Enzymatic sulfation of phenolic hydroxy groups of various plant metabolites by an arylsulfotransferase

Article abstract of DOI:10.1002/ejoc.201402875

The bacterial arylsulfate sulfotransferase (AST) from Desulfitobacterium hafniense was used as a catalytic tool to derivatize poorly soluble aromatic compounds (polyphenols). As examples, we sulfated the natural occurring compounds p-coumaric acid, 6-hydr

Full text of DOI:10.1002/ejoc.201402875

FULL PAPER
DOI: 10.1002/ejoc.201402875
Enzymatic Sulfation of Phenolic Hydroxy Groups of Various Plant
Metabolites by an Arylsulfotransferase
Michael A. van der Horst,[a][‡] Aloysius F. Hartog,[a][‡] Rabab El Morabet,[a]
Arthur Marais,[a] Menzo Kircz,[a] and Ron Wever*[a]
Keywords: Transferases / Enzyme catalysis / Biocatalysis / Synthetic methods / Phenols / Sulfation
The bacterial arylsulfate sulfotransferase (AST) from De-
sulfitobacterium hafniense was used as a catalytic tool to de-
rivatize poorly soluble aromatic compounds (polyphenols).
As examples, we sulfated the natural occurring compounds
p-coumaric acid, 6-hydroxyflavone, resveratrol, phloretin,
The sulfation of resveratrol resulted in two different mono-
sulfates (4Ј- and 3-sulfates), the 3,4Ј-disulfate, and the 3,5,4Ј-
trisulfate. Sulfation of phloretin resulted in a monosulfate (4Ј-
sulfate) and a disulfate (4,4Ј-disulfate). Although quercetin
has five hydroxy groups that could be sulfated, surprisingly
and quercetin, using p-nitrophenylsulfate as the sulfate do- this enzyme system primarily catalyses the sulfation only at
nor. The water-soluble sulfate esters were purified and char-
acterized. Depending on the nature of the compound, one or
more sulfate groups could be introduced in a stepwise order.
the 4Ј position. This simple enzymatic one-step sulfation
method is easy to use, and it allows a convenient and simple
production of sulfated compounds with improved solubility.
to be introduced. Deprotection methods have also been de-
veloped, e.g., by Ingram and Taylor.[4] However, in general,
both protection and deprotection require several steps,
which results in considerable amounts of waste.[5]
Introduction
Sulfated compounds play an important role in many
physiological processes, e.g., development, differentiation,
and regulation, and therefore they are of pharmacological
interest. In this paper, the sulfation of various plant phen-
olic compounds with a bacterial arylsulfotransferase (AST)
is explored. Many of these biologically active polyphenolic
compounds are present in fruits and vegetables, or are used
as food additives. Studies have shown that some of them
(may) work therapeutically against cancer, diabetes, and
various metabolic diseases.[1] The effects of these phenolic
compounds and their derivatives on human health have
been widely explored. In the human body, these compounds
often get sulfated, and the sulfated derivatives are needed
as reference compounds and standards. In addition, admin-
istration of these plant phenolic compounds is hampered
by their low solubility, and the use of sulfated derivatives
may circumvent this problem. However, in some cases, the
biological activity of the compound may change (either
positively or negatively) upon sulfation.[2,3]
It is also possible to use eukaryotic (aryl)sulfotransfer-
ases to sulfate compounds. These enzymes are substrate
specific, and use 3Ј-phosphoadenosine-5Ј-phosphosulfate
(PAPS) as the sulfate donor (for a review, see ref.[6]). Efforts
have been made to use enzymes of this type as catalytic
tools in biotransformations. However, the synthetic use of
these enzymes is hindered by the high cost of PAPS, inhibi-
tion by the PAP produced, and substrate specificity.[7,8]
Therefore, regeneration systems containing a second or
more enzymes have been developed to regenerate PAPS.[8,9]
An enzyme that uses a simpler and cheaper sulfate donor
and is able to sulfate a wide range of aromatic hydroxy com-
pounds would be an attractive alternative to both enzymatic
synthesis using PAPS and chemical synthesis of sulfate es-
ters. A bacterial arylsulfotransferase from Desulfitobacter-
ium hafniense that uses the simple sulfate donor para-
nitrophenyl sulfate (p-NPS) was expressed by us, and was
used as a catalytic tool for the sulfation of various steroids
and some aliphatic alcohols.[10] We previously showed that
this enzyme regiospecifically sulfates silybin and iso-
silybin,[11] and as another example, we sulfated zealorine at
the 14-position, resulting in a so-called masked myco-
toxin.[12] Recently, the sulfation of simple phenolic com-
pounds, and of nucleoside triphosphates as substrates for
PAPS regeneration, by an enzyme from the bacterium Hali-
angium ochraceum was studied.[13]
To synthesize sulfate esters of aromatic compounds,
chemical methods involving chlorosulfonic acid or triethyl-
amine–sulfur-trioxide complex have been developed. When
regiospecific modification is desired, protecting groups have
[a] Van’t Hoff Institute for Molecular Sciences, Faculty of Sciences,
University of Amsterdam,
Science Park 904, 1098 XH Amsterdam, The Netherlands
E-mail: r.wever@uva.nl
http://hims.uva.nl
[‡] These two authors contributed equally to this publication
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201402875.
In contrast to the mechanism of the reactions catalysed
by PAPS-dependent sulfotransferases, structural and mech-
534
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 534–541

Enzymatic Sulfation of Phenolic Hydroxy Groups
anistic understanding of the reactions catalysed by bacterial and grape juice, and it is thought to have beneficial health
arylsulfotransferases is quite limited. Recently, crystal struc- and anti-aging effects (for a review, see ref.[21]). In the hu-
tures of an AST from Escherichia coli CFT073 have become man body, resveratrol is rapidly metabolized, mainly by glu-
available.[14] These, and studies of an AST from Eubacte- curonidation and sulfation.[3] Because the concentrations of
rium A-44,[15] demonstrate a ping-pong bi-bi mechanism in the resulting metabolites are much higher than that of
which a conserved histidine residue forms a transient cova- resveratrol itself, it has been speculated that the biological
lent intermediate. However, even within the group of bacte- activity actually comes from these metabolites. The mole-
rial arylsulfotransferases, there may very well be subgroups cule has three potential sulfation sites. In humans, resvera-
with different reaction mechanisms and substrate-binding- trol 3-sulfate, resveratrol 4Ј-sulfate, and the disulfate could
pocket architectures.[10,16] Thus, it is not trivial to predict all be identified in metabolite analysis.[22] Phloretin is pres-
which types of substrates would be accepted by the AST’s ent in a high concentration in apples and apple-derived
as sulfate-acceptor molecules.
products, and it functions as an antioxidant.[23] The mol-
In this study, we have investigated the enzymatic sulf- ecule contains four hydroxy groups that could be sulfated.
ation of a number of phenolic compounds (Figure 1). The In humans, the main metabolites that are detected are phlo-
monohydroxyflavones (MHF’s) and p-coumaric acid (p- retin glucuronides, but also sulfates can be found. However,
CA) are structurally the simplest compounds in this study, in these cases, the site of sulfation remains unclear.[24] Quer-
having only a single phenolic hydroxy group that could be cetin is also ubiquitous in nature, and for example, it is pres-
sulfated. 6-Hydroxyflavone is a potential drug candidate to ent in high concentrations in capers. It has potentially bene-
treat anxiety-like disorders.[17] p-Coumaric acid is an anti- ficial effects on cardiovascular diseases.[25] The molecule has
oxidant that is found in a wide variety of edible plants.[18] five hydroxy groups that could potentially be sulfated. The
Furthermore, it is found as cofactor in the bacterial sensor major quercetin metabolite found in humans is quercetin-
protein photoactive yellow protein.[19]
3Ј-sulfate.[26] Interestingly, combinations of resveratrol and
quercetin have been shown to enhance the inhibition of ad-
ipogenesis.[27]
In this paper, we report the synthesis of all of these phen-
olic sulfated compounds on a 67 nmol to 1.3 mmol scale
using the bacterial arylsulfotransferase. By varying the con-
centration of the sulfate donor, p-NPS, with respect to the
acceptor molecules with more than one aromatic hydroxy
group, it was possible to study the regioselectivity and the
order in which the various OH groups were sulfated.
Results and Discussion
p-Coumaric Acid
The phenolic hydroxy group of p-coumaric acid was sulf-
ated by the AST when 1 equiv. p-NPS as sulfate donor was
added. When the reaction was followed by HPLC, a de-
crease in the concentrations of p-CA and p-NPS was ob-
served. At the same time, peaks corresponding to p-NP
(para-nitrophenol) and the presumed sulfated p-CA (zos-
teric acid) appeared (Figure 2; see Table S1 for an overview
of the reaction products and the corresponding HPLC re-
tention times). To synthesize and characterize the sulfated
compound, the reaction was carried out on a larger scale,
using 40 mm of the reactants in an incubation volume of
25 mL. According to the HPLC data, a conversion of 95%
was obtained. The product was purified by extraction with
ethyl acetate and silica gel chromatography (isolated yield
85%, for details see Exp. Sect. and Table 1). It was analysed
by 1H and 13C NMR spectroscopy, and it was proved to be
zosteric acid (see Supporting Information for details of this
Figure 1. Structures of the compounds used in this study.
Zosteric acid, sulfated p-coumaric acid, is found in sea and other compounds used and synthesized in this study).
grass, and it has been suggested that it plays a role as an Interestingly, the structurally comparable molecule 4-hy-
antibiofouling agent.[20] Resveratrol (3,5,4Ј-trihydroxystil- droxybenzoic acid is not accepted by the AST as a sulfate
bene) is widely distributed in nature, found in e.g., red wine acceptor.[10] Apparently, the enzyme can sulfate negatively
Eur. J. Org. Chem. 2015, 534–541
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
535

R. Wever et al.
FULL PAPER
charged substrate molecules, provided there is sufficient dis- hydroxyflavone, and 7-hydroxyflavone as sulfate acceptors
tance between the negative charge and the phenolic hydroxy was investigated. Because of the low water solubility of the
group.
hydroxyflavones, DMSO was used as a cosolvent (12.5%
v/v). It has previously been reported that the enzyme is
stable in the presence of up to 25% DMSO.[10] Upon incu-
bation of the three different hydroxyflavone molecules with
p-NPS and the AST, and analysis by HPLC, a new species
was formed with a concomitant decrease of p-NPS and the
appearance of p-NP. No difference was observed between
the sulfation rates of the different hydroxyflavones (not
1
shown). For analysis by H and 13C NMR spectroscopy, 6-
hydroxyflavone was sulfated on a larger scale, using 1 mm
of the acceptor molecule in a 150 mL reaction volume, and
using acetone as a cosolvent (yield 69 nmol). The NMR
spectroscopic data (see Supporting Information) show that,
as expected, the sulfation did indeed occur at the 6-position.
The sulfation of 5-hydroxyflavone by Streptomyces fulvis-
simus[28] has been reported, and human liver and duodenum
extracts have been shown to sulfate 7-hydroxyflavone, but
not 5-hydroxyflavone or 3-hydroxyflavone.[29] The in vitro
sulfation of monohydroxyflavones by liver fractions from
mouse, rat, dog, and human has been described.[30] In this
case, both 7-, 6-, and 4Ј-hydroxyflavones were sulfated, but
large differences were observed between the sulfation be-
haviour of the liver extracts from different species and gen-
ders. We conclude that the bacterial sulfotransferase from
Desulfitobacterium hafniense has a larger substrate toler-
ance with respect to monohydroxyflavones than do the eu-
karyotic sulfotransferases. This makes the enzyme suitable
for biocatalytic sulfation of this type of molecule.
Resveratrol
To study the regioselectivity of the AST in the sulfation
of polyphenolic compounds, the sulfation of E-resveratrol
(from here on referred to as resveratrol) was tested, since
this molecule has three hydroxy groups that could be sulf-
ated (note that the 3- and 5-positions are equivalent).
The reaction was carried out under the same conditions
that were used for the monohydroxyflavones, using acetone
(15%) to solubilize the resveratrol. HPLC analysis showed
that depending on the amount of sulfate donor used (1–
3 mol p-NPS/mol resveratrol) and the reaction time, four
different products could be observed in this reaction; four
of the five possible 3- and 4Ј-mono-, di-, and trisulfated
products. Figure 3, panel A shows the HPLC trace of an
incubation of resveratrol (t = 20.7 min) in the presence of
2 equiv. of p-NPS (t = 24.6 min) before the addition of the
AST. Figure 3, panel B shows the HPLC analysis after 20 h
incubation of the reaction mixture. As can be seen, after
this time, the resveratrol had been almost completely con-
sumed, and new peaks corresponding to mainly mono- and
smaller amounts of di- and trisulfated resvaratrol as well as
p-NP had appeared. Separate experiments were carried out
in which resveratrol and p-NPS were incubated in a 1:1
Figure 2. Sulfation of p-CA, and purification of zosteric acid. The
conversion of p-CA to zosteric acid and of p-NPS to p-NP was
followed by HPLC. The reaction was carried out as described in the
Experimental Section, using 40 mm of reactants and 0.25 UmL–1 of
the AST. Samples were taken at the start of the reaction (A), after
7 d (B), and after purification of the sulfated compound zosteric
acid (C). The following peaks were identified: 24.6 min, p-NPS;
12.5 min, p-CA; 22.4 min, p-NP; 21.2 min, zosteric acid.
Monohydroxyflavones
Hydroxyflavone was structurally the simplest flavone ratio, a 1:2 ratio, and a 1:3 ratio. After incubation, the four
used in this study, and the sulfation of 5-hydroxyflavone, 6- sulfated resveratrol derivatives were purified as described in
536
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 534–541

Enzymatic Sulfation of Phenolic Hydroxy Groups
Table 1. Reaction conditions and yields of the preparative trans-sulfations.
Acceptor (concentration
[mm])
Concentra-
tion of p-
NPS [mM]
Concentra-
tion of AST
[UmL–1]
Total vol-
ume [mL]
Time Product
Yield
[mg]
Yield
[mol-%][a]
p-Coumaric acid (20)
23
0.25
25
7 d
4-sulfocinnamic acid
(zosteric acid)
97
85
6-Hydroxyflavone[b] (1)
5
5.4
0.1
0.25
150
20
2 h
1 d
hydroxyflavone 6-sulfate
3-resveratrol monosulfate
4Ј-resveratrol monosulfate
3,4Ј-resveratrol disulfate
3,4Ј,5-resveratrol trisulfate
4-phloretin monosulfate
4,4Ј-phloretin disulfate
4-quercetin monosulfate
2-phenylphenol sulfate
2,2Ј-biphenol monosulfate
22
13
3
59
41
24
29
38
109
41
46
31
7
51
29
21
19
20
87
46
E-Resveratrol[b] (6.5)
E-Resveratrol (10)
32
2.5
0.2
30
8 d
Phloretin (16)
16
34
5.5
22
10
20
20
100
25
4 d
6 d
1 d
7 d
7 d
Quercetin (5)
2-Phenylphenol (20)
2,2Ј-Bisphenol (10)
0.1
0.2
0.2
25
[a] Yield is the yield after isolation. [b] Acetone (15%) was used as cosolvent. After the reaction was complete, the solution was flushed
with nitrogen to remove the acetone.
the Supporting Information. This allowed further charac- NPS were present, a single product was formed after a 27 h
terization by 1H and 13C NMR spectroscopy. Figure 3, pan- reaction (Figure 5, A–C). Purification (67 nmol, 21% iso-
els C, D, and E show the HPLC traces of the purified com- lated yield) and subsequent NMR spectroscopic analysis
pounds. Figure 4 summarizes the sequence of events. These (see Supporting Information) revealed this product to be
results show that the reactivities of the 3-OH and 4Ј-OH phloretin monosulfate, sulfated at the 4-position.
groups of resveratrol are more or less equivalent in the en-
A reaction was carried out with phloretin and p-NPS in
zyme-catalysed sulfation reaction. To obtain di- and tri- a 1:2.2 ratio, and the reaction was stopped after 117 h. Fig-
sulfated resvaratrol, more p-NPS is required, and by ure 5 (panels D–F) show that in addition to the peak corre-
choosing the concentration of the sulfate donor it was pos- sponding to the monosulfate, a second peak was observed.
sible to selectively obtain the various sulfated compounds The monosulfate and presumed disulfate could be sepa-
(Table 1).
rated on a silica column, as described in the Experimental
In animals, most resveratrol compounds detected are Section. Purification of the presumed disulfate yielded a
either glucuronide or sulfate conjugates. Wenzel et al.[31] single product with an isolated yield of 19%. NMR spectro-
showed that all five possible resveratrol metabolites could scopic analysis (see Supporting Information) showed that
be detected in rats, with only minimal traces of unchanged this product was indeed a phloretin disulfate, with sulfate
resveratrol. It has been thought that the biological response groups on both the 4 and 4Ј positions. These results show
in humans is not (only) elicited by nonsulfated resveratrol, that the reactivity of the 4-OH group is higher than that of
but also by the resveratrol metabolites, such as the sulfated the 4Ј-OH group. The OH groups in the ortho positions are
derivatives.[32] Using various sulfated resveratrol com- not sulfated, probably because they are less accessible for
pounds, which were obtained by chemical synthesis, it has the sulfating enzyme intermediate.
been shown that most of the derivatives show lower activi-
ties than unmodified resveratrol in in vitro assays.[3] How-
ever, resveratrol 3-sulfate showed a higher activity, and
Quercetin
Quercetin has five free hydroxy groups that could be sulf-
bearing in mind the relatively high amounts present in the
body, the less active derivatives may also play an important
role in the beneficial effects of resveratrol on human health.
Our easy enzymatic method for the preparation of sulfated
resveratrol esters facilitates these types of studies by giving
access to testing and reference compounds.
ated. A trial experiment was run [1 mL reaction volume, p-
NPS (5 mm), and quercetin (5 mm) at 30 °C], and samples
were taken for HPLC analysis. Upon incubation with the
AST, both of the substrates were consumed, and two new
species were formed; a compound with an elution time of
22.4 min, which is known to be p-NP, and a peak at
24.9 min, presumably corresponding to sulfated quercetin
Phloretin
Phloretin has four free hydroxy groups, all of which can, (not shown). This peak partially overlaps with the peak cor-
in theory, be sulfated. A test reaction monitored by HPLC responding to p-NPS, at 24.6 min. To obtain higher
showed the formation of two different products, so larger amounts of product for purification and further characteri-
scale reactions were performed under different conditions zation, a large-scale incubation (100 mL) was carried out.
to purify and analyse the sulfated products. When equi- The reaction was run for 4 h at 30 °C, with a small excess
molar amounts of the starting compounds phloretin and p- of p-NPS with respect to quercetin in the reaction mixture
Eur. J. Org. Chem. 2015, 534–541
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
537

R. Wever et al.
FULL PAPER
Figure 3. Formation of mono-, di-, and trisulfated resveratrol, as monitored by HPLC at 300 nm. Panel A, t = 0, and molar ratio
resveratrol/p-NPS = 1:2. Panel B, after incubation of this mixture for 24 h. Panels C, D, and E show the HPLC traces of purified
resveratrol 3-sulfate (23.4 min); resveratrol 3,4Ј-disulfate (24.7 min); and resveratrol 3,4Ј,5-trisulfate (25.9 min).
(5.5 mm and 5.0 mm, respectively). After purification of the taken from the reaction mixtures and analysed by HPLC
product as described in the Experimental Section, mass showed that during the enzymatic sulfation at pH 8–9, di-
spectrometry was used to confirm the formation of sulfated sulfated quercetin is not stable, and is probably sensitive to
quercetin. From the FAB-MS analysis, a product with an oxidation. Thus isolation, purification, and structural deter-
exact mass of m/z = 380.99 was found, which corresponds mination of this compound was not possible.
to monosulfated quercetin.[33] NMR spectroscopic analysis
Interestingly, another bacterial arylsulfotransferase has
(see Supporting Information) showed that the sulfate group been reported to sulfate quercetin.[34] This enzyme, from
had been introduced at the 4-position of quercetin. Upon Eubacterium A-44, was reported to seemingly only catalyse
longer incubations and addition of more p-NPS, a second the formation of quercetin 3,3Ј-disulfate (with equimolar
peak could be seen in the HPLC profiles. Presumably, this amounts of p-NPS and quercetin) and quercetin 3,3Ј,7-tris-
corresponds to a disulfated product. However, samples ulfate (with a tenfold excess of p-NPS).
538
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 534–541

Enzymatic Sulfation of Phenolic Hydroxy Groups
been shown already,[10] the enzyme is hardly deactivated by
these organic solvents, and it remained active during the
reaction for 2–7 d at 20–30 °C. This allowed the synthesis
of many compounds on useful scales (range 67 nmol–
1.3 mmol). By choosing the concentration of the reactants
and/or the enzyme, and the incubation time, it was possible
to obtain the desired sulfated products of p-coumeric acid,
6-hydroxyflavone, resveratrol, phloretin, and quercetin.
These sulfated compounds are of significant interest be-
cause of their formation as metabolites upon administration
of their biologically active parents, and also due to their
improved solubility. This bacterial arylsulfotransferase,
which uses the simple and cheap sulfate donor p-nitro-
phenyl sulfate, is a powerful tool in the synthesis of these
and other sulfated compounds.
Experimental Section
Heterologous Overproduction and Purification of AST: The AST
was expressed and purified using a method developed earlier[10]
with minor modifications. Briefly, E. coli BL21 (DE3) with
pJFT006 (carrying the D. hafniense AST gene) was grown, and
expression was induced using IPTG (isopropyl 1-thio-β-d-galacto-
pyranoside; 1 mm). Induction took place overnight at 25 °C. Rather
then direct lysis of the cells by sonication, the cells were first frozen
and stored at –80 °C. Freezing and thawing greatly increases the
enzyme yield. After cell lysis and centrifugation, DEAE (diethyl-
aminoethyl) Sephacel was added, and the pH was increased to pH
8. The DEAE Sephacel was poured into a column, the enzyme was
eluted and purified by FPLC (Poros column, Applied Biosystems).
The purified sample (10 Umg–1) was dialysed against Tris (50 mm,
pH 8.0), concentrated, and stored at –20 °C or –80 °C.
Figure 4. Order of sulfate transfer. The different sulfated species
that are formed in the AST-catalysed reaction of p-NPS with
resveratrol are shown, including their respective HPLC retention
times. The dashed arrows indicate the possible intermediate forma-
tion of a small amount of resveratrol 3,5-disulfate.
Chemicals: Chemicals were purchased from Sigma–Aldrich unless
otherwise stated.
Biotransformations: Typically, reactions were carried out by mixing
the sulfate acceptor with p-NPS in Tris–glycine buffer (pH 9.2),
followed by addition of the AST. The reaction mixture was stirred
at 30 °C in a closed round-bottomed flask. The concentrations of
the reactants used and other variable parameters are given in
Table 1. When the sulfate acceptor was poorly soluble in water,
acetone was used as cosolvent. This cosolvent could be easily re-
moved after or during the reaction.
Nonflavonoid Phenolic Acceptor Molecules
In addition to the flavonoids described above, a few un-
natural, nonflavonoid compounds were tested as substrates
for the AST, to investigate whether there is a trend regard-
ing the nature of the substrates that are accepted. These
reactions were carried out under the same conditions de-
scribed above; the addition of cosolvents was not necessary.
Both 2-phenylphenol and 2,2Ј-biphenol were accepted by
the enzyme as sulfate acceptors (Table 1), and 2-phen-
ylphenol sulfate and 2,2Ј-biphenol monosulfate were
formed. This shows that an OH group at an ortho position
can be sulfated, and that there is apparently no steric
hindrance preventing the 2-hydroxy groups from being
sulfated.
Purification of Products – General Procedure: After the conversion
was complete, the reaction mixture was extracted with ethyl acetate,
and the aqueous solution was acidified with acetic acid (1 m) to pH
5, and then freeze-dried in the presence of silica gel (10 g). This
material was then put onto a silica gel column, which was eluted
with n-butanol/acetic acid (99:1). In some cases, when the sulfate
esters were less soluble in n-butanol, an amount of methanol was
added to elute the pure product.
It was also possible to extract the crude sulfated compounds from
the acidified aqueous layer with n-butanol, but the best yields were
obtained by using direct column chromatography on silica gel. Suf-
ficient amounts of the sulfated compounds were isolated (scale
67 nmol–1.3 mmol) to allow characterization by NMR spec-
troscopy.
Conclusions
We have used the bacterial arylsulfotransferase (AST)
from Desulfitobacterium hafniense as a catalytic tool to ob-
tain sulfated natural flavonoid compounds. The flavonoid
compounds are sparingly soluble in water, and DMSO or
Analysis by HPLC, NMR spectroscopy, FAB-MS: Reactions were
monitored by HPLC. Samples from the reaction mixture were di-
acetone has to be present during the conversion. As has luted 100 times with acetonitrile/water (1:1), and 25 μL was injected
Eur. J. Org. Chem. 2015, 534–541
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
539

R. Wever et al.
FULL PAPER
Figure 5. Sulfation of phloretin, and purification of phloretin 4-sulfate and 4,4Ј-disulfate. The conversion of phloretin into phloretin 4-
sulfate (panels A and B) and phloretin 4,4Ј-disulfate (panels D–G), and of p-NPS into p-NP as followed by HPLC is shown. Two reactions
were performed in parallel, in one (panels A and B) the molar ratio phloretin/p-NPS was 1:1, in the second (panels d–G) the ratio was
1:2.2. Samples were taken at the start of the reaction (A and D), after 27 h (B), 20 h (E), and 117 h (F), and after the purification of the
sulfated compounds (C and G). The following peaks are identified: 22.8 min, phloretin; 24.6 min, p-NPS; 22.4 min, p-NP; 23.9 min,
phloretin 4-sulfate; 25.6 min, phloretin 4,4Ј-disulfate.
540
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 534–541

Enzymatic Sulfation of Phenolic Hydroxy Groups
[12]
M. W. F. Nielen, C. A. G. M. Weijers, J. Peeters, L. Weignerová,
H. Zuilhof, M. C. R. Franssen, World Mycotoxin J. 2014, 7,
107–113.
I. Ayuso-Fernández, M. A. Galmés, A. Bastida, E. García-Jun-
ceda, ChemCatChem 2014, 6, 1059–1065.
a) G. Malojcic, R. L. Owen, J. P. Grimshaw, M. S. Brozzo, H.
Dreher-Teo, R. Glockshuber, Proc. Natl. Acad. Sci. USA 2008,
105, 19217–19222; b) G. Malojcic, R. L. Owen, R. Glock-
shuber, Biochemistry 2014, 53, 1870–1877.
D. H. Kim, K. Kobashi, J. Biochem. 1991, 109, 45–48.
J. P. Grimshaw, C. U. Stirnimann, M. S. Brozzo, G. Malojcic,
M. G. Grutter, G. Capitani, R. Glockshuber, J. Mol. Biol.
2008, 380, 667–680.
L. H. Ren, F. Wang, Z. W. Xu, W. M. Chan, C. Y. Zhao, H.
Xue, Biochem. Pharmacol. 2010, 79, 1337–1344.
I. Gulcin, F. Topal, R. Cakmakci, M. Bilsel, A. C. Goren, U.
Erdogan, J. Food Sci. 2011, 76, C585–C593.
W. D. Hoff, P. Dux, K. Hard, B. Devreese, I. M. Nugteren-
Roodzant, W. Crielaard, R. Boelens, R. Kaptein, J. Van Beeu-
men, K. J. Hellingwerf, Biochemistry 1994, 33, 13959–13962.
a) F. Villa, D. Albanese, B. Giussani, P. S. Stewart, D. Daffon-
chio, F. Cappitelli, Biofouling 2010, 26, 739–752; b) S. Acham-
lale, B. Rezzonico, M. Grignon-Dubois, J. Appl. Phycol. 2009,
21, 347–352.
onto a Nucleosil 100-5 C-18 HD column (plus guard column),
equilibrated with NH4H2PO4 buffer (pH 6.3, 25 mm) containing
TBAP (ion-pairing reagent, tetrabutylammonium phosphate;
5 mm) and 5% acetonitrile. The HPLC dual pump system was an
[13]
[14]
Agilent 1100 system with
a 0–80% gradient (34 min run,
0.4 mLmin–1 flow rate) of acetonitrile (90% aq.), containing TBAP
(5 mm). The UV absorbance of the compounds was monitored at
300 nm. Peak areas of the detected substrates and products were
used to calculate the degree of sulfate transfer.
[15]
[16]
1H and 13C NMR spectroscopy was carried out with a Bruker
NMR instrument at 300, 400, and 500 MHz. Samples for 1H NMR
measurements were dissolved in [D6]DMSO, D2O, or CD3OD.
[17]
[18]
[19]
Fast-atom bombardment (FAB) mass spectrometry was carried out
with a JEOL JMS SX/SX 102A four-sector mass spectrometer, cou-
pled to a JEOL MS-MP9021D/UPD system program. Samples
were loaded in a matrix solution (3-nitrobenzyl alcohol) onto a
stainless-steel probe, and bombarded with Xenon atoms with an
energy of 3 keV. During the high-resolution FAB-MS measure-
ments, a resolving power of 10.000 (10% valley definition) was
used.
[20]
Acknowledgments
[21]
[22]
L. Biasutto, A. Mattarei, M. Zoratti, ChemBioChem 2012, 13,
1256–1259.
K. R. Patel, V. A. Brown, D. J. L. Jones, R. G. Britton, D. He-
mingway, A. S. Miller, K. P. West, T. D. Booth, M. Perloff, J. A.
Crowell, D. E. Brenner, W. P. Steward, A. J. Gescher, K. Brown,
Cancer Res. 2010, 70, 7392–7399.
Ing. E. Zuidinga and J. M. Ernsting are acknowledged for expert
help with NMR spectroscopic measurements. H. Peeters is ac-
knowledged for FAB-MS measurements. 2-Phenylphenol was
kindly provided by Dr. R. Wohlgemuth. This work was supported
by the Dutch National Research School Combination (NRSC Ca-
talysis).
[23]
T. Ridgway, J. O’Reilly, G. West, G. Tucker, H. Wiseman, Bio-
chem. Soc. Trans. 1996, 24, S391–S391.
[24] S. C. Marks, W. Mullen, G. Borges, A. Crozier, J. Agric. Food
Chem. 2009, 57, 2009–2015.
[1] a) A. Soto-Vaca, A. Gutierrez, J. N. Losso, Z. Xu, J. W. Finley, [25] M. G. L. Hertog, E. J. M. Feskens, P. C. H. Hollman, M. B.
J. Agric. Food Chem. 2012, 60, 6658–6677; b) T. Hirano, M.
Gotoh, K. Oka, Life Sci. 1994, 55, 1061–1069.
[2] M. Correia-da-Silva, E. Sousa, M. M. Pinto, Med. Res. Rev.
2014, 34, 223–279.
[3] J. Hoshino, E. J. Park, T. P. Kondratyuk, L. Marler, J. M. Pez-
zuto, R. B. van Breemen, S. Y. Mo, Y. C. Li, M. Cushman, J.
Med. Chem. 2010, 53, 5033–5043.
Katan, D. Kromhout, Lancet 1993, 342, 1007–1011.
[26] A. J. Day, F. Mellon, D. Barron, G. Sarrazin, M. R. Morgan,
G. Williamson, Free Radical Res. 2001, 35, 941–952.
[27] J. Y. Yang, M. A. Della-Fera, S. Rayalam, S. Ambati, D. L.
Hartzell, H. J. Park, C. A. Baile, Life Sci. 2008, 82, 1032–1039.
[28] A. R. Ibrahim, Y. J. Abulhajj, Appl. Environ. Microbiol. 1989,
55, 3140–3142.
[4] L. J. Ingram, S. D. Taylor, Angew. Chem. Int. Ed. 2006, 45, [29] M. Vietri, A. Pietrabissa, R. Spisni, F. Mosca, G. M. Pacifici,
3503–3506; Angew. Chem. 2006, 118, 3583–3586.
[5] a) R. A. Al-Horani, U. R. Desai, Tetrahedron 2010, 66, 2907–
2918; b) S. D. Taylor, A. Desoky, Tetrahedron Lett. 2011, 52,
3353–3357.
[6] E. Chapman, M. D. Best, S. R. Hanson, C. H. Wong, Angew.
Chem. Int. Ed. 2004, 43, 3526–3548; Angew. Chem. 2004, 116,
3610–3632.
Xenobiotica 2002, 32, 563–571.
[30] C. H. Yang, L. Tang, C. Lv, L. Ye, B. J. Xia, M. Hu, Z. Q. Liu,
J. Pharm. Pharmacol. 2011, 63, 967–970.
[31] a) E. Wenzel, T. Soldo, H. Erbersdobler, V. Somoza, Mol. Nutr.
Food Res. 2005, 49, 482–494; b) E. Wenzel, V. Somoza, Mol.
Nutr. Food Res. 2005, 49, 472–481.
[32] a) G. Kuhnle, J. P. E. Spencer, G. Chowrimootoo, H. Schroeter,
E. S. Debnam, S. K. S. Srai, C. Rice-Evans, U. Hahn, Biochem.
Biophys. Res. Commun. 2000, 272, 212–217; b) K. R. Patel, C.
Andreadi, R. G. Britton, E. Horner-Glister, A. Karmokar, S.
Sale, V. A. Brown, D. E. Brenner, R. Singh, W. P. Steward, A. J.
Gescher, K. Brown, Sci. Transl. Med. 2013, 5.
[33] a) H. van der Woude, M. G. Boersma, J. Vervoort, I. M. C. M.
Rietjens, Chem. Res. Toxicol. 2004, 17, 1520–1530; b) D. J. L.
Jones, R. Jukes-Jones, R. D. Verschoyle, P. B. Farmer, A.
Gescher, Bioorg. Med. Chem. 2005, 13, 6727–6731.
[34] M. Koizumi, M. Shimizu, K. Kobashi, Chem. Pharm. Bull.
1990, 38, 794–796.
[7] H. O. Gulcan, M. W. Duffel, Arch. Biochem. Biophys. 2011,
507, 232–240.
[8] C. H. Lin, G. J. Shen, E. Garcia-Junceda, C. H. Wong, J. Am.
Chem. Soc. 1995, 117, 8031–8032.
[9] a) M. D. Burkart, M. Izumi, E. Chapman, C. H. Lin, C. H.
Wong, J. Org. Chem. 2000, 65, 5565–5574; b) M. D. Burkart,
M. Izumi, C. H. Wong, Angew. Chem. Int. Ed. 1999, 38, 2747–
2750; Angew. Chem. 1999, 111, 2912–2915.
[10] M. A. van der Horst, J. F. T. van Lieshout, A. Bury, A. F. Har-
tog, R. Wever, Adv. Synth. Catal. 2012, 354, 3501–3508.
[11] P. Marhol, A. F. Hartog, M. A. van der Horst, R. Wever, K.
Purchartova, K. Fuksova, M. Kuzma, J. Cvacka, V. Kren, J.
Mol. Catal. B 2013, 89, 24–27.
Received: July 4, 2014
Published Online: December 12, 2014
Eur. J. Org. Chem. 2015, 534–541
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
541