
2006, 71, 502–504.