Effective enzymatic caffeoylation of natural glucopyranosides

Article abstract of DOI:10.1016/j.bmcl.2016.02.010

Reaction system was developed for enzymatic caffeoylation of model saccharidic acceptor methyl β-D-glucopyranoside to obtain exclusively methyl 6-O-caffeoyl-β-D-glucopyranoside. Reaction with starting concentration of acceptor 0.2?M provided 73% yield of

Full text of DOI:10.1016/j.bmcl.2016.02.010

Accepted Manuscript
Effective Enzymatic Caffeoylation of Natural Glucopyranosides
Andrej Chyba, Vladimír Mastihuba, Mária Mastihubová
PII:
S0960-894X(16)30113-5
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.02.010
BMCL 23561
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
16 December 2015
3 February 2016
4 February 2016
Please cite this article as: Chyba, A., Mastihuba, V., Mastihubová, M., Effective Enzymatic Caffeoylation of Natural
Glucopyranosides, Bioorganic & Medicinal Chemistry Letters (2016), doi: http://dx.doi.org/10.1016/j.bmcl.
2016.02.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Effective Enzymatic Caffeoylation of Natural Glucopyranosides
Andrej Chybaa, Vladimír Mastihubaa, and Mária Mastihubováa,
a Laboratory of Biocatalysis and Organic Synthesis, Institute of Chemistry, Slovak Academy of Sciences,
Dúbravská cesta 9, SK-845 38 Bratislava, Slovakia
ABSTRACT
Reaction system was developed for enzymatic caffeoylation of model saccharidic acceptor methyl β-D-
glucopyranoside to obtain exclusively methyl 6-O-caffeoyl-β-D-glucopyranoside. Reaction with starting
concentration of acceptor 0.2 M provided 73% yield of purified product within 17 days. Reactions with low
acceptor concentrations (0.04 and 0.08 M) run to the completion within 7 days. Such highly effective and
regioselective reaction was promoted by Lipozyme TL IM in tert-butanol, using vinyl caffeate as acylation
donor. The optimized reaction conditions were used in preparative caffeoylation of natural substances – arbutin
and salidroside, giving 75% of 6-O-caffeoylated arbutin (robustaside B) and 74% of 6-O-caffeoylated salidroside
as the only products after 12 and 16 days, respectively.
Keywords:
Regioselective caffeoylation
Lipozyme TL IM
Robustaside B
6-O-Caffeoyl-salidroside
Phenylpropanoid glycosides
Caffeic acid occurs in nature as a constituent of more complex molecules widely distributed in
medicinal plants, fruits, coffee grains, propolis etc. These substances possess a broad spectrum of
health-beneficial biological activities,1-4 their extraction from natural sources and purification use to be
however laborious and costly and there is a growing demand for their effective chemical synthesis.
Starting with caffeic acid phenethyl ester (CAPE)5 through chicoric and chlorogenic acid6 to
phenylpropanoid glycosides,7-9 their synthesis appears more and more complicated in view of
complexity of their chemical structure.
Robustaside B (6-O-caffeoyl-arbutin, p-hydroxyphenyl 6-O-caffeoyl-β-D-glucopyranoside) isolated
from Viburnum dilatatum10 and 6-O-caffeoyl-salidroside (p-hydroxyphenetyl 6-O-caffeoyl-β-D-
glucopyranoside) isolated from Prunus grayana11 belong to natural substances with remarkable
biological activities. Robustaside B expressed a stronger anti-melanin activity and lower toxicity than
 Corresponding author. Tel.: +421-2-59410246; fax: +421-2-59410222; e-mail: chemjama@savba.sk

arbutin,12 and its antioxidant13 and antibacterial activities were documented.14 6-O-Caffeoyl-
salidroside exhibited antibacterial activities against S. aureus and cytotoxic activities against human
tumor Hela and Siha cell lines.15
Direct selective enzymatic acylation of a respective acceptor with caffeic acid provides therefore
simple and straightforward alternative to chemical procedures. Given the complexity of acceptor, the
appropriate biocatalyst may provide the necessary chemo- or regioselectivity in one reaction step.
Different phenylalkyl esters of caffeic acid have been recently synthetized with use of enzymes. The
esterification of caffeic acid by lipophilic acyl acceptor - phenethyl alcohol was catalyzed by action of
Candida antarctica lipase B (CAL B, Novozym 435) in high yields.5,16,17 The enzymatic synthesis of
3-cyclohexylpropyl caffeate, a mimetic of CAPE, was carried out by a two-step reaction starting from
5-caffeoylquinic acid and catalyzed by chlorogenate hydrolase from Aspergillus japonicus and CAL
B.18
On the other side, caffeoylation of hydrophilic polyhydroxylated acceptors represents a more
complex problem. The enzyme should be active in polar organic solvents, which are necessary for
dissolving of such acceptor while lipase-catalyzed esterifications preferably proceed in non-polar
hydrophobic environment. Esterification and transesterification of caffeic acid or its methyl ester with
glycerol using feruloyl esterase from Aspergillus niger was not successful.19 The first direct syntheses
of 4-O-hydroxycinnamoyl derivatives of quinic and shikimic acids were accomplished by
regioselective esterification with lipase A from Candida antarctica,20 but no successful enzymatic
caffeoylation of quinic acid derivatives was reported so far.
Previous studies of enzymatic acylation of glycosides by derivatives of hydroxycinnamic acids with
lipases or esterases were mainly focused on feruloylations.21-24 Only a few reports of enzymatic
caffeoylations of saccharides have appeared, usually with very low or zero yields of products. Otto et
al.25 used variety of non-activated arylaliphatic acids for acylation of glucose and salicin catalyzed by
immobilized CAL B. They failed to obtain esters of aromatic carboxylic acids or unsaturated
arylaliphatic acids, like cinnamic acid and its derivatives. López-Munguía et al.26 used successfully the
same enzyme to acylate analogues of salidroside with saturated and unsaturated hydroxycinnamic
acids. Chemical yields of the resulting esters were moderate or very low, reaching 40% of ferulate and
only less than 5% of the respective caffeate. Cultured cells from sweet potato (Ipomoea batatas)
caffeoylated isoquercitrin in 38% yield to produce the corresponding 6-O-caffeate using caffeic acid,
coenzyme A and ATP.27 The enzyme system catalyzing this synthesis is however based on plant
transferase and not lipase or esterase. Several previous reports28,29 have demonstrated relative
feasibility of the enzymatic feruloylation of arbutin and another phenolic glycosides30 with low to
moderate yields. Formation of caffeoylated salidroside catalyzed by Novozyme 435 (CAL B) was
recently reported, the proof of the product synthesis is however demonstrated only by TLC, without its

isolation and structural characterization.31 To our best knowledge, the preparative enzyme-catalyzed
caffeoylation of arbutin or salidroside has not been so far described.
In our previous work,32 we have prepared a large scale of vinyl esters of phenolic acids and their
analogues for studies of substrate specificity of Lipolase 100T (lipase from Thermomyces lanuginosus)
in acylations of methyl α-D-glucopyranoside as a model acceptor. The synthetic reactivity of Lipolase
corresponds to the hydrolytic activity of feruloyl esterase type A proposed by Crepin et al.33 For
example the acylation of the acceptor with vinyl ferulate yielded 68% of 6-O-feruloylated product
while use of vinyl caffeate afforded only 16% of 6-O-caffeoylated glucoside.
To select a more specific candidate for enzymatic caffeoylations of saccharides, we have recently
developed a spectrophotometric method for assay of caffeoyl esterase and used it for screening this
activity within a scale of lipases, esterases and crude industrial glycanases.34 This paper presents our
study of enzymatic caffeoylation of methyl β-D-glucopyranoside (1) as an inexpensive model acceptor
manifesting reactivity of hydroxyls comparable to natural β-D-glucopyranosides (Scheme 1). Vinyl
caffeate (2)32 and 2,2,2-trifluoroethyl caffeate (3) were prepared and used as activated donors. The
latter donor was synthesized as a new substance. The conditions optimized for acylation of 1 were
applied in preparative caffeoylations of arbutin (5) and salidroside (6).
Scheme 1. Enzymatic caffeoylation of methyl-β-D-glucopyranoside.
Screening of enzymes displaying caffeoyl esterase activity was realized by the method published
recently.34 The protocol was modified to assay also immobilized enzymes insoluble in water. The
screening selected five enzyme candidates for more detailed study, namely Lipex 100T, Lipolase
100T, Lipozyme TL IM, Pentopan 500BG, Liquanase 2.5L. The results are summarized in Table 1.
These six enzymes were tested by model transcaffeoylations monitored by HPLC. It is interesting,
that their performance in synthetic reaction did not correspond to their hydrolytic caffeoyl esterase
activity. The far most active enzyme, Pentopan 500 BG, produced less than 5% of the caffeoylated 1
after 10 days, while conversion of the same reaction catalyzed by Lipozyme TL IM, a lipase from
Thermomyces lanuginosus, reached over 54% in 8 days (Table 1). The hydrolytic assay method did
not allow us to reliably estimate caffeoyl esterase activity of Novozyme 435 due to its physical
properties. Anyway, this enzyme is known to catalyze acylations of several acceptors with various

phenolic acids25 and therefore we tested its performance in the studied acylation, in which was
however Novozyme 435 only moderately effective catalyst (Table 1).
Table 1
Caffeoyl esterase activity in selected enzyme preparations and reaction conversions in caffeoylation of
1 with vinyl caffeate after 8 days in acetonitrile
Enzyme
Microbial origin
Enzyme class
Caffeoyl Conversion
esterase
(U/g)
1.9
of 1 to 4
(%)
Lipex 100T
Thermomyces lanuginosus Lipase
18.5
3.3
Pentopan 500BG
Lipozyme TL IM
Novozym 435
Lipolase 100T
Liquanase 2.5La
Humicola insolens
Xylanase
360
Thermomyces lanuginosus Lipase
2.3
54.3
19.3
8.9
Candida antarctica
Thermomyces lanuginosus Lipase
Bacillus licheniformis Protease
Lipase
n.d.
0.9
28.0
22.6
aUltrafiltered and lyophilized enzyme
Lipozyme TL IM was our obvious choice of biocatalyst for further studies. Its catalytic
performance was significantly influenced by the nature of reaction media. While only half of
glucoside 1 reacted after 6 days in acetonitrile, its complete caffeoylation in less toxic tert-butanol was
observed within the same time interval (Fig. 1). Other solvents allowed moderate to low conversions.
The reactivity of Lipozyme TL IM in tert.-butanol was assessed also for use of 2,2,2-trifluoroethyl
caffeate 3 as caffeoyl donor. The conversion within 8 days was only 78 % (Fig. 1), which makes vinyl
caffeate 2 the more favorable donor for enzymatic caffeoylations.
Figure 1. Effect of reaction media on caffeoylation of methyl-β-D-glucopyranoside (1) by Lipozyme
TL IM with vinyl caffeate (2) or 2,2,2-trifluoroethyl caffeate (3) as time course of its conversion to 4.

Reaction conditions: 0.04 M methyl-β-D-glucopyranoside, 0.06 M vinyl caffeate, 0.4 g molecular
sieves, 0.4 g Lipozyme TL IM, 5 mL of solvent, 40°C, 300 rpm.
Higher acceptor levels influenced negatively the rate of the reaction, anyway, its starting
concentrations up to 0.2 M allowed to achieve conversion on the level of 90 % in 9 days (Fig. 2).
Figure 2. Effect of methyl-β-D-glucopyranoside (1) concentration on its caffeoylation by Lipozyme
TL IM (expressed as time course of conversion 1 to 4). Reaction conditions: 1.5 eq. vinyl caffeate, 0.4
g molecular sieves, 0.4 g Lipozyme TL IM, 5 mL tert-butanol, 40°C, 300 rpm.
The preparative caffeoylation of 1 was carried out under the conditions selected in previous
experiments. The glucoside (0.2 M) was treated with 1.5 equiv. of vinyl caffeate in tert-butanol under
catalysis of Lipozyme TL IM. The reaction time was extended to 17 days until all glucoside was
consumed (HPLC). The yield of isolated product was 73%. Its structural analysis by NMR
spectroscopy confirmed regioselective caffeoylation of 1 at the primary hydroxyl on C-6 of the
glucopyranoside ring. No formation of dicaffeates or positional isomers was observed. Such strict
regioselectivity in acylation of primary hydroxyl of acceptor is consistent with our previous results
from feruloylation of methyl α-D-glucopyranoside catalyzed by Lipolase 100T.30)
The enzymatic reaction system was then applied in preparation of two bioactive plant metabolites
harboring the caffeate moiety: arbutin 5 and salidroside 6 were caffeoylated under the conditions. The
reaction proceeded again with absolute regioselectivity, 6-O-caffeoylated arbutin - robustazide B (p-
hydroxyphenyl 6-O-caffeoyl-β-D-glucopyranoside, 7) was isolated after 12 days of reaction in 75 %
and 6-O-caffeoylated salidroside (p-hydroxyphenetyl 6-O-caffeoyl-β-D-glucopyranoside, 8) after 16
days in 74 % yield (Scheme 2).

Scheme 2. Enzymatic caffeoylation of arbutin and salidroside by Lipozyme TL IM.
In conclusion, we have optimized enzymatic caffeoylation of methyl β-D-glucopyranoside as a
model glycoside for syntheses of natural glycophenolics. Vinyl and 2,2,2-trifluoroethyl caffeates were
studied as donors. Lipozyme TL IM in tert-butanol using vinyl caffeate catalyzed regioselective
formation of methyl 6-O-caffeoyl-β-D-glucopyranoside with 100% substrate conversion. The
optimized caffeoylation conditions were verified on reaction with arbutin and salidroside as acceptors.
Robustaside B and 6-O-caffeoylated salidroside as the only caffeoylated products were isolated in
75% and 74 % yield. The results give rise to a green enzymatic access to caffeoylated glycosides as
medicinally important plant metabolites.
Acknowledgments
This work was supported by the Slovak Research and Development Agency under the contract No.
APVV-0846-12. This publication is also the result of the grant from the Slovak Grant Agency for
Science VEGA under the Project No. 2/0138/12 and of the projects implementation ‘Centre of
Excellence on Green Chemistry Methods and Processes (CEGreenI, contract No. 26240120001)’ and
‘Amplification of the Centre of Excellence on Green Chemistry Methods and Processes (CEGreenII,
contract No. 26240120025)’ supported by the Research & Development Operational Programme
funded by the ERDF.
Supplementary data
Supplementary data (analytical methods, experimental procedures, NMR data for compounds 3–6)
associated with this article can be found, in the online version, at…

References and notes
1. Mølgaard, P.; Ravn, H. Phytochemistry 1988, 27, 2411.
2. Silva, T.; Oliveira, C.; Borges, F. (2009 - 2013). Expert Opin. Ther. Patents 2014, 24, 1.
3. Magnani, C.; Isaac, V. L. B.; Correa, M. A.; Salgado, H. R. N. Anal. Methods, 2014, 6, 3203.
4. Calis, I. In Biodiversity; Sener B. Ed.; Springer US, 2002, pp 137 149
5. Zhang, P.; Tang, Y.; Li, N. G.; Zhu Y.; Duan J. A. Molecules 2014, 19, 16458.
6. Clifford, M. N. J. Sci. Food Agric. 2000, 80, 1033.
7. Kurkin, V. A. Chem. Nat. Comp. 2003, 39, 123.
8. Pan, J.; Yuan, C.; Lin, C.; Jia, Z.; Zheng, R. Pharmazie 2003, 58, 767.
9. Fu, G.; Pang, H.; Wong, Y. H. Curr. Med. Chem. 2008, 15, 2592.
10. Machida, K.; Nakano, Y.; Kikuchi, M. Phytochemistry 1991, 30, 2013.
11. Shimomura, H.; Sashida, Y.; Adachi, T. Phytochemistry 1987, 26, 249.
12. Xu, M.; Lao, Q.-C.; Zhao, P.; Zhu, X.-Y.; Zhu, H.-T.; Luo, X.-L.; Yang, C.-R.; He, J.-H.; Li,
C.-Q.; Zhang, Y.-J. Nat. Prod. Res. 2014, 28, 932.
13. Adisa, R. A.; Olorunsogo, O. O. Mol. Med. Rep. 2013, 8, 1493.
14. Kostadinova, E. P.; Alipieva, K. I.; Kokubun, T.; Taskova, R. M.; Handjieva, N. V.
Phytochemistry 2007, 68, 1321.
15. Zeng, X.; Wang, H.; Gong, Z.; Huang,J.; Pei, W.; Wang, X.; Zhang, J.; Tang, X. Fitoterapia
2015, 101, 153.
16. Chen, H. C.; Chen, J. H.; Chang, C.; Shieh, C.-J. Ultrason. Sonochem. 2011, 18, 455.
17. Wang, J.; Gu, S. S.; Cui, H. S.; Wu, X. Y.; Wu, F. A. Bioresource Technol. 2014, 158, 39.
18. Kurata, A.; Takemoto, S.; Fujita, T.; Iwai, K.; Furusawa, M.; Kishimoto, N. J. Mol. Catal. B:
Enzym. 2011, 69, 161.
19. Tsuchiyama, M.; Sakamoto, T.; Fujita, T.; Murata, S.; Kawasaki, H. BBA Gen. Subjects. 2006,
1760, 1071
20. Armesto, N.; Ferrero, M.; Fernández, S.; Gotor, V. J. Org. Chem. 2003, 68, 5784.
21. Topakas, E.; Vafiadi, C.; Christakopoulos, P. Process Biochem. 2007, 42, 497.
22. Koseki, T.; Fushinobu, S.; Ardiansyah, Shirakawa, H.; Komai, M. Appl. Microbiol.
Biotechnol. 2009, 84, 803.
23. Faulds, C. B. Phytochem. Rev. 2010, 9, 121.
24. Zeuner, B.; Kontogeorgis, G. M.; Riisager, A.; Meyer, A. S. New Biotechnol. 2012, 29, 256.
25. Otto, R. T.; Scheib, H.; Bornscheuer, U. T.; Pleiss, J.; Syldatk, C.; Schmid, R. D. J. Mol.
Catal. B: Enzym. 2000, 8, 201.
26. López-Munguía, A.; Hernández-Romero, Y.; Pedraza-Chaverri, J.; Miranda-Molina, A.;
Regla, I.; Martínez, A.; Castillo, E. PLoS ONE 2011, 6, e20115.

27. Nakajima, N.; Ishihara, K.; Hamada, H.; Kawabe, S. I.; Furuya, T. J. Biosci. Bioeng. 2000, 90,
347.
28. Chigorimbo-Murefu, N. T. L.; Riva, S.; Burton, S. G. J. Mol. Catal. B: Enzym. 2009, 56, 277.
29. Ishihara, K.; Katsube, Y.; Kumazawa, N.; Kuratani, M.; Masuoka, N.; Nakajima, N. J. Biosci.
Bioeng. 2010, 109, 554.
30. Mastihubová, M.; Mastihuba, V.; Bilaničová, D.; Boreková, M. J. Mol. Catal. B: Enzym.
2006, 38, 54.
31. Yu, H.-L.; Xu, J.-H.; Su, J.-H.; Lu, W.-Y.; Lin, G.-Q. J. Biosci. Bioeng. 2008, 106, 65.
32. Mastihubová, M.; Mastihuba, V. Bioorg. Med. Chem. Lett. 2013, 23, 5389.
33. Crepin, V. F.; Faulds, C. B.; Connerton, I. F. Appl. Microb. Biotechnol. 2004, 63, 647.
34. Chyba, A.; Mastihubová, M.; Mastihuba, V. Anal. Biochem. 2014, 445, 49.