α-Hydroxylation of Carboxylic Acids Catalyzed by Taurine Dioxygenase

Article abstract of DOI:10.1002/cctc.201501244

Enzymes still have a limited application scope in synthetic organic chemistry. To expand this, different strategies exist that range from the de novo design of enzymes to the exploitation of the catalytic capabilities of known enzymes by converting different substrates; denoted as substrate promiscuity. We harnessed the synthetic potential offered by the taurine dioxygenase (TauD) from Escherichia coli (E. coli) by studying its promiscuous catalytic properties in the hydroxylation of carboxylic acid substrates. TauD showed high selectivities in the hydroxylation reaction but reduced levels of activity (26 % conversion, >96 % ee). We enhanced the enzyme substrate scope and improved the conversions for the tested substrates by introducing a point mutation at position 206 (F206Y). The conversions of the improved catalyst increased by at least 140 % compared to that of the wild-type enzyme. The number of carboxylic acids that accepted by the enzyme variant doubled from four to eight carboxylic acids.

Full text of DOI:10.1002/cctc.201501244

DOI: 10.1002/cctc.201501244
Full Papers
a-Hydroxylation of Carboxylic Acids Catalyzed by Taurine
Dioxygenase
Dennis Wetzl, Jennifer Bolsinger, Bettina M. Nestl, and Bernhard Hauer*[a]
Enzymes still have a limited application scope in synthetic or-
ganic chemistry. To expand this, different strategies exist that
range from the de novo design of enzymes to the exploitation
of the catalytic capabilities of known enzymes by converting
different substrates; denoted as substrate promiscuity. We har-
nessed the synthetic potential offered by the taurine dioxyge-
nase (TauD) from Escherichia coli (E. coli) by studying its promis-
cuous catalytic properties in the hydroxylation of carboxylic
acid substrates. TauD showed high selectivities in the hydroxyl-
ation reaction but reduced levels of activity (26% conversion,
>96% ee). We enhanced the enzyme substrate scope and im-
proved the conversions for the tested substrates by introduc-
ing a point mutation at position 206 (F206Y). The conversions
of the improved catalyst increased by at least 140% compared
to that of the wild-type enzyme. The number of carboxylic
acids that accepted by the enzyme variant doubled from four
to eight carboxylic acids.
Introduction
The taurine dioxygenase (TauD) from Escherichia coli (E. coli) be-
longs to the class of a-ketoglutarate (a-KG)-dependent en-
zymes that are known to catalyze a plethora of reactions.[1]
Members of this family are involved in DNA repair mecha-
nisms, modifications of amino acids, and the degradation of
xenobiotics.[1] TauD is part of the sulfur-starvation response of
E. coli and, therefore, part of the tauABCD gene cluster that is
involved in the utilization of aliphatic sulfur compounds such
as the natural substrate of TauD, taurine (1).[2,3] The mobiliza-
tion of the sulfonate from the organic compound is performed
by an intermediate hydroxylation reaction of the substrate,
which leads to an unstable a-hydroxy sulfonate compound
(Figure 1) that decomposes spontaneously. This decomposition
liberates the sulfite as the sulfur source. Eichhorn et al. have
shown that the substrate scope of TauD extends from the nat-
ural substrate 1 to longer-chained and sterically more demand-
ing sulfonic acid substrates that include N-morpholine pro-
panesulfonic acid (24).[4] Furthermore, TauD has been described
to accept aliphatic analogues of 1, which include butanesul-
fonic acid (25).[4] The crystal-structure solution and different
mechanistic investigations have helped to clarify the reaction
mechanism of this hydroxylation reaction.[5–9] Important resi-
dues for substrate anchoring and positioning in the active site
were also identified and include the residues H70, F159, F206,
and R270.[5,10,11] These studies have also revealed that the hy-
droxylation of TauD is performed by an intermediate iron(IV)
oxo species similar to that known for heme-type oxygenases.[9]
We speculated that the oxidative potential of TauD could
also be used for the a-hydroxylation of carboxylic acids by the
insertion of oxygen into a CÀH bond adjacent to a carboxylic
acid functional group. This type of reaction has already been
reported for long-chain fatty acids using cytochrome P450
monooxygenases and enzymes involved in the fatty acid me-
tabolism of different plants.[12–16] However, these enzymatic
routes are restricted in their application scope to longer-
chained fatty acid or arylic carboxylic acid substrates.[12–16] Clas-
sical synthetic routes to enantiomerically enriched a-hydroxy
carboxylic acids involve the diazotation of a-amino acids, the
stereoselective reduction of a-keto acids, the addition of ni-
triles to aldehydes with the concomitant hydrolysis of the ni-
trile, or the deracemization of racemic a-hydroxy acids.[17–22] Re-
cently, Kroutil and co-workers established an one-pot bienzy-
matic approach for the production of enantioenriched a-hy-
droxy carboxylic acids with aliphatic or arylic side chains.[18]
Furthermore, the groups of Schofield and Zaparucha have de-
scribed the hydroxylations of amino-substituted carboxylic acid
substrates by using related dioxygenases. Schofield and co-
workers demonstrated that the a-KG-dependent g-butyrobe-
taine hydroxylase (BBOX) could be applied in the desymmetri-
zation of achiral N,N’-dialkyl piperidine-4-carboxylates to the
corresponding hydroxylation products to show the versatility
of these enzymes to introduce chiral centers.[23] The group of
Zaparucha also discovered alternative members of this enzyme
family with altered regioselectivities in a genome-mining ap-
proach.[24] These current findings in combination with our work
indicate the not yet fully explored potential of this class of en-
zymes as catalysts for synthetic tasks.[23,24]
[a] Dr. D. Wetzl, J. Bolsinger, Dr. B. M. Nestl, Prof. Dr. B. Hauer
Institute of Technical Biochemistry
Universitaet Stuttgart
Allmandring 31, 70569 Stuttgart (Germany)
Fax: (+49)711-685-63196
E-mail: bernhard.hauer@itb.uni-stuttgart.de
We studied the TauD-catalyzed enantioselective a-hydroxyl-
ation of short-to-medium-chain-length w-amino carboxylic
Supporting information for this article can be found under http://
dx.doi.org/10.1002/cctc.201501244.
ChemCatChem 2016, 8, 1361 – 1366
1361
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
genase (TfdA) from Ralstonia eutropha that catalyzes the hy-
droxylation of the carboxylic acid substrate 2,4-dichlorophe-
noxyacetic acid, we assumed that TauD could catalyze the a-
hydroxylation of carboxylic acid substrates such as b-alanine
(4; Scheme 1 and Figure 1).[5,28]
Scheme 1. a-Hydroxylations of 1 and 4.
Thus, we performed an initial activity test with TauD and
1 as well as the carboxylic acid substrate 4. We used the assay
developed by Eichhorn et al.[4] to determine a specific activity
of (1.47Æ0.04) UmgÀ1 for 1 for our enzyme preparation,
which is similar to the activity reported in literature.[4] We were
pleased to find that TauD converts 4 to d-isoserine (5) selec-
tively with an overall conversion of (22.6Æ0.7)% (>96% ee,
Table 1). Encouraged by these preliminary findings, we ana-
Table 1. Conversions and selectivities of non-natural substrates 4–22
with wild-type (WT) TauD and mutant F206Y. The ee values are given for
the d enantiomer.
Substrate[a]
Conversion [%]
WT
Enantiomeric excess [%]
F206Y
WT
F206Y
4
6
8
22.6Æ0.7
54.2Æ3.8
37.7Æ4.9
53.5Æ1.1
28.1Æ2.8
5.2Æ0.6
4.1Æ0.3
3.6Æ0.2
–
>96[a]
>96[a]
>96[a]
>96[a]
>96[a]
n.d.
8.1Æ1.3
10Æ1
12.1Æ0.2
>96[a]
10
12
14
16
18
20[b,c]
<3
–
n.d.
–
–
–
–
–
–
–
–
n.d.
n.d.
Figure 1. Substrates and expected products of the TauD-catalyzed a-hydrox-
ylation reaction.
–
7.1Æ0.9
–
>96[b]
[a] The other enantiomer was not detected. [b] Conversion determined
by substrate consumption. No reference material was available for prod-
uct quantification. The stereochemistry of the product could not be con-
firmed. Enantiomeric excesses were calculated from the consumption of
the substrate enantiomers as only one enantiomer was consumed in the
course of the reaction. n.d.: not determined. [c] The proposed position
for hydroxylation of substrate 20 is derived from the position of hydroxyl-
ation of substrates 4–18.
acids. These 2-hydroxy-w-amino carboxylic acids are highly val-
uable products for different glycoside antibiotics, which in-
clude amikacin or arbekacin, and antibiotic peptides such as
edeines.[25–27] We created mutants to elucidate the binding
mode of the new substrates and to improve the catalytic prop-
erties of this promiscuous hydroxylation reaction.
Results and Discussion
lyzed TauD systematically for the conversion of further sub-
strates similar to 4. We tested a set of substrates (Figure 1)
that included longer-chained homologues of 4 such as 4-ami-
nobutyric acid (6) and 5-aminovaleric acid (8). In addition, we
tested valeric acid (12) and homologues thereof as well as a-
methyl-b-alanine (20). These analogues of 4 lack the terminal
amino functionality or possess a substitution on the position
adjacent to the carboxylic group (Figure 1). The substrate
scope of wild-type TauD is distinctly limited to 4 and its higher
homologues that range from w-amino carboxylic acid sub-
Reactivity of TauD from E. coli
Many studies were performed to determine the kinetics and
the mechanism of the TauD from E. coli with its natural sub-
strate 1 and analogues thereof.[3,5–10] However, to date, the cat-
alytic potential of TauD for the introduction of hydroxy func-
tionalities to chemically more distinct but structurally related
carboxylic acid substrates has not been exploited. As the TauD
is closely related to the 2,4-dichlorophenoxyacetic acid dioxy-
ChemCatChem 2016, 8, 1361 – 1366
www.chemcatchem.org
1362
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
strates of chain lengths C3–C6. The results of this screening are
summarized in Table 1. We found that substrates with even-
numbered chain lengths showed significantly lower conver-
sions than the substrates with odd-numbered chain lengths.
Aliphatic carboxylic acids of chain lengths C2–C5 (14–20) and
the substituted b-alanine analogue a-methyl-b-alanine (22) did
not show any detectable conversion. Biotransformations were
further subjected to chiral analyses. The enzyme could convert
nearly all accepted substrates almost exclusively to the d enan-
tiomeric product. Compound 6 was the only substrate to yield
a hydroxylation product with a poor enantiomeric excess of
only 10% ee (Table 1).
probably because of the loss of the ion-pairing ability of the
substrate binding pocket of TauD towards the tested substrate.
The mutation H70A also showed almost no conversion of 4
(conversion 1.9Æ0.2%), whereas the H70S mutant displayed
a dramatically decreased formation of product 5 of (4.8Æ
1.7)%.
Optimization of the substrate binding site for non-natural
substrates
To improve the activity of TauD towards new non-natural sub-
strates, we first aimed to gain a deeper understanding of the
binding mode of these planar carboxylic acid substrates. We
expected them to be similar to those of the natural sulfonate
substrates. Having shown that the model substrate 4 utilizes
the same anchoring mechanism in the active site as the natu-
ral substrate 1, we modeled the docking of 4 to TauD (PDB ID:
1OS7) by using Rosetta Design.
Binding mode of b-alanine to TauD
Previous work by Elkins and co-workers revealed that the resi-
dues that form the substrate binding motif are partially con-
served within a defined group of members of the a-KG-depen-
dent dioxygenases.[5] An arginine at position 270 (R270) found
within different members of this family is considered to be an
ion-pairing residue to the natural substrate. However, the histi-
dine at position 70 (H70) of TauD is only found in family mem-
bers that metabolize sulfonate or sulfate substrates and is,
thus, supposed to be involved in the discrimination of tetrahe-
dral substrates against planar ones through hydrogen
bridges.[5,11]
Residues that surround the catalytic side chains and transi-
tion state were repacked and redesigned to optimize steric,
Coulombic, and hydrogen-bonding interactions with the transi-
tion state and associated catalytic residues by using Rosetta
Design as described previously.[29,30] We intended to analyze
amino acids at the substrate binding site that hinder the bind-
ing of new substrates and to target amino acids that could
possibly be mutated to introduce additional interactions to im-
prove the binding of 4.[31–33] The docking of the model sub-
strate 4 indicated a positioning of the new substrate at the
same substrate binding site as for the natural substrate
1 (Figure 2). The docking also revealed that the carboxylic acid
substrate seems to have a distinct orientation that exposes
one hydrogen atom, namely, the pro-d-hydrogen atom, almost
exclusively towards the reactive Fe center. These results resem-
ble the enantiomeric excess values found experimentally for
the products of the tested conversion for the new substrates
as shown in Table 1.
To study the role of these amino acids in the conversion of
the planar carboxylic acid analogue 4, we replaced R270 and
H70 with alanine or serine amino acid residues. We created
single-point mutants H70A, H70S, R270A, and R270S and the
corresponding double mutants H70A_R270A, H70A_R270S,
H70S_R270A, and H70S_R270S. We hypothesized that these
substitutions would impair hydrogen bond formation to 4.
Based on results from previous work, we expected the mutants
with substitutions at position R270 to show a significantly re-
duced activity towards 4.[5,11] In contrast, the mutations at posi-
tion H70 were expected to have minor effects on the activity.
As expected, mutants R270A and R270S and the double mu-
tants showed no detectable conversion of 4 to 5. This is most
Furthermore, the docking study indicated that residues F159
and F206 in the active site of TauD play a key role in the sub-
strate orientation and positioning to bring the substrate in
Figure 2. Docking of 4 (blue sticks) into the wild-type TauD (on the left) and the mutant F206Y (on the right). The active site Fe is shown as an orange
sphere, and the active site residues F159 and F206/Y206 are depicted as green sticks. Acid-coordinating residues H70 and R270 are shown as orange sticks.
ChemCatChem 2016, 8, 1361 – 1366
www.chemcatchem.org
1363
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
close proximity to the reactive Fe center as already demon-
strated for the natural substrate 1.[10] McKusker and Klinman
showed that the amino acid F159 is involved in the coupling
of the oxygen activation and hydroxylation of 1.[10] The authors
demonstrated that the exchange of F159 with a smaller amino
acid led to a significant uncoupling of the oxygen activation
and the actual hydroxylation of 1 as well as the release of sul-
fite. The exact role of F206 is not known yet. Based on our
docking results with 4 using wild-type TauD, we assumed that
the exchange of F206 by a tyrosine will affect substrate bind-
ing through the introduction of an additional hydrogen bond.
We anticipated an increase of the enzyme affinity and the ac-
tivity towards this new substrate as binding of the substrate
constitutes a bottleneck for hydroxylation. We replaced both
residues F159 and F206 against leucine as well as tyrosine.
Both leucine variants showed no detectable conversion of 4,
which may be because of the unproductive binding of this
substrate in the active site.[34] In the case of the two tyrosine
variants, the mutant F159Y showed a dramatically reduced ac-
tivity with conversions that decreased to 3.9%, whereas the
mutant F206Y displayed an increased conversion of 4 of
ꢀ140% compared to that of the wild-type. Encouraged by
these findings, we re-assayed the initial substrate set
(Figure 2). Additionally, we performed a second docking study
with 4 using TauD variant F206Y (Figure 2, right). All results for
the tested conversions of the improved TauD variant F206Y are
summarized in Table 1. The docking study showed that the hy-
droxyl group of the new Y206 residue lies within a distance of
1.8 Š to the carboxylic acid functionality, which is in the typical
range of a hydrogen bridging interaction. The TauD mutant
F206Y displayed a twofold decreased activity towards 1. Specif-
ic activities of (0.71Æ0.04) UmgÀ1 for the TauD F206Y mutant
and (1.47Æ0.04) UmgÀ1 for the wild-type were determined (lit-
erature data from 1.64 to 14.0Æ1.5 UmgÀ1).[4,11] This indicates
that the introduction of the F206Y mutation enhanced the ac-
tivity towards new substrates and induced selectivity, which
started to switch the overall substrate preference of the
enzyme from 1 to the new substrate 4.
Surprisingly, no conversion of these analogues of 4 under the
tested conditions could be detected (data not shown).
The reason for the discrimination of substrates with odd-
numbered chain lengths versus those with even-numbered
chain lengths still remains unclear. Additionally, the enzyme se-
lectivity for substrates that possess an amino substituent distal
to the carboxylic acid is not yet fully understood. We assume
that the protonated amine group facilitates the binding of the
substrates in the active site, whereas substrates that lack this
feature are, therefore, disfavored. This hypothesis is underlined
by significant differences in the Michaelis constants (KM) de-
fined experimentally between aliphatic and amino-substituted
sulfonic acid substrates. The KM values of the amino-substitut-
ed substrates are one order of magnitude lower than those of
the aliphatic substrates.[4] However, further experiments are
needed to clarify the exact role of substrate substituents on
their acceptance. The fact that the enzyme does not accept
substrates that bear an amide instead of a carboxylic acid can
be explained by the acid-coordinating residues H70 and R270
in the active site of TauD. This might lead to a repulsion of the
amide group and, therefore, a nonproductive positioning in
the active site.
Conclusions
We studied the promiscuous activity of taurine dioxygenase
(TauD) from Escherichia coli in the hydroxylation of w-amino
carboxylic acids. We demonstrated that the native reaction
mechanism of the enzyme can be used for the direct a-hy-
droxylation of b-alanine (4) and its higher homologues. We fur-
ther improved the catalytic activity of TauD by the introduction
of rational mutations in the active site. The TauD variant F206Y
showed improved catalytic activities as well as a broadened
substrate scope.
Experimental Section
General remarks: All commercially available chemicals used in this
study were purchased from Sigma Aldrich or Alfa Aesar at the
highest purity level available and used without any further purifica-
tion. Restriction enzymes were purchased from Fermentas, and the
components used for culture media from Carl Roth company.
We demonstrated that our improved TauD variant F206Y in-
creased the conversion of the model substrate 4 and improved
the activity towards all of our substrates tested.
The increase in the conversion ranged from 140% for 4 to
up to >800% for 10 relative to that of the wild type. The
F206Y variant also displayed improved stereoselectivities for
substrates that had formerly showed not very pronounced
enantiomeric excesses in biotransformations with wild-type
enzyme. Remarkably, mutant F206Y not only displayed im-
proved catalytic properties towards the substrates already con-
verted by the wild-type enzyme but also displayed a wider
substrate scope as it accepted homologues of the model sub-
strate 4 as well as aliphatic carboxylic acids that were not ac-
cepted by TauD wild-type. Conversions of the aliphatic carbox-
ylic acid substrates 12, 14, and 16 were between (5.2Æ0.6)
and (3.6Æ0.2)%. In the second screening, we tested further
analogues of 4, namely, 3-hydroxypropionic acid and 3-amino-
propionic amide in which the amine functionality was replaced
by a hydroxyl group and the acid by an amide, respectively.
DNA modification and sequencing: Wild-type TauD was amplified
from genomic DNA from E. coli JM109 by a standard polymerase
chain reaction (PCR) technique using the following primers (restric-
tion sites are highlighted in bold) TauD_f: GGGATTTCCATATGAGT-
GAACGTCTGAGCATTACC with the NdeI restriction site and TauD_r:
CGGAATTCTTACCCCGCCCG ATAAAACGG with the EcoRI restriction
site. The resulting PCR fragment was purified by agarose gel elec-
trophoresis, double digested using the corresponding restriction
enzymes, and subsequently cloned into a digested pET22b(+)
vector system using T4 ligase (pITB_1022). The stop codon of the
TauD was removed using site-directed mutagenesis to result in
a C-terminally His-tagged TauD construct (pITB_988) that was used
throughout this study.
Site-directed mutagenesis: Mutants were created by using the
standard QuikChangeꢁ PCR technique. The following primers were
ChemCatChem 2016, 8, 1361 – 1366
www.chemcatchem.org
1364
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
used for the construction of the mutants mentioned in this work
with modified sequences highlighted in bold italic letters. The for-
ward primers are always given, and the reverse primers were con-
structed by inverting the sequence. The resulting plasmid con-
structs are given in brackets behind the primer sequence. TauD-
solutions were prepared fresh weekly. All reagents besides the
iron(II) solution were adjusted to pH 7 before use. The reaction
mixture was incubated at 308C for 18 h by using an Eppendorf
ThermoMixer at a shaking speed of 850 rpm.
His6:
CCGTTTTATCGGGCGGGGAATTCGAGCTCCGTCGACAAG
Synthesis of reference substances for HPLC analysis: Substrate
13 was not commercially available and was, therefore, synthesized
as a reference for analysis according to Ref. [35]. The purification
procedure of the product, however, was altered from that de-
scribed in the literature. After the reaction, the pH of the reaction
mixture was adjusted to 1 by adding concentrated hydrochloric
acid, and the mixture was extracted by adding a Dowex SCX
strong cation exchanger. The Dowex pearls were then collected by
filtration, and the bound products were eluted by adding aqueous
NH3 (25%). The aqueous phase was evaporated to dryness, and
the resulting oil was re-dissolved in water. In the next step, the re-
sulting aqueous product mixture was separated by using HPLC
and a Reprosil Chiral AA column (250”4.6 mm; 8 mm; Dr. Maisch
GmbH, Ammerbuch-Entringen, Germany) and the method de-
scribed below for the analysis of amino carboxylic substrates.
(pITB 988); F159 L: GCGGAGCATGATTTCCGTAAATCGTTGCCGGAA-
TACAAATACCGCAAAAC (pITB_998); F159Y: GCGGAGCATGATTTCCG-
TAAATCGTACCCGGAATACAAATACCGCAAAAC (pITB_997); F206L:
CAGGCGCTGTTTGTGAATGAAGGCTTGACTACGCGAATTGTTGATG
(pITB_1000);
F206Y:
CAGGCGCTGTTTGTGAATGAAGGCTACAC-
TACGCGAATTGTTGATG (pITB_999); H70 A: CCCAGCGTTTTGGC-
GAATTGCATATTGCGCCTGTTTACCCGCATGCC (pITB_989); H70S:
CCCAGCGTTTTGGCGAATTGCATATTTCCCCTGTTTACCCGCATGCC
(pITB_990); R270 A: GCCACAGCGACGGATAATGCATGCGGCGAC-
GATCCTTGGGG (pITB_991); R270S: GCCACAGCGACGGATAATG-
CATTCCGCGACGATCCTTGGGG (pITB_992).
Protein expression and purification: Expression and purification
of the His6-tagged wild-type TauD and the corresponding TauD
mutants were performed in E. coli BL21(DE3).Cells were grown at
378C in LB-medium (low salt) until the OD600 reached 0.4–0.6 and
isopropyl b-thiogalactoside (IPTG) was added to a final concentra-
tion of 0.5 mm. After induction, the cells were grown at 308C for
another 5 h and harvested by centrifugation. Cells were resuspend-
ed in 50 mm BisTris buffer pH 7 (5 mL per g cell wet weight) and
disrupted by sonification (Branson sonifier W250, Danbury, USA).
The resulting solution was clarified by centrifugation to remove
cell debris. The supernatant was filtered through a 0.2 mm syringe
filter. In the next step, the protein was purified by using an ¾KTA-
purifier system (GE Healthcare, Chalfont St. Giles, UK) through
a 5 mL HiTrap Chelating HP column (GE Healthcare, Chalfont St.
Giles, UK). After the sample was applied, the column was washed
with two column volumes of binding buffer (20 mm phosphate
buffer pH 7, 150 mm NaCl, and 20 mm imidazole). In the second
step a linear gradient over eight column volumes was applied and
the buffer concentration was varied from 0 to 80% elution buffer
(20 mm phosphate buffer pH 7, 150 mm NaCl, and 500 mm imida-
zole) to elute the target protein. The protein preparation was
>90% pure as judged by using sodium dodecyl sulfate polyacryl-
amide gel electrophoresis (SDS-PAGE). After the elution of the pro-
tein, the concentration of the elution buffer was increased to
100% for two column volumes to wash the column. After purifica-
tion, the resulting enzyme preparation was dialyzed twice against
50 mm BisTris buffer pH 7 to remove salts from the purification
and then concentrated with VivaSpin concentrator columns until
the targeted protein concentration was reached. The protein con-
centration was determined by using a Pierce BCA protein kit
(Thermo Scientific) following the manufacturer’s instructions. The
activity of the enzyme preparations were determined as described
elsewhere.[4]
Chiral HPLC analysis of w-amino carboxylic substrates: The anal-
ysis of 4 and its higher homologues was performed by using an
Agilent HPLC system (1200 series) equipped with an Agilent Infini-
ty 1260 evaporative light scattering detector (ELSD) and Reprosil
Chiral AA column (250”4.6 mm; 8 mm; Dr. Maisch GmbH, Ammer-
buch-Entringen, Germany). The mobile phase was an isocratic mix-
ture of methanol and water (50:50), the column temperature was
set to 308C, and the flow was set to 0.8 mLminÀ1. The retention
times of the reagents were proven by the injection of standard ref-
erences.
HPLC analysis of carboxylic substrates: The analysis of the ali-
phatic carboxylic acids such as acetic acid and its higher homo-
logues was performed by using an Agilent HPLC system (1200
series) equipped with an Agilent Infinity 1200 refractive index de-
tector (RID) and an Aminex HPX-87H column (7.8”300 mm; 9 mm;
BioRad, Hercules, CA). The mobile phase was 5 mm aqueous sulfu-
ric acid, the column temperature was set to 508C, and the flow
was set to 0.5 mLminÀ1
.
Acknowledgements
The authors thank Dr. Rainer Stürmer from Stürmer Scientific as
well as Dr. Alexander Seiffert for thoughtful comments and Dr.
Juliane Stahmer for help with the docking. Funding by the Euro-
pean Union’s Seventh Framework Programme FP7/2007–2013
under grant agreement no. 266025 is gratefully acknowledged.
Keywords: biocatalysis · biotransformations · carboxylic acids ·
enzymes · hydroxylation
Biotransformations of alternative substrates: All biotransforma-
tions were performed in 1.5 mL Eppendorf tubes. For the biotrans-
formations, a reaction setup adapted from the standard reaction
setup reported by Eichhorn et al. for taurine was used.[4] The reac-
tion mixture contained the corresponding purified protein in
a final concentration of 1.2 mgmLÀ1 and 10 mm BisTris buffer pH 7.
The target substrate and the cosubstrate a-ketoglutarate were
both added in a final concentration of 10 mm. Ascorbic acid was
added to the reaction mixture at a final concentration of 0.4 mm.
The reaction was started by adding iron(II) sulfate to a final con-
centration of 0.1 mm. The iron solution was prepared freshly
before addition to prevent the iron from oxidation. Ascorbic acid
[1] R. P. Hausinger, Crit. Rev. Biochem. Mol. Biol. 2004, 39, 21–68.
[2] J. R. van der Ploeg, M. A. Weiss, E. Saller, H. Nashimoto, N. Saito, M. A.
Kertesz, T. Leisinger, J. Bacteriol. 1996, 178, 5438–5446.
[3] M. R. Uria-Nickelsen, E. R. Leadbetter, W. G. Iii, Arch. Microbiol. 1994, 161,
434–438.
[4] E. Eichhorn, J. R. van der Ploeg, M. A. Kertesz, T. Leisinger, J. Biol. Chem.
1997, 272, 23031–23036.
[5] J. M. Elkins, M. J. Ryle, I. J. Clifton, J. C. D. Hotopp, J. S. Lloyd, N. I. Bur-
zlaff, J. E. Baldwin, R. P. Hausinger, P. L. Roach, Biochemistry 2002, 41,
5185–5192.
ChemCatChem 2016, 8, 1361 – 1366
www.chemcatchem.org
1365
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[6] P. K. Grzyska, E. H. Appelman, R. P. Hausinger, D. A. Proshlyakov, Proc.
Natl. Acad. Sci. USA 2010, 107, 3982–3987.
[22] P. Brewster, E. D. Hughes, C. K. Ingold, P. A. D. S. Rao, Nature 1950, 166,
179–180.
[23] A. M. Rydzik, I. K. H. Leung, G. T. Kochan, M. A. McDonough, T. D. W. Clar-
idge, C. J. Schofield, Angew. Chem. 2014, 126, 11105–11107.
[24] D. Baud, P.-L. Saaidi, A. Monfleur, M. Harari, J. Cuccaro, A. Fossey, M. Bes-
nard, A. Debard, A. Mariage, V. Pellouin, J.-L. Petit, M. Salanoubat, J.
Weissenbach, V. de Berardinis, A. Zaparucha, ChemCatChem 2014, 6,
3012–3017.
[7] P. K. Grzyska, M. J. Ryle, G. R. Monterosso, J. Liu, D. P. Ballou, R. P. Hau-
singer, Biochemistry 2005, 44, 3845–3855.
[8] J. C. Price, E. W. Barr, T. E. Glass, C. Krebs, J. M. Bollinger, J. Am. Chem.
Soc. 2003, 125, 13008–13009.
[9] J. C. Price, E. W. Barr, B. Tirupati, J. M. Bollinger, C. Krebs, Biochemistry
2003, 42, 7497–7508.
[25] G. Dinos, D. N. Wilson, Y. Teraoka, W. Szaflarski, P. Fucini, D. Kalpaxis,
K. H. Nierhaus, Mol. Cell 2004, 13, 113–124.
[10] K. P. McCusker, J. P. Klinman, J. Am. Chem. Soc. 2010, 132, 5114–5120.
[11] J. R. O’Brien, D. J. Schuller, V. S. Yang, B. D. Dillard, W. N. Lanzilotta, Bio-
chemistry 2003, 42, 5547–5554.
[12] M. Girhard, S. Schuster, M. Dietrich, P. Dürre, V. B. Urlacher, Biochem. Bio-
phys. Res. Commun. 2007, 362, 114–119.
[13] M. Hamberg, I. Ponce de Leon, M. J. Rodriguez, C. Castresana, Biochem.
Biophys. Res. Commun. 2005, 338, 169–174.
[14] M. Hamberg, I. Ponce de Leon, A. Sanz, C. Castresana, Prostaglandins
Other Lipid Mediators 2002, 68–69, 363–374.
[26] M. K. Kharel, B. Subba, D. B. Basnet, J. S. Woo, H. C. Lee, K. Liou, J. K.
Sohng, Arch. Biochem. Biophys. 2004, 429, 204–214.
[27] A. Czerwinski, H. Wojciechowska, A. R. Andruszkiewicz, J. Grzybowska, J.
Gumeniak, E. Borowski, J. Antibiot. 1983, 36, 1001–1006.
[28] J. C. Dunning Hotopp, R. P. Hausinger, Biochemistry 2002, 41, 9787–
9794.
[29] S. A. Combs, S. L. DeLuca, S. H. DeLuca, G. H. Lemmon, D. P. Nanne-
mann, E. D. Nguyen, J. R. Willis, J. H. Sheenan, J. Meiler, Nat. Protoc.
2013, 8, 1277–1298.
[15] M. Landwehr, L. Hochrein, C. R. Otey, A. Kasrayan, J.-E. Bäckvall, F. H.
Arnold, J. Am. Chem. Soc. 2006, 128, 6058–6059.
[30] J. Meiler, D. Baker, Proteins Struct. Funct. Bioinf. 2006, 65, 538–548.
[31] J. Kirchmair, G. Wolber, C. Laggner, T. Langer, J. Chem. Inf. Model. 2006,
46, 1848–1861.
[16] W. Adam, W. Boland, J. Hartmann-Schreier, H. Humpf, M. Lazarus, A. Saf-
fert, C. R. Saha-Mçller, R. V. April, J. Am. Chem. Soc. 1998, 120, 11044–
11048.
[32] J. Bostrçm, J. R. Greenwood, J. Gottfries, J. Mol. Graphics Modell. 2003,
21, 449–462.
[17] S. Lüttenberg, T. D. Ta, J. von der Heyden, J. Scherkenbeck, Eur. J. Org.
Chem. 2013, 1824–1830.
[33] D. Kutlµn, P. Presits, I. Molnµr-Perl, J. Chromatogr. A 2002, 949, 235–248.
[34] K. P. McCusker, J. P. Klinman, Proc. Natl. Acad. Sci. USA 2009, 106,
19791–19795.
[18] E. Busto, N. Richter, B. Grischek, W. Kroutil, Chem. Eur. J. 2014, 20,
11225–11228.
[19] W. Adam, M. Lazarus, C. R. Saha-Moller, P. Schreier, Acc. Chem. Res. 1999,
32, 837–845.
[35] J.-P. Gouesnard, Bull. Soc. Chim. Fr. 1989, 88–94.
[20] Y.-P. Xue, Y.-G. Zheng, Y.-Q. Zhang, J.-L. Sun, Z.-Q. Liu, Y.-C. Shen, Chem.
Commun. 2013, 49, 10706–10708.
[21] K. Faber, Biotransformations in Organic Chemistry, Springer Verlag, Berlin,
2011.
Received: November 12, 2015
Revised: January 18, 2016
Published online on February 23, 2016
ChemCatChem 2016, 8, 1361 – 1366
www.chemcatchem.org
1366
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim