Glycosidase-Catalyzed Synthesis of Glycosyl Esters and Phenolic Glycosides of Aromatic Acids

Article abstract of DOI:10.1002/adsc.201900259

Phenolic glycosides occur naturally in many plants and as such are often present in the human diet. Their isolation from natural sources is usually laborious due to their presence in complex matrices. Their chemical and enzymatic syntheses have been found complex, time-consuming, and costly, yielding only small amounts of glycosylated products. In quest of a convenient biocatalytic route to structurally complex phenolic glycosides, we discovered that the rutinosidase from Aspergillus niger not only efficiently converts hydroxylated aromatic acids (e. g. coumaric and ferulic acids) into the respective phenolic rutinosides, but surprisingly also catalyzes the formation of the respective glycosyl esters. We report here the results of a systematic study presenting the unique synthesis of naturally occurring glycosyl esters and phenolic glycosides accomplished by glycosidase catalysis. A panel of aromatic acids was tested as glycosyl acceptors and the crucial structural features required for the formation of glycosyl esters were identified. In the light of the present structure-activity relationship study, a plausible reaction mechanism was proposed. All the products were fully structurally characterized by NMR and MS. (Figure presented.).

Full text of DOI:10.1002/adsc.201900259

Title: Glycosidase-Catalyzed Synthesis of Glycosyl Esters and
Phenolic Glycosides of Aromatic Acids
Authors: Ivan Bassanini, Jana Kapesova, Lucie Petraskova, Helena
Pelantova, Kristina Markosova, Martin Rebros, Katerina
Valentova, Michael Kotik, Kristyna Kanova, Pavla Bojarová,
Josef Cvacka, Lucie Turkova, Erica E. Ferrandi, Ikram
Bayout, Sergio Riva, and Vladimir Kren
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Adv. Synth. Catal. 10.1002/adsc.201900259
Link to VoR: http://dx.doi.org/10.1002/adsc.201900259

10.1002/adsc.201900259
Advanced Synthesis & Catalysis
VERY IMPORTANT PUBLICATION
FULL PAPER
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Glycosidase-Catalyzed Synthesis of Glycosyl Esters and
Phenolic Glycosides of Aromatic Acids
Ivan Bassanini,b,f Jana Kapešová,a Lucie Petrásková,a Helena Pelantová,a Kristína
Markošová,c Martin Rebroš,c Kateřina Valentová,a Michael Kotik,a Kristýna Káňová,a
Pavla Bojarová,a Josef Cvačka,e Lucie Turková,a Erica E. Ferrandi,b Ikram Bayout,b,d
Sergio Riva,b Vladimír Křena,*
a
Institute of Microbiology of the Czech Academy of Sciences, Vídeňská 1083, CZ 14220, Prague 4, Czech Republic
Phone: (+420) 296442510; Fax: (+420) 244471286; E-mail: kren@biomed.cas.cz
Istituto di Chimica del Riconoscimento Molecolare, Consiglio Nazionale delle Ricerche, Via Mario Bianco 9, I 20131
b
Milano, Italy
c
Institute of Biotechnology, Slovak University of Technology, Radlinského 9, SK 81237, Bratislava, Slovakia
Asymmetric Catalysis Laboratory (LCAE), Badji Mokhtar Annaba-University, B.P. 12, 23000 Annaba, Algeria
Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences, Flemingovo nám. 2, CZ 16610,
d
e
Prague 6, Czech Republic
f
Dipartimento di Scienze Farmaceutiche, Università degli Studi di Milano, Via Mangiagalli 25, I 20131 Milano, Italy
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.
Abstract. Phenolic glycosides occur naturally in many We report here the results of a systematic study presenting
plants and as such are often present in the human diet. Their the unique synthesis of naturally occurring glycosyl esters
isolation from natural sources is usually laborious due to and phenolic glycosides accomplished by glycosidase
their presence in complex matrices. Their chemical and catalysis. A panel of aromatic acids was tested as glycosyl
enzymatic syntheses have been found complex, time- acceptors and the crucial structural features required for the
consuming, and costly, yielding only small amounts of formation of glycosyl esters were identified. In the light of
glycosylated products. In quest of a convenient biocatalytic the present structure-activity relationship study, a plausible
route to structurally complex phenolic glycosides, we reaction mechanism was proposed. All the products were
discovered that the rutinosidase from Aspergillus niger not fully structurally characterized by NMR and MS.
only efficiently converts hydroxylated aromatic acids (e.g.
coumaric and ferulic acids) into the respective phenolic Keywords: natural product; glycosylation; rutinosidase;
rutinosides, but surprisingly also catalyzes the formation of carboxylic glycoside; coumaric acid; ferulic acid
the respective glycosyl esters.
sources but most of them are not readily obtainable
due to their occurrence in complex matrices.
Introduction
Phenolic acids and their glycosides are frequently
found as acyl moieties in esters of a plethora of
polyphenols. Phenolic acids form esters either with
phenolic OH groups or, in some cases, with the
glycosidic moieties of polyphenol glycosides.[1]
Acylation of polyphenol glycosides can be well
accomplished (at their glycone moiety) using
chemical or chemoenzymatic synthesis.[2] Glycosides
of hydroxycinnamic acid derivatives were first
described in the sixties of the last century[3] and later
they were found to be ubiquitous in plants.[4a,b] Then
Hydroxyphenyl propenoic acids (hydroxycinnamic
acids), e.g. ferulic acid(s), coumaric acids, sinapic
acid, caffeic acid, etc., are common components of
the human diet. These compounds also occur in
plants in the form of various glycosides, such as
glucosides, galactosides, rhamnosides, arabinosides,
and rutinosides. As the diets rich in polyphenols have
repeatedly been related to low incidence of
cardiovascular, neurodegenerative, and oncological
diseases, various food supplements containing these
compounds are becoming increasingly popular
among the general population. Some phenolic
glycosides can quite easily be isolated from natural
1-O-hydroxycinnamoyl-β-glucopyranoses
were
identified to be substrates of acyltransferases (known
as serine carboxypeptidase-like acyltransferases) and
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900259
Advanced Synthesis & Catalysis
in plants they serve as activated acyl donors (besides
acyl-CoA) for the biosynthesis of many secondary
metabolites.[4c] They also play an important role in
cell-wall lignification. There is a limited number of
in-detail-described examples (with full spectral
identification) showing that the carboxyl of the acid
is involved in the formation of the glycosidic linkage
(glycosyl ester). A series of ferulic acid glycosides, as
were given. Here it should be noted that sucrose
phosphorylases have rather different reaction
mechanism than glycosidases and they can produce
only -glucopyranosides, therefore they are not as
flexible as glycosidases.
Glycosylation of phenolic OH groups is a rather
challenging reaction especially for glycosidases.[11]
Leloir glycosyltransferases that synthesize phenolic
glycosides in vivo can be used for preparatory in vitro
synthesis of phenolic glycosides as well;[15] on the
other hand, there exist only a few examples of phenol
glycosylation with non-Leloir glycosyltransferases
such as phosphorylases.[16] These enzymes use
sucrose as an inexpensive glycosyl donor; however,
they typically synthesize only α-glucopyranosides.
Recently, we described a robust diglycosidase,
which glycosylated a number of phenolic acceptors.
This enzyme, the rutinosidase from Aspergillus niger
(EC 3.2.1.168, CAZY GH 5_23), is capable of
transferring rutinosyl (6-O-α-L-rhamnopyranosyl-β-
D-glucopyranosyl) residue onto various alcohols, but
most interestingly also to phenolic acceptors, which
are generally difficult to glycosylate with
e.g.
1-O-(E)-
and
(Z)-feruloyl-β-sophorose
(tuberonoid A and B, respectively), was isolated from
Allium tuberosum (garlic chives) leaves.[5] 1-O-(E)-
Caffeoyl-β-rutinose (swertiamacroside) was isolated
from the plant Swertia macrosperma;[6] a mixture of
the (E)- and 1-O-(Z)-p-coumaroyl-β-rutinoses was
isolated from an acetone extract of aspen knot
wood.[7] Recently, 1-O-acyl-β-glucopyranose, which
is a metabolite of the synthetic herbicide 2,4-
dichlorophenoxyacetic acid (2,4-D) was described in
wild radish (Raphanus raphanistrum).[8]
Glycosylated hydroxycinnamates are generally not
easily accessible. Their isolation from plant material
is tedious, time demanding and low-yielding. Kosuge
and Conn described the glucosylation of coumaric
acid after feeding it to higher plants such as
Meliototus alba and Hierochloe odorata.[9]
Alternatively, these compounds can be produced in
vitro, exploiting either classical organic synthesis
protocols or biocatalytic approaches. For instance a
5-step synthesis of the above-mentioned 1-O-
glycosidases.[17]
A
major advantage of this
transglycosylation reaction is the use of inexpensive
and biocompatible rutin (1) as the glycosyl donor,
while quercetin (2) as a byproduct of the reaction
precipitates and can be easily removed from the
reaction mixture by filtration. The produced
rutinosides can be conveniently transformed in situ
via a telescoping reaction with α-L-rhamnosidase to
yield the respective β-glucopyranosides.[18] These
biotransformations are of interest because certain
industrial applications of carboxy group-containing
compounds are limited by the strong sour smell and
taste and also by their low solubility.[16a]
feruloyl-
and
1-O-β-sinapoyl-β-glucopyranoses
yielded gram amounts of these compounds.[10]
Enzymatic methods are a straightforward route to
the glycosylation of various natural products mostly
due to their simplicity (single-step methods), stereo-,
and regioselectivity.[11] Enzymatically produced
glycosides are usually obtained under mild conditions
in quite pure form as unprotected compounds, leading
to their easy isolation using e.g. gel filtration or other
chromatographic techniques. Glycosyltransferases,
which are natural biosynthetic enzymes, are very
selective, but they employ expensive and labile sugar
nucleosides and as typically intracellular proteins are
often rather unstable under in vitro conditions. This is
illustrated, e.g., in the synthesis by sinapate
glucosyltransferase from Gomphrena globasa (globe
amaranth).[12] This method relied on a relatively
expensive UDP-glucose and a not readily accessible
enzyme, and as a result, only small amounts of 1-O-
feruloyl- and 1-O-sinapoyl-β-glucopyranoses were
prepared, probably due to product inhibition. In
contrast to glycosyltransferases, extracellular
glycosidases are readily available robust enzymes,
operating under broad reaction conditions with much
cheaper glycosyl donors. Glycosylation of carboxylic
acids (e.g. benzoic acid) was described with sucrose
phosphorylase from Streptococcus mutans; here the
We present here an enzymatic glycosylation of a
series of hydroxyphenyl propenoic acids, and we
describe, besides phenolic OH glycosylation,
probably the first in vitro glycosidase-catalyzed
glycosylation at the carboxy moiety.
Results and Discussion
At the beginning we embarked on a project to test the
transglycosylation capacity of the rutinosidase from A.
niger with a variety of cinnamic acid derivatives and
their analogues. The recombinant rutinosidase was
prepared using
a
Pichia pastoris expression
system.[17] Rutin was used as a universal glycosyl
donor throughout all experiments. The primary
screening was performed under the same conditions
that proved to be effective for polyphenol
rutinosylation.[18]
(E)-p-Coumaric acid (3) and (E)-ferulic acid (4)
(Table 1) were tested as two pioneer substrates to
products
were
only
-glucopyranosides.[13]
Glycosylation of caffeic acid with recombinant
sucrose phosphorylase from Bifidobacterium longum
was demonstrated in supercritical CO2.[14] Authors
claimed formation of 1-O-caffeoyl--glucopyranose
based on MS studies only, no NMR data nor yields
examine
the
general
feasibility
of
the
transglycosylation concept. Both acceptors gave
transglycosylation products in a good yield as clearly
shown by TLC. After HPLC analysis and a detailed
1
inspection of the corresponding H NMR spectra, we
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900259
Advanced Synthesis & Catalysis
could observe that at least two glycosylation products
were formed. Two peaks in proportions of about 1 : 3
and 1 : 2 were observed in each HPLC chromatogram
for acceptors 3 and 4, respectively. In both reactions
the respective peaks had identical molecular masses
(determined by LCMS) and were found to correspond
to the respective O-monorutinosyl derivatives. This
result was very surprising as both acceptors have only
a phenolic group. Reactions such as C-glycosylation,
oligoglycosylation and de-rhamnosylation were
excluded by MS analysis. One remaining possibility,
hitherto considered to be rather improbable, was the
formation of glycosyl esters (Scheme 1), i.e.
anomeric proton H-1 of rhamnose. The position of
the rutinosyl moiety was unambiguously determined
using the HMBC correlations of the anomeric proton
of glucose (Fig. S1d, Supporting Information). As a
major product (about 79%) 4-O-rutinosyl (E)-p-
coumaric acid (3b) was identified, in which the
anomeric proton of glucose was correlated to the
para-positioned aromatic carbon of the aglycone. Ten
percent of the sample accounted for 1-O-(E)-p-
coumaroyl--rutinose (3a) whose structure was
unequivocally confirmed by the correlation of the
glucose-based anomeric proton with the C=O at the
carboxyl group (details of the HMBC spectrum are
glycosylation of the OH group at the carboxyl moiety. given in Fig. S1e in the Supporting Information).
As previously described, several glycosyl esters of
phenolic acids (but not rutinosyl esters) were already
isolated from natural sources.[4c,5,7] However,
glycosylation of a carboxy group with glycosidases
has never been described and it had always been
considered highly improbable.
However, the sample contained two additional
products, which were so far not observed even by
chromatographic techniques (LC-MS); later we found
that they were also inseparable using preparative
HPLC. These products were glycosylated products of
(Z)-p-coumaric acid, i.e. 4-O-rutinosyl (Z)-p-
coumaric acid (3c, 6 %) and 1-O-(Z)-p-coumaroyl--
rutinose (3d, 5 %), as determined by the respective
coupling constants of the isolated olefinic protons in
the aglycone (J = 15.9 ~ 16.0 Hz in (E)-p-coumaroyl;
J = 12.9 Hz in (Z)-p-coumaroyl). It should be noted
that all starting materials (cinnamic acid derivatives)
were always uniformly of the E-configuration.
We were thus facing another enigma: “How the
(Z)-isomers are formed during the reactions?”
Control experiments – incubation of the respective
(E)-acid under the same reaction conditions (without
addition of the enzyme, under the same mild
conditions) – confirmed that no (E)-(Z) isomerization
occurred. Derivatives of cinnamic acid, such as e.g.
coumaric or ferulic acid occur in nature both as (E)-
and (Z)-isomers; they both were isolated e.g. from
corn steep liquor[19]. Moreover, phenolic glycosides
containing both (E)- and (Z)-p-coumaric acid were
isolated from Salix glandulosa.[20] It is interesting to
note that some simple glycosylated derivatives of (Z)-
p-coumaric acid act as leaf-opening factors in
nyctinastic plants (e.g. Albizzia julibrissin or
Mimosa).[21] Isomerization between (E)- and (Z)-
isomers of cinnamic acid derivatives occurs in nature
typically under direct UV-A light. Mäkilä et al.
demonstrated that (E)-p-coumaric acid was converted
to the corresponding (Z)-isomer under UV-A
radiation (one hour) and direct sunlight (one week)
by 25 and 34%, respectively, when investigating the
stability of these acids in black currant juice.[22] At
preparatory scale, (Z)-cinnamates are readily
accessible from the respective (E)-isomers, which are
commercially available, by irradiation with a Pyrex-
filtered high-pressure mercury lamp (λem >300 nm) or
monochromatic light at 313 nm, resulting in the
formation of a photostationary mixture consisting of
about 67% (E)- and 33% (Z)-isomers.[23]
Scheme 1. Rutinosylation of p-coumaric acid (3) with
rutinosidase from Aspergillus niger (the products of (Z)-
configuration are not shown).
Glycosylation at the carboxy moiety yields esters,
which, as such, are apt to hydrolyze (saponify) under
alkaline conditions, while glycosides of alcohols
including phenols are quite stable under these
conditions. Conversely, it is well-known that all O-
glycosides hydrolyze under acidic conditions. As a
test of our hypothesis whether one of the products
was a glycosyl ester, we incubated the glycosylation
products of the enzymatic reactions with acceptors 3
and 4 under alkaline conditions (10% NaOH, r.t.).
Indeed, in both cases one of the products hydrolyzed,
yielding rutinose and the respective acid, whereas the
other product remained intact. Under acidic
conditions (1 N HCl, short boiling) all products
hydrolyzed to yield the respective acid acceptor
together with rhamnose and glucose, indicating
complete cleavage of all glycosidic linkages in the
molecules.
Advanced NMR methods allowed a detailed
analysis of the mixture of glycosylated products of p-
coumaric acid (3), pre-purified by extraction with
Amberlite XAD-4 resin, and the structural
determination of each component (Table S1,
Supporting Information). Anomeric configurations of
the rutinosyl moieties were determined from the
We speculated that for (E)-(Z) isomerization the
formation of a quinone intermediate might be
required (see Scheme 2). This requires a free phenolic
OH group in p- or possibly also o-position as well as
magnitude of JH1,
glycosidic linkage was proved by HMBC contact of
the downfield shifted carbon C-6 of glucose with the
coupling constants; the 16
H2
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900259
Advanced Synthesis & Catalysis
the presence of a conjugated 2,3-double bond and a
carboxyl function.
Scheme 3. Proposed mechanism of rutinosylation of
hydroxycinnamic acids at the carboxy moiety.
Table 1. Screening of acceptors for rutinosylation (isolated
yields).
Scheme 2. Glycosylation of (E)-p-coumaric acid (3).
To support this hypothesis and to investigate the
characteristics required for a particular phenolic acid
substrate to be glycosylated, a large panel of aromatic
acids bearing various structural features was tested
(Table 1). As previously shown, 3 and 4 yielded both
the (E) and (Z)-isomers of phenolic glycosides and
glycosyl esters while m-coumaric acid (5) unable to
form a quinone intermediate yielded both the
phenolic glycosides and glycosyl esters exclusively in
their original (E)-forms (5a, b). o-Coumaric (6) and
caffeic acid (7), both of which are able to form o- or
p-quinones, respectively, yielded only the (E)-
isomers of the glycosylated products.
Transglycosylation in retaining glycosidases
occurs by the mechanism depicted in Figure S45A
(Supporting Information), where the acceptor alcohol
approaches covalently bound glycosyl-enzyme
intermediate and forms a transition state stabilized by
the catalytic amino acid in the ionized state.[24] Acids
are ionized and thus negatively charged and therefore
they are repulsed electrostatically from the active site
carboxylic acid (Glu) acting as acidobasic catalyst
(see Figure S45B). This is probably the reason why
most glycosidases do not glycosylate also phenols
being weak acids. Rutinosidases seem to be rather
special in its ability to glycosylate phenols.[25,26]
It appears that for the glycosylation of the carboxy
group to occur, the complete structure of
hydroxycinnamic
acid
is
required.
p-
Hydroxyphenylacetic acid (8) lacking the conjugation
of the carboxy function with the aromatic system
yielded only the phenolic rutinoside 8a. This was
corroborated by the use of phenylacetic (9) and
phenylpropionic (10) acids as acceptors where no
glycosylation products were detected. Therefore, we
propose that the formation of glycosyl esters is
limited to the presence of a conjugated acid enriched
in nucleophilicity via mesomeric resonances or
inductive effects of a hydroxylated aromatic system
(Scheme 3).
Rut = 6-O-α-L-rhamnopyranosyl-β-D-glucopyranosyl-
(rutinosyl)
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900259
Advanced Synthesis & Catalysis
Moreover, the experiments performed with cinnamic
acid (11) and with its two derivatives p-
methoxycinnamic (12) and m-nitrocinnamic acid (13),
in which the -electronic density is enriched due to
the presence of the electron donating methoxy group
or reduced by the electron withdrawing nitro group,
respectively, corroborated our hypothesis since no
glycosyl esters were obtained.
Finally, sinapic (14) and quinic (15) acids, two
sterically hindered substrates, were tested. Acid
acceptor 14, which was thought to be a potential
substrate (in terms of our above-mentioned
hypothesis), was not glycosylated at all, probably due
to the steric hindrance of OH and/or due to hydrogen
bond formation with adjacent methoxy groups; acid
15 bearing an esterified caffeoyl moiety was not
substituted either.
Enzymatic Derhamnosylation of Rutinosylated
Aromatic Acids
To prove larger applicability of our enzymatic
procedure we have tested the previously developed
concept of a sequential two-enzyme synthesis of aryl
glucopyranosides.[18]
-L-Rhamnosidase
from
Aspergillus terreus[28] was shown to selectively “trim”
the rutinosyl unit to yield the respective β-
glucopyranoside. Here, we have used a preparation
containing both rutinosides 3a (rutinosyl ester) and
3b (phenolic rutinoside). After treatment by
rhamnosidase we could observe
a
virtually
quantitative conversion of rutinosides into respective
β-glucopyranosides by LCMS analysis (see Figs. S41
and S42, Supporting Information; Scheme 4).
Structures of the products were determined by UV
spectra analysis and MS (loss of rhamnose). No free
acid was observed in LCMS chromatogram, which
demonstrates a clean and selective reaction.
With regard to the above results we were interested
to see whether, due to their particular nature,
hydroxycinnamic acids can also yield phenolic or
carboxylic glycosides with other glycosidases. p-
Coumaric and ferulic acid, which gave the best
transglycosylation yields with rutinosidase, were
tested as substrates for glycosylation using a panel of
β-N-acetylhexosaminidases,
all
retaining
monoglycosidases. These enzymes were shown to
afford high transglycosylation yields with various
substrate types such as carbohydrates and/or aliphatic
alcohols.[27] We tested seven synthetically potent -N-
acetylhexosaminidases from Acremonium persicinum,
Aspergillus oryzae, Fusarium oxysporum, Penicillium
chrysogenum, P. oxalicum, Talaromyces flavus, and
Trichoderma harzianum. However, both aromatic
Scheme 4. Enzymatic derhamnosylation of the products of
enzymatic rutinosylation.
acids
strongly
inhibited
all
-N-
acetylhexosaminidases tested (the glycosylation rate
was reduced by more than 10 times) and despite the
addition of higher amounts of enzyme no
glycosylation products were observed using LCMS
analysis.
We have to note that Katayama et al. prepared 4-
O-rutinosides of vanillic, sinapic, ferulic, and caffeic
acids using crude rutinosidase from tartary
buckwheat (Fagopyrum tataricum) seeds, however
no glycosylation of the carboxy moiety was
described.[25] Although spectral characteristics of all
compounds were given, overall yields were not
provided. The resulting compounds were tested for
antiviral activity against feline calicivirus, and it was
found that rutinosylation, especially in the case of
sinapic acid, improved antiviral activity.
Finally, we were interested whether the products of
enzymatic glycosylation were substrates of the
rutinosidase. Previously we found that some
transglycosylation products, e.g. methyl rutinoside,
were virtually resistant to rutinosidase hydrolysis. We
have tested the hydrolytic cleavage of both
glycosylation products of p-coumaric acid 3a and 3b
using the rutinosidase under analogous conditions as
the transglycosylation reactions, however in the
absence of rutin, and indeed both rutinosides were
hydrolyzed to yield acid 3 as determined by LC-MS
analysis.
Conclusion
Many
retaining
glycosidases
exhibit
transglycosylation activities resulting in the facile
synthesis of various glycosides, which greatly
expands the use of these enzymes for synthetic
applications. Typically, the great potential of
enzymes in chemistry consists in their high selectivity,
simple reaction conditions and the possibility of
tuning their catalytic properties by selecting natural
or generating artificial enzyme variants. Although
mostly specific, enzymes may have promiscuous
activities, and thus enable the exploration of novel
reactions, possibly with surprising outcomes. Here
we describe for the first time the enzymatic synthesis
of glycosyl esters using a glycosidase as a catalyst.
We present a detailed structure-activity study of a
panel of aromatic acids as glycosyl acceptors and
identify the crucial structural features required for
this unique enzymatic activity as well as a plausible
reaction mechanism. At the same time, this work
highlights the rutinosidase from Aspergillus niger as
a remarkable catalyst with unparalleled synthetic
capabilities. Our study may serve as an initiation
platform for future research in this uncharted area.
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900259
Advanced Synthesis & Catalysis
follows: ESI interface voltage, 4.5 kV; detector voltage,
1.15 kV; nebulizing gas flow, 1.5 mL/min; drying gas 15
mL/min; heat block temperature 200 C; DL temperature
250 C; SCAN mode 150-680 m/z. The data were
processed by LabSolutions software ver. 5.75 SP2
(Shimadzu).
Experimental Section
Materials
Preparative HPLC
Media components were from Oxoid (UK) or Carl-ROTH
(DE). Rutin was from Alchimica (Prague, CZ) and other
chemicals of the highest purity available were purchased
from Sigma-Aldrich. All cinnamic acid derivatives used
were of (E)-configuration.
Preparative HPLC was performed with the same Shimadzu
Prominence system as described above; additionally,
fraction collector FRC-10A (Shimadzu, Kyoto, Japan) was
used. General procedure: the sample (10 mg) was
dissolved in mobile phase (200 µL), centrifuged and
injected (25 µL) in to the Chromolith SemiPrep (100  10
mm, Merck, DE) column equipped with Chromolith RP-
18e (5× 4.6 mm) guard cartridge. The fractions (1.2 mL)
were collected. Mobile phase: 10 % acetonitrile in water,
0.1 % formic acid, flow rate 1.2 mL/min, 25 °C, isocratic
elution.
Strains
An EasySelect Pichia pastoris KM71H Expression Kit was
obtained from Invitrogen (US). This expression system
employs methanol inducible AOX1 promoter, which is
repressed by, e.g., glucose. The recombinant protein is
produced as a fusion product into the growth medium in
response to the secretion signal of the α-mating factor from
S. cerevisiae included in the pPICZαA vector.
Cultivation Media
The culture of Aspergillus niger K2 CCIM is stored in the
Collection of Microorganisms of the Institute of
Microbiology of the Czech Academy of Sciences, Prague.
The fungal strains producing β-N-acetylhexosaminidases
(EC 3.2.1.52) originated from the Culture Collection of
Fungi (CCF), Department of Botany, Charles University,
Prague. The strains were cultivated and the enzymes were
prepared as described previously.[29]
P. pastoris cultivation on agar plates was performed in
YPD medium (Yeast Extract Peptone Dextrose Medium)
[g/L]: yeast extract OXOID, 10; bacteriological peptone
OXOID, 20; glucose LACHNER, CZ, 20.
The inoculum for P. pastoris cultivation was prepared in
BMGY medium (Buffered Glycerol-Complex Medium)
[g/L]: yeast extract, 10; peptone, 20; 100 mM potassium
phosphate, pH 6.0; YNB, 13.4; biotin 0.0004; glycerol, 10.
Buffered Methanol-Complex Medium (BMMY) has the
same composition as BMGY but instead of 1% (v/v)
glycerol methanol (0.5%, v/v) is added.
For large-scale productions, minimal media were used.
BMGH medium (Buffered Minimal Glycerol medium)
[g/L]: 100 mM potassium phosphate pH 6.0; YNB, 13.4;
biotin 0.0004; glycerol, 10) for overnight preculture.
BMMH medium (Buffered Minimal Methanol medium)
[g/L]: 100 mM potassium phosphate pH 6.0; YNB, 13.4;
biotin 0.0004; methanol, 5) was used for the main culture.
Fed-batch fermentations were carried out in BSM medium
(Basal Salt Medium) [g/L]: 85 % H3PO4, 26.7 mL; CaSO4 .
2 H2O, 1.17; K2SO4, 18.2; MgSO4 . 7 H2O, 14.9; KOH,
4.13; and glycerol, 40; and supplemented with 4.35 mL/L
of PTM1 (trace salts solution [g/L]: CuSO4 . 5 H2O, 6; NaI,
0.08; MnSO4 . H2O, 3; Na2MoO4 . 2 H2O, 0.2; H3BO3,
0.02; CoCl2, 0.5; ZnCl2, 20; FeSO4 . 7 H2O, 65; biotin, 0.2;
H2SO4 conc. 9.2 g). Methanol added in fed-batch
experiments was also supplemented with PTM1 (1.2 mL/L
pure methanol).
The production medium for A. niger cultivation of
consisted of [g/L]: rutin, 5.0; KH2PO4, 15.0; NH4Cl, 4.0;
KCl, 0.5; yeast extract, 5.0; casein hydrolysate, 1.0; and
1.0 mL of trace element Vishniac solution[30] at pH 5.0.
The pH of the medium was adjusted to 5.0. After
sterilization, each flask was supplemented with 1.0 mL of
sterile 10% MgSO4 .7 H2O (w/v).
Structure Determination
NMR spectra were recorded on a Bruker Avance III 600
1
MHz spectrometer (600.23 MHz for H, 150.93 MHz for
13C) and a Bruker Avance III 700 MHz spectrometer
1
(700.13 MHz for H, 176.05 MHz for 13C) in DMSO-d6 at
o
30 C using the manufacturer’s software (data and the
spectra in extenso in the Supporting Information).
Individual spin systems of acceptors and monosaccharide
units of rutinosyl were identified by COSY experiments,
and then transferred to carbons by HSQC. The HMBC
experiment enabled to assign quaternary carbons, connect
partial structures, and thus determine the glycosidic
linkage and the position of the aglycones.
Mass spectra in the negative ion mode were measured
using LTQ Orbitrap XL hybrid mass spectrometer
(Thermo Fisher Scientific, Waltham, MA, USA) equipped
with an electrospray ion source. The samples were
dissolved in methanol and introduced into the mobile
phase flow (methanol/water 4:1; 100 µL/min) using a 5 µL
loop. Spray voltage, capillary voltage, tube lens voltage
and capillary temperature were 4.6 kV, -25 V, -125 V and
275 °C, respectively.
Analytical HPLC and LCMS
The HPLC and LCMS analyses were accomplished using
the Shimadzu Prominence system consisting of a DGU-
20A3 mobile phase degasser, two LC-20AD solvent
delivery units, a SIL-20AC cooling auto sampler, a CTO-
10AS column oven, SPD-M20A diode array detector, and
LCMS-2020 mass detector with single quadrupole
equipped with an electrospray ion source (Shimadzu,
Kyoto, Japan). The sample (ca 0.5 mg) was dissolved in
mobile phase A (60 µL), filtered and analyzed on
Chromolith Performance RP-18e column (100 × 3 mm,
Merck, DE) equipped with Chromolith RP-18e (5× 4.6
mm) guard cartridge. Binary gradient elution was used:
mobile phase A = 5 % acetonitrile in water, 0.1 % formic
acid; mobile phase B = 80% acetonitrile in water, 0.1 %
formic acid. Gradient used: start 10 % B, 30 % B for 0–3
min; 30 % B for 3–8 min; 10 % B 8-9 min, end at 15 min;
flow rate 0.4 mL/min. temperature 25 °C, injection volume
1 µL; detection 200 nm (p-hydroxyphenylacetic acid), 330
nm (ferulic acid), 280 nm (m-coumaric acid), 285 nm (o-
coumaric acid, caffeic acid). The MS parameters were as
Heterologous Expression of Rutinosidase in Pichia
pastoris
The expression vector pPICZαA-RUT, obtained as
described previously, was linearized with restriction
endonuclease SacI and then the prepared competent P.
pastoris cells KM71H were transformed with the
linearized expression vector by electroporation according
to the manufacturer’s instructions (EasySelect Pichia
Expression Kit, Invitrogen).[17] The electroporated cells
were grown at various concentrations under the selection
pressure of zeocin (100 µg/mL) on YPD agar plates for 2
days at 28 °C.
Recombinant rutinosidase production was done according
to the manufacturer’s instructions: the colonies were
inoculated into 100 mL of BMGY medium, pH 6.0, and
incubated overnight with shaking at 28 °C. The cells were
then collected by centrifugation (5000 × g, 10 min, 20 °C)
and the pellet was resuspended in 30 mL of BMMY
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900259
Advanced Synthesis & Catalysis
medium in a 300 mL baffled flask. The production of
rutinosidase was induced by the addition of methanol
(0.5% v/v) every 24 h for 4 days. The flasks were
incubated at 28 °C and 220 rpm.
hydroxybenzaldehyde in acidic MeOH, 15 mg/mL) or by
charring with 5 % H2SO4 in EtOH.
This reaction setup was used (without reaction workup) for
screening of the rutinosylation acceptors.
For the preparatory reaction, 300 mg of acceptor acid were
used under the above conditions. When the reaction was
completed, the reaction mixture was diluted with ca 20 mL
of 50 mM citrate-phosphate buffer pH 4.5, centrifuged,
supernatant was collected and the solids were resuspended
in the same buffer volume and centrifuged again.
Combined supernatants (pH 4.5) were extracted with
AcOEt to remove the bulk of unreacted acceptor, rutin and
quercetin. The aqueous phase was partially evaporated in
vacuo to remove dissolved AcOEt. Non-ionic resin
Amberlite XAD-4 (Sigma) was washed overnight with
acetone, then washed extensively with water to remove all
traces of organic solvents and filled into the column. The
partially evaporated supernatant was loaded onto the
column and washed with 4 column volumes of water. Then
the resin was washed with MeOH to elute the
glycosylation products and the methanol solution was
evaporated to yield a crude mixture of products, which
were separated by preparative HPLC.
Large Scale Enzyme Production
The inoculum was prepared in BMGY medium. Fed-batch
fermentations were carried out in BSM medium. Methanol
added in fed-batch experiments was also supplemented
with PTM1 (1.2 mL/L pure methanol).
The inoculum for the fermentation cultivation was
prepared in 100 mL of BMGY medium. The fermentation
was performed as described previously in 3-L laboratory
fermenters (Brunswick BioFlo® 115, Eppendorf, DE).[31]
Then, 1.5 L of BSM media supplemented with 6.53 mL of
PTM1 was inoculated with inoculum (OD600 approx. 10-
12) to a concentration of 5 % v/v. The fermentation
conditions were as follows: 30 °C, pH 5 maintained with
ammonia solution (28-30 %), DO (dissolved oxygen) 20 %
maintained by agitation cascade from 50 to 1000 rpm,
aeration 0.66 vvm (volume of air per volume of medium
per minute) with the addition of 200 µL of Struktol J650
(Schill + Seilacher “Struktol” GmbH, Hamburg, DE) as an
antifoaming agent. After the complete consumption of
glycerol (approx. 20 h), two methanol doses of 3 g/L were
added in 20th and 31st h. After the utilization of the second
dose, the agitation was set at 600 rpm, the agitation
cascade was stopped and additional methanol (3 g/L) was
added. The methanol feed was connected to the actual
level of dissolved oxygen as described previously.[31]
Whenever the level of DO rose above 20 %, the methanol
feed was turned on by an automated program, and when
the DO level rose above 30 %, signaling an excess of
methanol and the inability of the culture to utilize it, the
pump was turned off again.
p-Coumaric Acid (3) Rutinosylation
(E)-p-Coumaric acid (3, 300 mg) was rutinosylated as
described in the general procedure. The reaction was
extracted by XAD-4 adsorption and the methanol eluate
was evaporated to yield 170 mg (20 %) of the glycosylated
products (mixture of glycosides). The above mixture of the
products was analyzed by HPLC (LCMS) (see Figure S2
Supporting Information) and it showed two peaks of
potential products with the same UV spectrum and m/z.
This mixture was directly subjected to NMR analysis,
which revealed the presence of (all possible) four
rutinosides (see Tables S1a-d and Figures S1a-e,
Supporting Information) in the following proportion
Purification of Recombinant Rutinosidase
1
(calculated from H integration of the meta-proton signals
1
of the aromatic moieties of respective aglycones in the H
Recombinant rutinosidase was purified from the culture
medium of Pichia pastoris after 6 days of cultivation with
methanol induction. The cells were harvested by
centrifugation (5000 × g, 10 min at 4 °C). The supernatant
was dialyzed against 6 L of 10 mM sodium acetate buffer,
pH 3.6, for 2 h (dialysis tubing cellulose membrane,
Sigma-Aldrich, cut-off 10 kDa). The pH of the solution
was then adjusted to 3.6 with 10 % acetic acid and filtered.
NMR spectrum of this mixture): 4-O-rutinosyl (E)-p-
coumaric acid (3b, 79 %), 1-O-(E)-p-coumaroyl--rutinose
(3a, 10 %), 4-O-rutinosyl (Z)-p-coumaric acid (3c, 6 %),
and 1-O-(Z)-p-coumaroyl--rutinose (3d, 5 %).
Preparative HPLC of the reaction mixture after XAD-4
workup (125.3 mg) yielded two fractions. 4-O-Rutinosyl
(E)-p-coumaric acid (3b, 38.7 mg, 30.9 %) white powder,
1
LCMS Fig. S3a-b, H and 13C NMR see Table S2, Fig.
-
S5a-b; MS (ESI−): ([M − H]−, m/z 471.1); HRMS (ESI−):
calcd for C21H27O12 471.15080, measured 471.14974
(−2.26 ppm), see Fig. S4.
This solution was loaded into a Fractogel EMD SO3
column (15 × 100 mm) in 10 mM sodium acetate buffer,
pH 3.6. The protein was eluted using a linear gradient of 0
– 1 M NaCl (5 mL/min). Fractions were collected and then
analyzed for rutinosidase activity using p-nitrophenol
rutinoside as a substrate. The fractions containing
rutinosidase activity were concentrated by ultrafiltration
using cellulose membranes with a 10 kDa cut-off
(Millipore, USA). The concentrated protein was then
purified to homogeneity by gel filtration in a Superdex 200
10/300 GL column (10 × 300 mm, 10 mM citrate-
phosphate buffer, pH 5.0, 150 mM NaCl).
1-O-(E)-p-Coumaroyl-β-rutinose (3a, 3.9 mg, 3.1 %) white
powder, LCMS (Fig. S6a-b). This preparation contained as
an inseparable minority ca 25 % of 1-O-(Z)-p-coumaroyl-
1
β-rutinose (3d) as determined by H and 13C NMR (see
Table S3a,b, Figs. 8a,b). MS (ESI−) spectrum: ([M − H]−,
m/z 471.1). HRMS (ESI−): calcd for C21H27O12 471.15080,
measured 471.14987 (-1.97 ppm), see Fig. S7.
Ferulic Acid (4) Rutinosylation
Protein concentrations were determined by Bradford assay
calibrated for BSA. The purity of recombinant rutinosidase
was checked by 12 % SDS-PAGE.
(E)-Ferulic acid (4, 300 mg) was rutinosylated as described
in the general procedure to yield 162 mg (21 %) of the
glycosylated products (mixture of glycosides). This
mixture (45 mg) was analyzed by LCMS to show two
peaks of products (Supporting Information, Fig. S9). They
was separated by preparative HPLC to yield 4-O-rutinosyl
(E)-ferulic acid (4b, 2.7 mg, 6 %), white powder, for
Enzymatic Glycosylation – General Procedure
The acid acceptor (1 equivalent) was dissolved in DMSO
(90 mg/mL), the resulting solution was added to the crude
rutinosidase medium (0.3 U/mL, 60 mL/g acceptor) and
pH was adjusted to 3.0. Rutin (0.5 eq) was added followed
by three additions (each 0.5 eq) of rutin every 90 minutes
(final concentration of DMSO was 15 %, v/v). The
resulting heterogeneous mixture was shaken (180 rpm,
Thermoshaker, Eppendorf, D) at 40 °C for a total of 6-7 h.
Reaction was monitored by TLC: the sample was diluted
5× with MeOH, analyzed on silica plates using
AcOEt/MeOH/HCO2H (4 : 1 : 0.05) as a mobile phase or
on reverse phase plates (MeCN/H2O, 4 : 6) and visualized
under UV light and by Komarowski reagent (solution of p-
1
HPLC see Fig. S10a,b, H and 13C NMR see Table S4 and
Fig. 12Sa,b; MS (ESI−): ([M − H]−, m/z 501.2); HRMS
(ESI−): calcd for C22H29O13 501.16136; found 501.16046 (-
1.80 ppm), Fig. S11 and 1-O-(E)-feruloyl-β-rutinose (4a,
1
3.7 mg, 8.2 %), white powder, HPLC see Fig. S13a,b, H
and 13C NMR see Table S5a and Figures S15a,b; MS
(ESI−): ([M − H]−, m/z 501.2); HRMS (ESI−) calcd. for
C22H29O13 501.1613; found 501.16136, see Fig. S14. The
latter preparation contained ca 25% of inseparable minority
1
1-O-(Z)-feruloyl-β-rutinose (4c) as determined by H and
13C NMR (see Table S5b).
7
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900259
Advanced Synthesis & Catalysis
m-Coumaric Acid (5) Rutinosylation
ferulic acid (4) and left overnight at room temperature
under stirring. Then the resulting mixture was analyzed by
HPLC – see Fig. S40. Mixture of glycosides contains two
product peaks (Fig. S40A as determined by UV scan,
LCMS and by co-chromatography with isolated products
4a and 4b). After alkaline hydrolysis (Fig. S40B) product
4a disappeared, while the phenolic glycoside 4b remained
intact and free acid 4 appeared. Liberated ferulic acid (4) in
the alkaline milieu probably partially underwent oxidative
oligomerization forming an unidentified peak (Fig. S40B).
(E)-m-Coumaric acid (5, 300 mg) was rutinosylated as
described in the general procedure to yield 184 mg (21 %)
of the glycosylated products (mixture of glycosides). This
mixture (125.3 mg) was analyzed by LCMS to show two
peaks of products (Fig. S16). They were separated with
preparative HPLC to yield 4-O-rutinosyl (E)-m-coumaric
acid (5b, 76.1 mg, 46 %) white powder, HPLC Fig. S17a,b,
1H and 13C NMR see Table S6 and Figs. S19a,b; MS
(ESI−): ([M − H]−, m/z 471.1); HRMS (ESI−): calcd for
C21H27O12 471.15080, measured 471.14951 (-2.74 ppm),
see Fig. S21 and 1-O-(E)-m-coumaroyl-β-rutinose (5a,
Enzymatic Derhamnosylation of Rutinosides
1
14.2 mg, 8.6 %), white powder, HPLC Fig. S20a,b, H and
Mixture (20 mg) of 4-O-rutinosyl (E)-p-coumaric acid (3b,
79 %), 1-O-(E)-p-coumaroyl-β-rutinose (3a, 10 %) was
incubated with recombinant α-L-rhamnosidase from
Aspergillus terreus (2.19 U/mL, 30.1 mg/mL)[26] in 4 mL
of 50 mM citrate-phosphate buffer, pH 5, for 4 hours at
37 °C. Then, the reaction was terminated by enzyme
denaturation (90 °C, 5 min) and the reaction mixture was
analyzed by HPLC and LCMS (Figures S41 and S42,
Supporting Information); a sample of the compound 3e
was isolated by preparatory HPLC to obtain clean NMR
spectra (Table S13 and Figure S44, Supporting
Information), compound 3f was not isolated due to its
paucity.
13C NMR see Table S7 and Figs. S22a,b; MS (ESI−): ([M −
H]−, m/z 471.1); HRMS (ESI−): calcd for C21H27O12
471.15080, measured 471.14993 (-1.84 ppm), see Fig. S21.
No (Z)-side product was observed in the NMR spectra.
o-Coumaric Acid (6) Rutinosylation
(E)-o-Coumaric acid (6, 300 mg) was rutinosylated as
described in the general procedure to yield 4-O-rutinosyl
(E)-o-coumaric acid (6b, 138 mg, 16 %) yellowish
1
powder; HPLC see Fig. S23a,b, H and 13C NMR see
Table S8a and Fig. S25a,b; MS (ESI−): ([M − H]−, m/z
471.1); HRMS (ESI−): calcd for C21H27O12 471.15080,
measured 471.14978 (-2.16 ppm), see Fig. S24. This
product contained 7 % of inseparable minority 1-O-(E)-o-
coumaroyl-β-rutinose (6a) as inferred from NMR spectra;
1H and 13C NMR see Table S8b. No (Z)-configured
product was observed in the NMR spectra.
Glycosylation of p-Coumaric and Ferulic Acid with
Hexosaminidases
Reaction mixtures comprised of (final. conc.) p-coumaric
or ferulic acid (300 mM), 30 mM p-nitrophenyl β-N-
acetylglucosaminide (30 mM), McIlvaine buffer pH 5 /
Caffeic Acid (7) Rutinosylation
40
% (v/v) acetonitrile and the respective β-N-
acetylhexosaminidase (1.5 U/mL). Fungal extracellular β-
N-acetylhexosaminidases with a good transglycosylation
potential[32] from the following sources were tested:
Acremonium persicinum CCF 1850, Aspergillus oryzae
CCF 1066, Fusarium oxysporum CCF 371, Penicillium
chrysogenum CCF 1269, P. oxalicum CCF 2315,
Talaromyces flavus CCF 2686, Trichoderma harzianum
CCF 2687. Reactions were incubated at 35 C and 850 rpm
for 24 hours and monitored by TLC (propan-2-ol : H2O :
NH4OHaq., 7:2:1) and LCMS. A control reaction containing
only 30 mM p-nitrophenyl β-N-acetylglucosaminide (30
mM) donor in McIlvaine buffer pH 5 / 40 % (v/v)
acetonitrile and the enzyme (0.15 U/mL) was run under the
same conditions to verify the enzyme activity in the
absence of acid acceptor.
(E)-Caffeic acid (7, 300 mg) was rutinosylated as
described in the general procedure. Reaction was worked
up with XAD-4 adsorption and methanolic eluate was
evaporated to yield 224 mg (28 %) of the glycosylated
product, which after LCMS analysis showed three peaks
having the UV spectrum and m/z of expected rutinosylated
product (for HPLC see Fig. S16). This reaction mixture
(109.3 mg) was separated with preparatory HPLC to yield,
4-O-rutinosyl (E)-caffeic acid (7b, 8.3 mg, 7.6 %) as a
white powder; HPLC see Fig. S27a,b, 1H and 13C NMR see
Table S9 and Figs. 29a,b; MS (ESI−): ([M − H]−, m/z
487.1); HRMS (ESI−): calcd for C21H27O13 487.14571,
measured 487.14478 (-1.93 ppm), see Fig. S28; 3-O-
rutinosyl (E)-caffeic acid (7c, 29.9 mg, 27.3 %) as a white
powder, HPLC see Fig. S30a,b, 1H and 13C NMR see Table
S10 and Figs. S32a,b; MS (ESI−): ([M − H]−, m/z 487.1);
HRMS (ESI-): calcd for C21H27O13 487.14571, measured
487.14442 (-2.66 ppm), see Fig. S31; 1-O-(E)-caffeoyl-β-
rutinose (7a, 4.5 mg, 4.1 %) as a white powder, for LCMS
Acknowledgements
1
see Fig. S33a,b, for H and 13C NMR see Table S11 and
Authors acknowledge support from Czech Science Foundation
project 18-00150S (V.K.) and 19-00091S (M.K.), the joint Czech-
Italian AVČR-CNR mobility project No. CNR-16-30 (V.K. & S.R.)
and the EDRF project from Slovak Research Agency grant
number ITMS 26230120009 (M.R).
Figs. S35a,b, MS ESI−: ([M − H]−, m/z 487.1); HRMS
ESI−: calcd for C21H27O13 487.14571, measured 487.14511
(-1.24 ppm), see Fig. S34.
p-Hydroxyphenylacetic Acid (8) Rutinosylation
p-Hydroxyphenylacetic acid (8, 300 mg) was rutinosylated
as described in the general procedure to yield 145 mg
(16 %) of the glycosylated product, which after LCMS
analysis showed one peak of rutinosylated product (for
HPLC see Fig. S36). This product (52.9 mg) was was
further purified by preparative HPLC to yield 4-O-
rutinosyl phenylacetic acid (8a, 6.6 mg, 12.4 %; r.t. 2.660
min) as a white powder, LCMS see Fig. S37a,b, 1H and 13C
NMR see Table S12 and Figs. S39a,b; MS ESI(−): ([M −
H]−, m/z 459.1;); HRMS ESI(−): calcd for C20H27O12
459.15080, measured 459.14993 (-1.89 ppm), see Fig. S38.
References
[1] N. C. Veitch, I. Regos, G. C. Kite, D. Treutter,
Phytochemistry 2011, 72, 423-429.
[2] E. Heřmánková-Vavříková, A. Křenková, L.
Petrásková, C. S. Chambers, J. Zápal, M. Kuzma, K.
Valentová, V. Křen, Int. J. Mol. Sci. 2017, 18, 1074.
[3] a) J. B. Harborne, J. J. Corner, Biochem. J. 1961, 81,
242-250; b) S. Asen, S. L. Emsweller,
Phytochemistry 1962, 1, 169-174.
[4] a) B. Schuster, K. Herrmann, Phytochemistry 1985,
24, 2761-2764; b) D. Strack, N. Marxen, H. Reznik,
H.-D. Ihlenfeld, Phytochemistry 1990, 29, 2175-
Alkaline Hydrolysis
Aqueous solution of NaOH (10 % w/v, 3 mL) was added
to the mixture of dry glycosylated products (32 mg) of
8
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900259
Advanced Synthesis & Catalysis
2178; c) C. Milkowsky, D. Strack, Phytochemistry
2004, 65, 517-524.
[19] T. Niwa, U. Doi, Y. Kato, T. Osawa, J. Agric. Food
Chem. 2001, 49, 177-182
[5] S. H. Han, W. S. Suh, K. J. Park, K. H. Kim, K. R.
Lee, Arch. Pharm. Res. 2015, 38, 1312–1316.
[6] H. M. Zhou, Y. L. Liu, Acta Pharm. Sinica 1990, 25,
123-126.
[7] M. Neacsu, V. Micol, L. Pérez-Fons, S. Willför, B.
Holmbom, Molecules 2007, 12, 205-217.
[20] C. S. Kim, O. W. Kwon, S. Y. Kim, S. U. Choi, J. Y.
Kim, J. Y. Han, S. I. Choi, J. G. Choi, K. H. Kim, K.
R. Lee, J. Nat. Prod. 2014, 77, 1955-1961.
[21] M. Ueda, Y. Sawai, Y. Shibazaki, C. Tashiro, T.
Ohnuki, S. Yamamura, Biosci. Biotechnol. Biochem.
1998, 62, 2133-2137.
[8] D. E. Goggin, G. L. Nealon, G. R. Cawthray, A.
Scaffidi, M. J. Howard, S. B. Powles, G. R. Flematti,
J. Agric. Food Chem. 2019, 66, 13378-13385.
[9] T. Kosuge, E. E. Conn, J. Biol. Chem. 1961, 236,
1617-1621.
[22] L. Mäkilä, O. Laaksonen, A. L. Alanne, M.
Kortesniemi, H. Kallio, B. Yang, J. Agric. Food
Chem. 2016, 64, 4584-4598.
[23] M. L. Salum, C. J. Robles, R. Erra-Balsells, Org. Lett.
2010, 12, 4808-4811.
[10] Y. Zhu, J. Ralph, Tetrahedron Lett. 2011, 52, 3729–
3731.
[24] P. Bojarová, V. Křen, Trends Biotechnol. 2009, 27,
199-209.
[11] T. Desmet, W. Soetaert, P. Bojarová, V. Křen, L.
Dijkhuizen, V. Eastwick-Field, A. Schiller, Chem.
Eur. J. 2012, 18, 10786-10801.
[25] S. Katayama, F. Ohno, Y. Yamauchi, M. Kato, H.
Makabe, S. Nakamura, J. Agric. Food Chem. 2013,
61, 9617-9622.
[12] Y. Matsuba, Y. Okuda, Y. Abe, Y. Kitamura, K.
Terasaka, H. Mizukami, H. Kamakura, N. Kawahara,
Y. Goda, N. Sasaki, Y. Ozeki, Plant Biotechnol. 2008,
25, 369–375.
[13] K. Sugimoto, K. Nomura, H. Nishiura, K. Ohdan, T.
Nishimura, H. Hayashi, T. Kuriki, J. Biosci. Bioeng.
2007, 104, 22–29.
[14] M. H. Shin, N.Y. Cheong, J. H. Lee, K. H. Kim,
Food Chem. 2009, 115, 1028–1033.
[15] V. Křen, L. Martínková, Curr. Med. Chem. 2001, 8,
1313–1338.
[16] a) C. Goedl, T. Sawangwan, P. Wildberger, B.
Nidetzky, Biocat. Biotrans. 2010, 28, 10-21; b) K. De
Winter, L. Van Renterghem, K. Wuyts, H. Pelantová,
V. Křen, W. Soetaert, T. Desmet, J. Agric. Food
Chem. 2015, 63, 10131–10139.
[17] D. Šimčíková, M. Kotik, L. Weignerová, P. Halada,
H. Pelantová, K. Adamcová, V. Křen, Adv. Synth.
Catal. 2015, 357, 107-117.
[26] L. S. Mazzaferro, G. Weiz, L. Braun, M. Kotik, H.
Pelantová, V. Křen, J. D. Breccia, Biotechnol. Appl.
Biochem. 2019, 66, 53-58.
[27] K. Slámová, P. Bojarová, L. Petrásková, V. Křen,
Biotech. Adv. 2010, 28, 682-693.
[28] D. Gerstorferová, B. Fliedrová, P. Halada, P. Marhol,
V. Křen, L. Weignerová, Process Biochem. 2012, 47,
828-835.
[29] P. Fialová, L. Weignerová, J. Rauvolfová, V.
Přikrylová, A. Pišvejcová, R. Ettrich, M. Kuzma, P.
Sedmera, V. Křen, Tetrahedron 2004, 60, 693-701.
[30] W. Vishniac, M. Santer, Bacteriol. Rev. 1957, 21,
195–213.
[31] K. Markošová, L. Weignerová, M. Rosenberg, V.
Křen, M. Rebroš, Front. Microbiol. 2015, 6, 1140.
[32] a) P. Bojarová, N. Kulik, M. Hovorková, K. Slámová,
H. Pelantová, V. Křen, Molecules 2019, 24, 599; b) K.
Slámová, R. Gažák, P. Bojarová, N. Kulik, R. Ettrich,
H. Pelantová, P. Sedmera, V. Křen, Glycobiology
2010, 20, 1002-1009.
[18] I. Bassanini, J. Krejzová, W. Panzeri, V. Křen, D.
Monti, S. Riva, ChemSusChem 2017, 10, 2040-2045.
9
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900259
Advanced Synthesis & Catalysis
FULL PAPER
Glycosidase-Catalyzed Synthesis of Glycosyl
Esters and Phenolic Glycosides of Aromatic Acids
Adv. Synth. Catal. Year, Volume, Page – Page
Ivan Bassanini,b,f Jana Kapešová,a Lucie
Petrásková,a Helena Pelantová,a Kristína
Markošová,c Martin Rebroš,c Kateřina Valentová,a
Michael Kotik,a Kristýna Káňová,a Pavla
Bojarová,a Josef Cvačka,e Lucie Turková,a Erica
Ferrandi,b Ikram Bayout,b,d Sergio Riva,b Vladimír
Křena,*
10
This article is protected by copyright. All rights reserved.