Sulfation of various alcoholic groups by an arylsulfate sulfotransferase from desulfitobacterium hafniense and synthesis of estradiol sulfate

Article abstract of DOI:10.1002/adsc.201200564

Bacterial arylsulfate sulfotransferases (AST) are enzymes that catalyse the transfer of a sulfate group from p-nitrophenyl sulfate (p-NPS) to a phenolic acceptor molecule. By screening of the NCBI protein database a gene coding for an AST was found in Desulfitobacterium hafniense. After expression the enzyme was purified and characterised. This AST efficiently sulfates various acceptor molecules (estrone, estradiol, enkephalin and non-phenolic alcohols) using p-NPS as sulfate donor. The purified AST has a pH optimum of 9.6, it is stable in the presence of 10% of DMSO, and depending on the conditions it has a melting temperature of up to 47°C. Surprisingly, and in great contrast to all other known bacterial ASTs, this enzyme was able to use a variety of non-phenolic alcohols as sulfate acceptor. Because of these properties, this unique enzyme is a promising tool for biotransformation processes, providing a green and simple method to specifically sulfate compounds without need for functional group protection.

Full text of DOI:10.1002/adsc.201200564

FULL PAPERS
DOI: 10.1002/adsc.201200564
Sulfation of Various Alcoholic Groups by an Arylsulfate
Sulfotransferase from Desulfitobacterium hafniense and
Synthesis of Estradiol Sulfate
a
a
a
Michael A. van der Horst, Johan F. T. van Lieshout, Aleksandra Bury,
a
a,
Aloysius F. Hartog, and Ron Wever *
Van ꢀt Hoff Institute for Molecular Sciences, Faculty of Sciences, University of Amsterdam, Science Park 904, 1098 XH
a
Amsterdam, The Netherlands
E-mail: r.wever@uva.nl
Received: July 6, 2012; Published online: December 4, 2012
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201200564.
Abstract: Bacterial arylsulfate sulfotransferases conditions it has a melting temperature of up to
(
AST) are enzymes that catalyse the transfer of a sul- 478C. Surprisingly, and in great contrast to all other
fate group from p-nitrophenyl sulfate (p-NPS) to known bacterial ASTs, this enzyme was able to use
a phenolic acceptor molecule. By screening of the a variety of non-phenolic alcohols as sulfate accept-
NCBI protein database a gene coding for an AST or. Because of these properties, this unique enzyme
was found in Desulfitobacterium hafniense. After ex- is a promising tool for biotransformation processes,
pression the enzyme was purified and characterised. providing a green and simple method to specifically
This AST efficiently sulfates various acceptor mole- sulfate compounds without need for functional group
cules (estrone, estradiol, enkephalin and non-phenol- protection.
ic alcohols) using p-NPS as sulfate donor. The puri-
fied AST has a pH optimum of 9.6, it is stable in the Keywords: biocatalysis; green chemistry; regioselec-
presence of 10% of DMSO, and depending on the tivity; steroids; sulfation
Introduction
of sulfur trioxide (SO ) with tertiary amines or
amides. However, use of this reactive reagent suffers
3
Sulfate monoesters are of great importance to many from numerous disadvantages such as harsh reaction
biological systems. Sulfation (also: sulfurylation) of conditions and lack of reaction selectivity. Additional-
a molecule changes many physico-chemical properties ly, in order to prevent side reactions of labile func-
of the acceptor molecule, such as charge, size, and im- tionalities and to enhance the chemo- or regioselectiv-
portantly, solubility. In the synthesis of these com- ity of the overall reaction, these functionalities have
[
9–11]
pounds sulfotransferases are involved that catalyse to be protected.
Enzymatic methods for sulfation
the transfer of a sulfuryl group to a hydroxy or amine of compounds under mild conditions may have clear
moiety on various molecules including small-molecule advantages and the possibility to use enzymes for this
[
12–14]
drugs, steroids, hormones, carbohydrates, and pro- purpose has been studied.
[1,2]
teins.
Sulfate conjugation in mammals is a major
Two families of sulfotransferase enzymes can be
pathway for the biotransformation of phenolic and distinguished for the transfer of a sulfate group from
catechol drugs and forms a detoxification route in the a donor molecule to the hydroxy group of an acceptor
[3]
[1]
liver. Recent studies have implicated roles of sulfo- molecule (for a review, see ref. ). In eukaryotes,
transferases in a number of disease states including a family is found that uses 3’-phosphoadenosine 5’-
[
4,5]
[6]
entry of HIV,
chronic inflammation, and various phosphosulfate (PAPS) as the sulfate-donor molecule
[7]
forms of cancer. Because of this great biological for the sulfation of for example, sugars, hormones and
relevance, there is growing interest in the synthesis of antibiotics. Use of these enzymes is problematic since
sulfated molecules.
[
8]
PAPS is very expensive and has to be regenerated, in-
Sulfated compounds are in general synthesised by troducing another enzyme and sulfate donor in the
the pharmaceutical/chemical industry using complexes system and further these sulfotransferases are in gen-
Adv. Synth. Catal. 2012, 354, 3501 – 3508
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3501

Michael A. van der Horst et al.
FULL PAPERS
[
1]
eral donor- and substrate-specific. The other family bacterium hafniense can be found, annotated as aryl-
of sulfotransferases found in bacteria uses phenolic sulfotransferase. This sequence codes for a protein of
sulfate esters such as p-nitrophenyl sulfate or p-ace- 628 residues corresponding to a molecular mass of
tylphenyl sulfate as a sulfate-donor. The bacterial sul- 71.4 kDa. Sequence alignment of ASTs from different
fotransferase [arylsulfate sulfotransferase; A(S)ST, organisms shows there might be different classes of
EC 2.8.2.22] was originally discovered in the bacteri- AST proteins (an alignment of some examples from
[
15,16]
um Eubacterium A-44 from human intestine
more recently many more bacteria
and both classes are shown in the Supporting Information,
have been Figure S1). One class (Class I) represents proteins
found to contain PAPS-independent aryl sulfotrans- mainly from proteobacteria and the sequences show
[
17–21]
[2]
ferases (see ref. for a review). The biological role of a high similarity. Depending on the amount of disul-
these bacterial ASTs has not been elucidated. Only fide bridges in the protein, and the genomic context
for the recently discovered AST from Streptomyces of the encoding gene, this class can be further subdi-
sp. MK730-62F2 a physiological role was shown. This vided. The other class (Class II) consists of proteins
enzyme acts as a sulfotransferase in liponucleoside which show internally less sequence similarity and
biosynthesis: in this organism it is responsible for the originate from firmicutes. According to the sequence
[
27]
[
22]
formation of sulfated liponucleoside antibiotics.
similarity AST from D. hafniense belongs to Class II.
Kinetic studies on AST from Eubacterium A-44 The sequence similarity between the two main classes
demonstrated a ping-pong bi-bi reaction mecha- of AST is relatively low (for example, 23% identity
[23]
nism. A crystal structure of the homodimeric AST between the AST proteins from Eubacterium and E.
from the uropathogenic strain E. coli CFT073 in the coli). Furthermore, the sequences of Class II are
[
24]
presence of various sulfuryl donors showed that the about 30 residues longer than Class I sequences. The
[
19]
sulfate group is bound transiently to a catalytic histi- tyrosine residue identified by Kwon et al. to be es-
dine residue. This mechanism bears resemblance to sential for enzymatic activity in AST from Enterobact-
the mechanism of transphosporylation catalysed by er ambigenus is only conserved in the sequences of
a group of acid phosphatases. During transfer of Class I enzymes. The region of the sequence in which
a phosphate donor to a phosphate acceptor molecule this residue is located is highly non-homologous be-
by these enzymes an activated enzyme phospho-histi- tween the two classes. Another difference that can be
[
25,26]
dine intermediate is formed.
Interestingly, earlier observed is the number, and location, of cysteine resi-
site-directed mutagenesis studies on AST from Enter- dues. Class I contains three conserved cysteine resi-
obacter ambigenus implied transient sulfurylation of dues (not shown), whereas Class II enzymes contain
[
19]
a tyrosine residue. Different classes of AST can be eight, and several cysteines are found that are ran-
[27]
discerned, possibly having different reaction mecha- domly present. In Class I the cysteines are involved in
nisms (also see Results and Discussion section). disulfide bond formation and are important for cata-
[
28]
The bacterial ASTs have been much less character- lytic activity. Interestingly the spatial arrangement
ised than the eukaryotic PAPS-dependent sulfotrans- of the active site residues in PAPS dependent sulfo-
ferases. However, the ASTs are highly interesting for transferases is the same as that in the bacterial
[
29]
biotransformation purposes because of the lower cost ASTs.
and higher stability of p-NPS compared to PAPS.
Since the enzymes also have a broader substrate spe-
[12]
cificity these enzymes may be potentially useful in Cloning, Expression and Purification
the selective and mild sulfation of many compounds
such as peptides, steroids and carbohydrates. Here we To determine the functionality of AST, its gene was
report successful cloning and expression of the astA amplified from D. hafniense genomic DNA and
gene from Desulfitobacterium hafniense and explored cloned in the vector pET26b. The resulting plasmid,
its donor and acceptor specificity.
pJFT006, was used to transform E. coli BL21 ACHTUNGTRENNUNG( DE3)
cells. The temperature during induced expression was
lowered to 258C to reduce the amount of expressed
protein appearing as inclusion bodies. Though cloning
in pET26b promotes heterologous expression to the
periplasm, AST was found in large amounts in the cy-
toplasm as well. Accordingly, the enzyme was purified
Results and Discussion
Identification of the ast Gene
In the NCBI protein database (http://www.ncbi.nlm. from whole cells, by following a conventional purifica-
nih.gov/protein), the number of protein sequences tion strategy, as described in the Experimental Sec-
that encode (putative) ASTs is growing rapidly. These tion. The gene codes for a protein with a molecular
either have been annotated as AST, or can be found mass of 71.4 kDa and in agreement with this, SDS-
manually using BLAST analyses. Among these, a se- PAGE of the purified AST resulted in a single band
quence from the dehalorespiring bacterium Desulfito- of about 70 kDa (not shown). This molecular mass is
3502
asc.wiley-vch.de
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2012, 354, 3501 – 3508

Sulfation of Various Alcoholic Groups by an Arylsulfate Sulfotransferase
[
2]
close to that of the subunit of the E.coli sulfotrans-
ferase.
Enzymatic Activity, pH Optimum and Temperature
Stability
Arylsulfate sulfotransferases require two different
substrates for activity, one to act as a sulfate donor,
the other as sulfate acceptor. Absence of either ac-
ceptor or donor does not result in any measurable ac-
tivity. Among the different sulfated compounds tested
as a donor, only the arylsulfates could be used (see
Figure S2, Supporting Information) and the alkylsul- Figure 1. Sulfation activity of AST at different pH values,
fates were inactive. This may have a thermodynamic both with phenol (closed circles) and tyramine (closed
squares) as a sulfate acceptor. Relative activities are shown.
origin but selectivity of the active site for an aromatic
Activity at pH 9.75 is taken as 100%. Reactions were car-
group cannot be excluded. In all further experiments,
ried out using 5 mM of p-NPS and 5 mM of acceptor in
p-nitrophenyl sulfate was used as the sulfate donor. A
1
00 mM Tris-HCl at the various pH values.
variety of phenols were tested for their potential as
sulfate acceptor for this enzyme. The results, which
[
15–19,30]
[31]
are in line with the results found previously
are pK for histidine is usually lower. Mozhaev et al. re-
a
summarised in Table 1. Except for p-hydroxybenzoic ported that the AST from Clostridium innocuum has
[
12]
acid, all phenolic compounds were good sulfate ac- optimal activity at slightly lower pH (8.0–8.5).
ceptors. Since many potential substrates for AST are poorly
In reactions with phenol and tyramine a sharp, al- soluble in water, the use of organic solvents is often
kaline, pH dependency was found with optimal sulfa- required to guarantee adequate substrate concentra-
tion activity at pH 9.5–9.8 (Figure 1). The sharp in- tion. A range of organic solvents was used to study
crease in activity suggests that an acid base group their effect on the catalytic activity. Addition of 10%
with a high pK is involved in catalysis and points to ethanol, methanol and methylene chloride reduced
a
the involvement of phenolic OH group (which has the enzymatic activity to 50–65%. In DMF, THF, di-
a pK value of about 10). The involvement of a histi- oxane and acetone the activity of the enzyme de-
a
dine residue having this value is less likely, since the creased to 30–34% of the original activity. However,
1
2
0% DMSO had no effect on the activity and even in
5% of DMSO, the remaining activity was still 73%.
Table 1. Phenolic sulfate acceptors. Relative activity of AST Therefore in general for example, as in Table 2, 10%
with different sulfate acceptors. Reactions were performed DMSO was used to ensure substrate solubility. For
in 50 mM Tris-glycine buffer (pH 9.0), with 5 mM of p-NPS prolonged incubations, and for preparative catalytic
as the sulfate donor and 5 mM of acceptor.
reactions, 10% acetone was used instead of DMSO
because of its easy removal.
[a]
Sulfate acceptor
Relative reaction rate
For synthetic applications the AST should be suffi-
ciently stable and therefore the thermostability of
AST was investigated at different pH values and in
the absence or presence of substrates. The enzyme
was incubated exactly 5 min at a given temperature
after which the activity was measured. From a plot of
the remaining activity against the incubation tempera-
ture, the midpoint of the denaturation curve (Tm,
melting temperature) was determined (Figure 2). A
melting temperature of 39.28C was found at the near
optimal pH of 9.0 that increased to 41.78C at pH 8
(not shown).
[
b]
phenol
100, 106
70
195
138
540
resorcinol
p-Cl-phenol
m-Cl-phenol
dopamine
p-cresol
200
tyramine
60[
b]
1
2
4
5
4
-naphthol
-naphthol
-hydroxybenzoic acid
-hydroxyindole
-hydroxyindole
50
[
b]
86
[b]
0.7
[
b]
b]
74[
88
[b]
p-aminophenol
o-aminophenol
102[
At both higher (pH 9.75) or lower pH (6.5) values
the melting temperatures decreased to 30.98C and
b]
136
[b]
3,5-dimethoxyphenol
33
36.88C, respectively. The presence of a high concen-
[
a]
tration of salt (500 mM NaCl) at pH 6.5 had hardly
Initial reaction rates for different substrates are normal-
ised to the rate of sulfation with phenol (100%).
These transformations were performed in 10% DMSO.
any effect on the melting temperature (DT =0.98C).
m
[
b]
The stability of AST in prolonged catalytic incuba-
Adv. Synth. Catal. 2012, 354, 3501 – 3508
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3503

Michael A. van der Horst et al.
FULL PAPERS
Figure 2. Thermostability of AST. Thermal denaturation
curve of arylsulfate sulfotransferase at pH 9.0 in the absence
and presence of acetone and p-nitrophenyl sulfate. Samples
were incubated for 5 min at the given temperatures. pH 9.0
(
closed circles), containing 5 mM p-nitrophenyl sulfate
(
open circles), 10% acetone (closed squares), 10% acetone
and 5 mM p-nitrophenyl sulfate (open squares). Remaining
activity was measured at standard assay conditions.
tions containing acetone was investigated by assessing
the melting temperature of AST in the presence of p-
NPS and 10% acetone at pH 9.0. As is evident from
Figure 2, the addition of 5 mM p-NPS causes a signifi-
cant increase in melting temperature, whereas 10% of
acetone results in a decreased melting temperature
(
T =46.98C and 32.58C, respectively). When both p-
m
NPS and acetone were present in the incubation mix-
ture, the destabilising effect of acetone was dimin-
ished and the melting temperature (T =44.98C) was Figure 3. Steroids used in this study. The compounds in
m
considerably higher than that without the addition of panel B are good sulfate acceptors for AST, whereas the
compounds depicted in panel A do not accept sulfate from
either.
AST. Reactions were performed in 50 mM Tris-glycine
buffer (pH 9.0), with 5 mM of p-NPS as the sulfate donor
and 5 mM of acceptor. Reaction progress was judged by the
increase of absorption at 410 nm in the reaction mixture, in-
dicative of p-NP formation.
Sulfation of Steroids and Complete Conversion of
Estrone
Sulfotransferases and sulfated steroids are of consid-
erable toxicological, pharmacological and pharma-
In addition to this qualitative test of sulfation of
ceutical interest and therefore the sulfation of a set of steroids the extent of sulfation of estrone was deter-
steroids as substrates for AST was studied. As mined quantitatively. For this, the p-NPS and estrone
Figure 3 shows steroids that have an aromatic 3’-OH were incubated in equimolar amounts (5 mM) and the
were good sulfate acceptors. This test was based on reaction was followed by HPLC over time (Figure 4).
the formation of a yellow colour due to the formation While estrone and p-NPS gradually disappeared, two
of p-NP, when the steroid was incubated with p-NPS new products were simultaneously formed, which as
and AST. As a control, hydrolysis of p-NPS was judged by the elution time are p-nitrophenol and es-
monitored in the absence of the steroid and in the trone sulfate, respectively. After 24 h 90% of the es-
presence of the enzyme, however, hydrolysis of the p- trone was sulfated. When another aliquot of 5 mM p-
NPS did not occur under those conditions and thus nitrophenyl sulfate was added the residual estrone
the aryl sulfotransfer reaction was strictly dependent was sulfated within 2 h and less than 1% remained.
on the presence of the steroid. No measurable activity This demonstrates that the enzyme is stable during
was detected with steroids like nandrolone and deso- the incubation, also under turnover conditions.
gestrel that lack the 3’- hydroxy group. This also
shows that the 17’-OH group is not sulfated.
To test if the sulfation reaction using AST could be
up scaled, a 100-mg scale conversion of 17b-estradiol
was performed. This also allows the isolation and
3504
asc.wiley-vch.de
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2012, 354, 3501 – 3508

Sulfation of Various Alcoholic Groups by an Arylsulfate Sulfotransferase
Sulfation of Enkephalin
Previously, some ASTs have been reported to be able
[
10,20,32]
to sulfate tyrosine-containing small peptides.
In
line with this, also AST from D. hafniense was able to
sulfate peptides like Leu-enkephalin. Analogous to
estrone and estradiol, the reaction with p-nitrophenyl
sulfate (10 mM) and Leu-enkephalin (10 mM) was
followed by HPLC over time. Within 48 h 91% of
Leu-enkephalin was sulfated which reached 98%
after 6 days of incubation (see the Supporting Infor-
mation, Figure S3).
Sulfation of Aliphatic Alcohols
In contrast to all previously reported ASTs, AST
from D. hafniense was able to use a variety of non-
phenolic alcohols as sulfate acceptor. Table 2 shows
a number of these.
The qualitative test, in which 100 mM of the accept-
or was used, was again based on the formation of
a yellow colour due to the formation of p-NP. As
a control, hydrolysis of p-NPS was monitored in the
absence of the acceptor and in the presence of the
enzyme, however, hydrolysis of the p-NPS did not
occur under those conditions and thus the aryl sulfo-
transfer reaction was strictly dependent on the pres-
Table 2. Aliphatic sulfate acceptors. The sulfation of various
aliphatic alcohols was determined semi-quantitatively. Reac-
tions were carried out in 50 mM Tris-glycine, pH 9.0, with
Figure 4. Sulfation of estrone followed by HPLC. The con-
version of estrone to estrone-sulfate and p-NPS to p-NP as
followed by HPLC. At the time intervals given, samples
were taken and analysed by HPLC. The following peaks
could be identified: 28.1 min p-NP; 33.4 min p-NPS; 36.2 es-
1
00 mM of acceptor and 10 mM of donor. Reaction progress
was judged by the increase of absorption at 410 nm in the
reaction mixture, indicative of p-NP formation. + + + cor-
responds to 50% conversion in 7 days, + + corresponds to
1
0% conversion in 7 days, + corresponds to 2–10% conver-
trone sulfate and at 37.3 min estrone. Reaction conditions:
sion in 7 days.
À1
0
.1 unitmL AST, 5 mM p-NPS, 5 mM estrone in 100 mM
Tris-HCl (pH 8.0).
Sulfate acceptor
Conversion
1
2
1
1
2
1
2
-butanol
-butanol
-pentanol
-octanol
-octanol
+ +
+
[a]
analysis of the reaction products. After 24 h about
+ +
+ +
8
0% of the 17b-estradiol was converted, but longer
incubation even for another day or addition of
.25 mM AST and 1 mM p-NPS did not result in addi-
[
a]
+
+
0
[b]
-phenylethanol
-phenylethanol
tional conversion. The isolated yield of the final prod-
uct (17ß-estradiol 3-sulfate as a sodium salt) was
+ + +
+
dihydroxyacetone
glycerol
79 mg (57%) and, after analysis by NMR and HPLC,
+ +
+ +
+ +
+ +
+
was calculated to be >90% pure (see the Supporting cyclohexanol
1
Information, Figure S4 for the H NMR spectrum of furfuryl alcohol
1
7ß-estradiol 3-sulfate). The impurities consisted of p- benzyl alcohol
d-glucose
octyl glucoside
NP, p-NPS, and less than 1% 17ß-estradiol.
+
N-hydroxysuccinimide
+
[
[
a]
b]
For these compounds S(+) was the preferred isomer.
For these compounds S(À) was the preferred isomer.
Adv. Synth. Catal. 2012, 354, 3501 – 3508
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3505

Michael A. van der Horst et al.
FULL PAPERS
ence of the aliphatic alcohol. It should be noted that
the reactions are very slow, requiring days.
In addition to sulfation of aliphatic OH groups,
AST was able to sulfate the N-substituted hydroxy
group of N-hydroxysuccinimide. Of the compounds
with non-phenolic hydroxy groups, 2-phenylethyl al-
cohol was sulfated with the highest rate and its con-
version was examined at different pH values and ana-
lysed by HPLC (Figure 5).
The pH dependence was similar to that seen in the
sulfation of phenol and tyramine as acceptors; higher
conversions were observed at increasing pH values.
When 10 mM p-NPS were incubated with 100 mM 2-
phenylethyl alcohol for 4 days at pH 9.0 the sulfate
donor was completely desulfated and about 10 mM of
sulfated 2-phenylethyl alcohol was formed, as judged
from HPLC analysis. This was confirmed using sulfat-
ed 2-phenylethyl alcohol which was separately syn-
[
33]
thesised chemically . Surprisingly, secondary alco-
hols (Table 2) are also sulfated though the rate is sig-
nificantly slower. Further when 1-phenylethyl alcohol
was used as acceptor, clear preference for the S(À)
isomer can be observed.
Conclusions
We cloned the astA gene from Desulfitobacterium
hafniense and successfully expressed the encoded
AST protein. We characterised the biochemical prop-
erties of the recombinant protein and we tested the
acceptor specificity and found a wide range of not
only phenolic, but surprisingly also aliphatic alcohols
that can act as sulfate acceptor for this enzyme. The
rate of sulfate transfer to aliphatic alcohols is substan-
tially slower than to phenolic alcohols. However, by
varying reaction parameters such as enzyme and ac-
ceptor concentration, we estimate that this can be sig-
nificantly improved. Furthermore, enzyme modifica-
tions, for example, by using a directed evolution ap-
proach, will most likely result in improved rates of
sulfate transfer to aliphatic alcohols and this would
expand further the scope of the application of this
enzyme.
Experimental Section
Figure 5. Sulfation of 2-phenylethyl alcohol at different pH
values, followed by HPLC. Sulfation of 2-phenylethyl alco-
hol by AST at several pH values. 0.3 unitmL AST was in-
Chemicals
À1
cubated with 10 mM p-NPS and 100 mM 2-phenylethyl alco-
hol in 100 mM acetate (pH 5) or Tris-HCl (pH 6, 7, 8, 9 or
All steroids, including estrone sulfate, were kindly provided
by Diosynth (Organon, Oss, The Netherlands). 4-Methylum-
belliferyl sulfate and p-nitrophenyl sulfate were obtained
from Sigma. Octyl sulfate and phenyl sulfate were prepared
1
3
0) containing 10% acetone. After 96 h of incubation at
08C samples were taken and analysed by HPLC. The
[33]
peaks eluting at 29.0 min and 33.4 min correspond to p-NP
and p-NPS, respectively. The peak at 31 min has the same
elution time as 2-phenylethyl sulfate that was synthesised
chemically.
from the corresponding alcohol using triethylamine-SO3.
All other sulfate donors were prepared as previously de-
[34]
scribed. Leu-enkephalin and leu-enkephalin sulfate were
obtained from Bachem.
3506
asc.wiley-vch.de
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2012, 354, 3501 – 3508

Sulfation of Various Alcoholic Groups by an Arylsulfate Sulfotransferase
Bacterial Hosts and Vector
prewarming the cuvette in a water bath for at least 5 min,
0 mL AST extract were added and the release of p-nitro-
phenol was monitored at 405 nm on a Varian Cary50 spec-
trophotometer. The rate of enzymatic sulfate transfer was
calculated from the initial linear part of the kinetic curve,
5
The T7 expression vector pET26b was obtained from Nova-
gen. Escherichia coli XL-1 blue (Stratagene) was used as an
initial host for cloning while E. coli BL21 (DE3) (Strata-
gene) was used as an expression host for the pET-derivative.
E. coli was grown in TY medium in a rotary shaker at 378C.
using the molar extinction coefficient for p-NP (e
=
405
À1
À1
1
6.6 mM cm ). One unit of activity is defined as the
Kanamycin was added to
a
final concentration of
amount of enzyme catalysing the formation of 1 mmol p-NP
per minute. All activities were corrected for non-enzymatic
release of p-nitrophenol.
À1
3
0 mgmL .
Cloning of the astA Gene
The pH optimum of the enzyme was determined in a dis-
continuous assay in Tris-glycine buffer over pH range 4–11
at 308C. Enzyme was added to buffered solutions of 5 mM
p-NPS and 5 mM phenol or tyramine and after 5 min the re-
action was stopped by placing the tubes on ice, followed by
the addition of NaOH and the absorbance at 405 nm was
measured.
For measurements to determine the thermostability, sam-
ples were incubated at the indicated temperatures for 5 min.
After that, activity was measured as described above.
D. hafniense genomic DNA, a kind gift from Krisztina
Gabor (Laboratory of Microbiology, Wageningen Universi-
ty, Wageningen, The Netherlands), was used as template in
a PCR reaction. A primer set (BspHI and SacI restriction
sites underlined) was designed to amplify astA by PCR on
an iCycler Thermal Cycler (BioRad): 5’-GCGCGCTCAT
GAATCCGATCAAAAGTGAACAGAT-3’ and 5’ GCGCGCGAGCT
CAAGCTGTAATAGACACTCCGGTC-3’. In addition to 10 ng of
D. hafniense chromosomal DNA and 100 ng of each primer,
the 50 mL mixture contained 0.2 mM dNTPs, Pfu buffer and
HPLC
2.5 U Pfu DNA polymerase. The PCR product with the ex-
pected size was digested and ligated into NcoI/SacI-digested
vector pET26b, resulting in pJFT006. E. coli XL1-blue and
BL21 (DE3) were transformed with this plasmid. Sequence
analysis of pJFT006 by MWG Biotech AG (Martinsried,
Germany) confirmed the correct cloning of the astA gene.
Enzymatic assays with donor substrates other than p-NPS or
with a steroid as acceptor substrate were monitored by
HPLC. 5 mM of donor substrate were mixed in an Eppen-
dorf tube with 5 mM of acceptor substrate in a 50 mM Tris-
glycine buffer (pH 9.0) and incubated at 308C in an Eppen-
dorf Thermomixer compact. After 5 min, 50 mL of AST ex-
tract were added and samples were taken at regular time in-
tervals. The sample was diluted twenty times in 10% aceto-
nitrile/water and injected on a Nucleosil 100–5 C-18 HD
column (plus guard column), equilibrated with 25 mM
NH H PO buffer (pH 6.3) containing 5 mM TBAP (tetra-
Expression and Purification of AST
An overnight culture of E. coli BL21 (DE3) cells with
pJFT006 was used to inoculate LB medium containing
À1
5
0 mgmL kanamycin. When the OD600 reached 0.5, the
4
2
4
cells were induced with 1 mM IPTG and subsequently
grown at 258C for approximately 16 h. Cells were harvested
by centrifugation (4000ꢂg for 10 min.) and resuspended in
a 50 mM Tris-HCl buffer (pH 8.0). Cells were lysed by soni-
cation (4 min, output 7, duty cycle 50%) on a Branson soni-
fier. After removal of the cell debris by centrifugation
butylammonium phosphate) as an ion-pairing reagent, and
5% acetonitrile. The HPLC dual pump system was equipped
with 2 LKB 2150 pumps and a LKB 2152 controller. An in-
jection loop of 50 mL was used. A 0–60% gradient (43 min
À1
run, 0.4 mLmin flow rate) of 90% acetonitrile, containing
5 mM TBAP, was applied. The UV absorbance of the eluate
was monitored both at 210 nm (Figure 4 and Figure 5) and
at 254 nm using a Pharmacia LKB VWM2141 dual diode
array variable wavelength detector. For data collection and
integration the Borwin 1.21 software was used. Peak areas
of the detected substrates and products were used to calcu-
late the degree of sulfate transfer.
(
20,000ꢂg for 30 min.) Sephacel DEAE was added to the
supernatant and stirred for 2 h. After settling the Sephacel
was poured into a column, which was washed extensively
with 100 mM NaCl in 50 mM Tris (pH 8). The AST was
eluted by a stepwise NaCl gradient. Fractions containing
AST activity were applied onto a Poros column (Applied
Biosystems) using an Amersham Pharmacia FPLC system.
The protein was eluted using a linear gradient towards 1M
NaCl. Fractions containing AST activity were pooled, con-
centrated using Amicon Centricon concentrators, and buf-
fered in 50 mM Bis-Tris (pH 6.0). This sample was applied
onto a Q-FF column (GE Healthcare), equilibrated with
“
Large” Scale Estradiol Sulfation (Reaction
Conditions, Product Isolation and Analysis)
To a 100-mL round-bottom flask containing 3.7 mL acetone,
9.25 mL 400 mM Tris/glycine (pH 8.5), 11.95 mL water,
10 mL 40 mM p-NPS, 100 mg 17b-estradiol were added and
the reaction was started by the addition of 2.1 mL 8.8 mM
AST. This resulted in a solution of 37 mL containing 10%
acetone/10 mM 17b-estradiol/10.8 mM p-NPS and 0.5 mM
AST in 100 mM buffer.
50 mM Bis-Tris (pH 6.0). In a linear gradient towards 1M
NaCl, AST eluted between 400 and 500 mM NaCl. The
eluted fractions were analysed using SDS-PAGE and used
for enzymatic assays and biotransformations.
This solution was kept at of 308C and stirred in a closed
round-bottom flask. The reaction was followed by taking
samples and analysis by HPLC after dilution in 10% aceto-
nitrile/water (see above). When maximal conversion was
reached, the solution was extracted with three times 20 mL
CH Cl which removed acetone, 50% of the p-NP and the
Enzymatic Assays
Standard enzymatic assays were performed at 308C. A
1200 mL quartz cuvette typically contained 50 mM Tris-gly-
cine buffer (pH 9.0), 5 mM of p-nitrophenyl sulfate as the
sulfate donor and 5 mM of acceptor (usually phenol). After
2
2
Adv. Synth. Catal. 2012, 354, 3501 – 3508
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3507

Michael A. van der Horst et al.
FULL PAPERS
remaining traces of 17b-estradiol (the purification was
adapted from ref. ). The water solution was extracted
three times with 20 mL n-butanol resulting in extraction of
[10] B. Guilbert, N. J. Davis, M. Pearce, R. T. Aplin, S. L.
Flitsch, Tetrahedron: Asymmetry 1994, 5, 2163–2178.
[11] G. K. Tꢃth, B. Penke, M. Zarꢄndi, K. Kovꢄcs, Int. J.
Peptide Protein Res. 1985, 26, 630–638.
[35]
95% of the product and some p-NP and p-NPS. This organic
solution was washed with 5 mL 5% NaCl solution, the or-
ganic phase removed and evaporated. The product was
washed with ethyl acetate and then solubilised in 3 mL
water. The pH was brought to 11 with a small amount of 1N
NaOH. The solution was extracted again with two times
[12] V. V. Mozhaev, Y. L. Khmelnitsky, F. Sanchez-Riera, J.
Maurina-Brunker, R. A. Rosson, A. D. Grund, Biotech-
nol. Bioeng. 2002, 78, 567–575.
[
[
[
[
[
[
13] C. H. Lin, G. J. Shen, E. Garcia-Junceda, C. H. Wong,
J. Am. Chem. Soc. 1995, 117, 8031–8032.
14] C. Niehrs, W. B. Huttner, D. Carvallo, E. Degryse, J.
Biol. Chem. 1990, 265, 9314–9318.
10 mL n-butanol, which was evaporated to dryness. The
solid product was washed with ethyl acetate and the materi-
al was solubilised in 3 mL water. After concentration to
15] K. Kobashi, Y. Fukaya, D. H. Kim, T. Akao, S. Takebe,
1
mL and cooling a precipitate was obtained which was col-
Arch. Biochem. Biophys. 1986, 245, 537–539.
16] D. H. Kim, L. Konishi, K. Kobashi, Biochim. Biophys.
Acta 1986, 872, 33–41.
lected by centrifugation. After discarding the water solution
the solid product was dried under vacuum.
17] M. C. Baek, S. K. Kim, D. H. Kim, B. K. Kim, E. C.
Choi, Microbiol. Immunol. 1996, 40, 531–537.
18] J. W. Kang, Y. J. Jeong, A. R. Kwon, H. J. Yun, D. H.
Kim, E. C. Choi, Arch. Pharmacal Res. 2001, 24, 316–
Acknowledgements
3
22.
This work was supported in part by the Dutch National Re-
search School Combination (NRSC-Catalysis) and was per-
formed under the ACTS-NWO program Integration of Bio-
synthesis & Organic Synthesis (IBOS). The former compa-
nies Organon and Diosynth which are now MSD (Oss, NL),
and Syncom (Groningen, NL) are kindly acknowledged for
financial support and Organon and Diosynth for the gift of
steroids. We also wish to thank Prof. Dr. D. Kellog (Syncom),
Dr. M. Ostendorp (Diosynth, Organon), Dr. M. Huibers, Dr.
F.L. van Delft (Radboud University, Nijmegen) for valuable
discussions and help.
[
19] A. R. Kwon, T. G. Oh, D. H. Kim, E. C. Choi, Protein
Expression Purif. 1999, 17, 366–372.
[20] K. Kobashi, D. H. Kim, T. Morikawa, J. Protein Chem.
1987, 6, 237–244.
[21] J. W. Kang, A. R. Kwon, D. H. Kim, E. C. Choi, Biol.
Pharm. Bull. 2001, 24, 570–574.
[22] L. Kaysser, K. Eitel, T. Tanino, S. Siebenberg, A. Mat-
suda, S. Ichikawa, B. Gust, J. Biol. Chem. 2010, 285,
12684–12694.
[23] D. H. Kim, K. Kobashi, J. Biochem. 1991, 109, 45–48.
[
24] 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.
References
[
25] N. Tanaka, Z. Hasan, R-J. Sanders, A. F. Hartog, T. van
Herk, R. Wever, Org. Biomol. Chem. 2003, 1 , 2833–
2839.
[
1] E. Chapman, M. D. Best, S. R. Hanson, C. H. Wong,
Angew. Chem. 2004, 116, 3610–3632; Angew. Chem.
Int. Ed. 2004, 43, 3526–3548.
[26] S. de Macedo-Ribeiro, R. Renirie, R. Wever, A. Mes-
[
2] G. Malojcic, R. Glockshuber, Antioxid. Redox Signaling
serschmidt, Biochemistry 2008, 47, 929–934.
2
010, 13, 1247–1259.
[27] J. P. Grimshaw, C. U. Stirnimann, M. S. Brozzo, G. Ma-
lojcic, M. G. Grutter, G. Capitani, R. Glockshuber, J.
Mol. Biol. 2008, 380, 667–680.
[
3] N. Gamage, A. Barnett, N. Hempel, R. G. Duggleby,
K. F. Windmill, J. L. Martin, M. E. McManus, Toxicol.
Sci. 2006, 90, 5–22.
[
28] A. R. Kwon, E. C. Choi, Arch. Pharmacal Res. 2005,
8, 561–565.
[
[
[
[
4] E. G. Cormier, M. Persuh, D. A. D. Thompson, S. W.
Lin, T. P. Sakmar, W. C. Olson, T. Dragic, Proc. Natl.
Acad. Sci. USA 2000, 97, 5762–5767.
2
[
29] T. Teramoto, R. Adachi, Y. Sakakibara, M. C. Liu, M.
Suiko, M. Kimura, Y. Kakuta, FEBS Lett. 2009, 583,
5] M. Farzan, T. Mirzabekov, P. Kolchinsky, R. Wyatt, M.
Cayabyab, N. P. Gerard, C. Gerard, J. Sodroski, H.
Choe, Cell 1999, 96, 667–676.
3
091–3094.
[
[
[
[
30] L. Konishi-Imamura, D. H. Kim, K. Kobashi, Biochem.
Int. 1992, 28, 725–734.
31] S. P. Edgcomb, K. P. Murphy, Proteins Struct. Funct.
6] K. G. Bowman, S. Hemmerich, S. Bhakta, M. S. Singer,
A. Bistrup, S. D. Rosen, C. R. Bertozzi, Chem. Biol.
Genet. 2002, 49, 1–6.
1
998, 5, 447–460.
32] B. Delhom, G. Alvaro, G. Caminal, J. L. Torres, P.
Clapes, Biotechnol. Lett. 1996, 18, 609–614.
33] M. Pogorevc, K. Faber, Tetrahedron: Asymmetry 2002,
7] M. Miksits, K. Wlcek, M. Svoboda, T. Thalhammer, I.
Ellinger, G. Stefanzl, C. N. Falany, T. Szekeres, W.
Jaeger, Cancer Lett. 2010, 289, 237–245.
1
3, 1435–1441.
[
8] L. S. Simpson, T. S. Widlanski, J. Am. Chem. Soc. 2006,
[
[
34] M. Huibers, Dissertation, Univ. Nijmegen, 2009.
35] R. Y. Kirdani, Steroids 1965, 6, 845–853.
1
28, 1605–1610.
9] S. D. Taylor, A. Desoky, Tetrahedron Lett. 2011, 52,
353–3357.
[
3
3508
asc.wiley-vch.de
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2012, 354, 3501 – 3508