Asymmetric enzymatic hydration of hydroxystyrene derivatives

Article abstract of DOI:10.1002/anie.201207916

More than one activity: Owing to their hydratase activity, phenolic acid decarboxylases catalyze the regio- and stereoselective addition of H ₂O across the C=C double bond of hydroxystyrene derivatives yielding (S)-4-(1-hydroxyethyl)phenols with up to 82 % conversion and 71 % ee. Based on structure analysis and molecular docking simulations, a catalytic mechanism for this novel enzymatic reaction is proposed. Copyright

Full text of DOI:10.1002/anie.201207916

Angewandte
Chemie
DOI: 10.1002/anie.201207916
Enzyme Catalysis
Asymmetric Enzymatic Hydration of Hydroxystyrene Derivatives**
Christiane Wuensch, Johannes Gross, Georg Steinkellner, Karl Gruber, Silvia M. Glueck,* and
Kurt Faber*
=
The stereoselective addition of water across C C bonds
Asymmetric addition of water across isolated or conju-
transforms prochiral alkenes to nonracemic alcohols and
represents a major challenge in synthetic organic chemistry.
In general, alkene hydration is an equilibrium reaction
slightly favoring the alcohol side in 1,4-additions and some-
what disfavored on isolated C C bonds. Acid-catalyzed
gated C C bonds is an important process in biology,[1] which is
=
catalyzed by lyases (termed “hydro-lyases” or “hydratases”).
Mechanistically, these enzymes can be divided into two
categories: 1) acting through (Lewis) acid-catalyzed 1,2-
addition and 2) acting through (Michael-type) nucleophilic
1,4- or 1,6-addition involving quinone methide enolates.[6]
Among group (1), acetylene hydratase (AH) is a rare
tungsten-dependent protein that catalyzes the hydration of
acetylene to furnish acetaldehyde.[7] Likewise, several
[1]
=
alkene hydration, which follows the rule of Markovnikov,
usually proceeds with low regioselectivity and is often
accompanied by rearrangement yielding regioisomeric prod-
uct mixtures; with a few exceptions,[2] no generally applicable
protocol has been developed so far. Likewise, base-catalyzed
1,4-addition of water to a,b-unsaturated (Michael) acceptors
is impeded by the poor nucleophilicity of water.[3] Overall, an
astonishingly limited number of asymmetric alkene-hydration
protocols are reported: 1) The stereoselective hydration of
a,b-unsaturated carboxylic acids by using a heterobimetallic
chiral biopolymer (wool–PdII–CoII) catalyst furnished b-
hydroxy carboxylic acids in high optical purities,[4] and
2) the asymmetric syn-hydration of a,b-unsaturated acyl
imidazoles while applying a DNA-based CuII catalyst yielded
b-hydroxy carbonyl compounds with moderate ee values.[5] To
compensate for the insufficient nucleophilicity of water,
indirect methods using strong nucleophiles (alkoxides, N-
silyloxycarbamates, oximes, silicon and boron reagents) have
been employed, which require cumbersome reductive or
oxidative follow-up chemistry to yield the desired b-hydroxy
carbonyl compounds.[3]
=
enzymes catalyze the hydration of nonconjugated C C
bonds in fatty acids (for example, oleate hydratase),[8] natural
products (kievitone and phaseollidin hydratase,[9] carotenoid
1,2-hydratase),[10] and terpenoids (linalool dehydratase-iso-
merase).[11] Unfortunately, the high substrate specificity of
these enzymes severely limits their practical applicability.
In contrast, enzymes of group (2) appear to be more
flexible: Fumarase is industrially applied for the anti-hydra-
tion of fumarate yielding (S)-malate (ca. 2000 t/a).[12] How-
ever, fumarase and its relatives malease, citraconase, and
mesaconate hydratase exhibit a very narrow substrate spec-
trum.[13] Enoyl-CoA hydratases are key enzymes in the b-
oxidation pathway and require the (ATP-consuming) activa-
tion of their monoacid substrates through formation of
a thioester bond to the cofactor coenzyme A (CoA).[14]
Despite their complexity, a few processes operate on indus-
trial scale.[12] Owing to the dependence on ATP, whole cells
are employed. In a related fashion, hydroxycinnamoyl-CoA
hydratase-lyase (HCHL) requires ATP-dependent substrate
activation with CoA and catalyzes the two-step degradation
[*] C. Wuensch, J. Gross, Dr. S. M. Glueck
ACIB GmbH c/o Department of Chemistry
Organic & Bioorganic Chemistry, University of Graz
Heinrichstrasse 28, 8010 Graz (Austria)
E-mail: si.glueck@uni-graz.at
=
of feruloyl-CoA through stereospecific hydration of the C C
bond, followed by retroaldol C–C-cleavage to yield vanillin
and acetaldehyde.[6,15] Recently, Michael hydratase alcohol
dehydrogenase (MhyADH)[1] was shown to be a bifunctional
enzyme requiring Mo, Fe, and Zn as cofactors.[16] MhyADH
catalyzes the 1,4-hydration of a range of a,b-unsaturated
carbonyl compounds followed by oxidation of the b-hydroxy
moiety to yield the corresponding b-oxo-aldehydes or
-ketones in the presence of an oxidant; in the absence of an
electron acceptor, the hydration product could be identi-
fied.[17] In view of the general narrow substrate tolerance of
hydratases, MhyADH is exceptional owing to its broad
substrate spectrum.
Dr. G. Steinkellner
ACIB GmbH c/o Institute of Molecular Biosciences
University of Graz, Humboldtstrasse 50, 8010 Graz (Austria)
Prof. K. Gruber
Institute of Molecular Biosciences
University of Graz, Humboldtstrasse 50, 8010 Graz (Austria)
Prof. K. Faber
Department of Chemistry, Organic & Bioorganic Chemistry
University of Graz, Heinrichstrasse 28, 8010 Graz (Austria)
E-mail: Kurt.Faber@Uni-Graz.at
[**] This work has been supported by the Austrian BMWFJ, BMVIT, SFG,
Standortagentur Tirol, and ZIT through the Austrian FFG-COMET-
Funding Program. Byung-Gee Kim (School of Chemical and
Biological Engineering, Seoul National University, Seoul, South
Korea) is cordially thanked for the generous donation of phenolic
acid decarboxylase plasmids.
Herein we describe the unprecedented stereoselective
asymmetric hydration of hydroxystyrene-type substrates by
employing phenolic acid decarboxylases (PADs; Table 1).
The promiscuous[18] catalytic “hydratase activity” of these
enzymes was discovered during studies on the regioselective
b-carboxylation of p-vinylphenol,[19] which unexpectedly
furnished (S)-1-(p-hydroxyphenyl)ethanol derived through
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201207916.
Angew. Chem. Int. Ed. 2013, 52, 2293 –2297
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2293

.
Angewandte
Communications
Table 1: Conversions and stereoselectivities of the asymmetric enzymatic hydration of hydroxystyrenes 1a–6a.[a]
Entry Substrate
Product
PAD_Lp
PAD_Ba
PAD_Mc PAD_Ms PAD_Ps
PAD_Ll
FDC_Es
Control[b]
conv. [%] conv. [%] conv. [%] conv. [%] conv. [%] conv. [%] conv. [%] conv. [%]
(ee [%])
(ee [%])
(ee [%])
(ee [%])
(ee [%])
(ee [%])
(ee [%])
(ee [%])
82
(43 (S))
64
(53 (S))
2
77[c]
(40 (S))
75[c]
(36 (S))
73[c]
(25 (S))
77[c]
(41 (S))
6
(rac)
1
(n.d.)
1a
1b
2b
3b
4b
5b
<1
(–)
30
(50 (S))
<1
(–)
24[c]
(28 (S))
39[c]
(12 (S))
45[c]
(28 (S))
27[c]
(8 (S))
14
(rac)
2
2a
73
(28 (S))
73
(36 (S))
<1
(–)
63
(27 (S))
76
(10 (S))
73
(20 (S))
74[c]
(8 (S))
17
(rac)
3
3a
46
(29 (S))
46
(44 (S))
<1
(–)
45[c]
(37 (S))
44[c]
(32 (S))
47[c]
(18 (S))
34[c]
(71 (S))
39
(rac)
4
4a
8
3
<1
(–)
5
12
35
17
10
(rac)
5
(n.d.)
(n.d.)
(n.d.)
(3)[d]
(8)[d]
(10)[d]
5a
<1
(–)
<1
(–)
<1
(–)
<1
(–)
<1
(–)
<1
(–)
<1
(–)
<1
(–)
6
6a
6b
[a] All conversion data in Table 1 are corrected for the control in the absence of biocatalyst; reaction conditions: whole lyophilized cells containing the
overexpressed enzyme (30 mg), substrate (10 mm), KHCO3 (3m), phosphate buffer (pH 8.5, 100 mm), 308C, 120 rpm (orbital shaker), 24 h;
n.d.=not determined owing to low conversion; phenolic acid decarboxylases from Lactobacillus plantarum (PAD_Lp), Bacillus amyloliquefaciens
(PAD_Ba), Mycobacterium colombiense (PAD_Mc), Methylobacterium sp. (PAD_Ms), Pantoea sp. (PAD_Ps), Lactoccocus lactis (PAD_Ll), and ferulic
acid decarboxylase from Enterobacter sp. (FDC_Es) were used; [b] spontaneous (nonenzymatic) hydration in the absence of biocatalyst; [c] the
corresponding (substituted) benzaldehyde was detected as minor side product; [d] absolute configuration not determined owing to low ee.
=
enzymatic hydration of the C C bond, next to the expected
(PAD_Mc), Methylobacterium sp. (PAD_Ms), Pantoea sp.
(PAD_Ps), Lactoccocus lactis (PAD_Ll), ferulic acid decar-
boxylase from Enterobacter sp. (FDC_Es), and (for reason of
comparison) p-hydroxycinnamoyl CoA hydratase-lyase from
Pseudomonas fluorescens (HCHL). All genes were synthe-
sized by Geneart AG (Regensburg) and subcloned in a pET
vector (pET 21a or pET 28a). A standard E. coli BL21(DE3)
host was transformed with the obtained plasmids and IPTG
was used to induce overexpression. According to SDS-PAGE
analysis, successful overexpression was obtained for all
enzymes, which were employed as lyophilized whole-cell
biocatalyst to a range of substrates. The absence of competing
hydratase activity of empty E. coli host cells was verified by
separate control experiments. To our delight, almost all
decarboxylases were able to catalyze the hydration of
hydroxystyrene-type substrates 1a–5a except for PAD_Mc
from Mycobacterium colombiense (Table 1). The broad range
of hydratase activities was even more remarkable, since the
carboxylation product p-coumaric acid.
Detailed analysis revealed that recombinant phenolic acid
decarboxylases from Lactobacillus plantarum (PAD_Lp) and
from Bacillus amyloliquefaciens (PAD_Ba) were able to
=
hydrate the C C bond of p-vinylphenol (1a), thereby forming
(S)-4-(1-hydroxyethyl)phenol (1b) with good conversions and
remarkable stereoselectivities (PAD_Lp: conv. 82%, ee 43%;
PAD_Ba: conv. 64%, ee 53%, Table 1, entry 1) in the
presence of carbonate buffer (3m, pH 8.5) used as CO2
source for carboxylation. Interestingly, the expected carbox-
ylation product (p-coumaric acid) was observed only as minor
[
]
*
side product (ꢀ 5%).
To exploit the potential of this
method, the following enzymes were chosen based on
a similarity search (40–80% sequence identity): phenolic
acid decarboxylases from Mycobacterium colombiense
[*] For enzymes, which showed very low hydration activity, such as
PAD_Mc, the carboxylation reaction was dominant.
2294 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 2293 –2297

Angewandte
Chemie
ꢀrealꢁ hydratase, hydroxycinnamoyl-CoA hydratase-lyase did
not show any hydratase activity at all.
literature studies revealed puzzling similarities between the
(proposed) mechanism of p-coumaric acid decarboxylase
from Lactobacillus plantarum,[20] ferulic acid decarboxylase
from Enterobacter sp.,[21] and hydroxycinnamoyl-CoA hydra-
tase-lyase from Pseudomonas fluorescens:[1,6,21] All enzymes
All active enzymes displayed similar levels of conversion
with respect to a given substrate (conv. 64–82% for 1a, conv.
24–45% for 2a, conv. 63–76% for 3a, conv. 34–47% for 4a,
conv. 3–35% for 5a, conv. < 1% for 6a). The hydroxy group
in the para-position appears to be mandatory for enzyme
activity, meta-analogues were unreactive, and ortho-substi-
tuted substrates turned out to be unstable. Overall, the data
indicate that steric effects play a major role, because p-
vinylphenol (1a) and the chloro analogue 3a gave the highest
conversions (conv. 63–82%, Table 1, entries 1,3), while the
conversion continuously dropped upon increasing size and/or
number of substituents. An additional methyl or methoxy
group (2a, 4a) led to moderate conversions (conv. 24–47%,
Table 1, entries 2,4), and more bulky substituents like on 5a
resulted in a significant decrease of conversion (Table 1,
entry 5). Compound 6a, carrying three substituents on the
phenyl ring, was not accepted (conv. < 1, Table 1, entry 6).
Comparison of results obtained with substrates bearing alkyl,
alkoxy, and chloro groups shows that electronic effects play
a minor role.
[
]
*
act via a central quinone methide intermediate, which
represents a Michael-acceptor for the nucleophilic 1,6-addi-
tion of water. Based on literature[1,6,15a,20,21] and experimental
results, in silico docking and energy minimization studies
(based on the crystal structure of ferulic acid decarboxylase
from E. sp.) were performed,[21] which result in the mecha-
nistic proposal depicted in Scheme 1.
To elucidate the effect of bicarbonate, a set of experiments
with various buffer systems were performed using p-vinyl-
phenol (1a) and whole cells of E. coli harboring PAD_Lp
(Figure 1). The data clearly show that the hydration indeed
&
Figure 1. Hydration (resulting in 1b; c c) versus carboxylation
~
(resulting in p-coumaric acid; c c) of p-vinylphenol (1a)
Scheme 1. Proposed catalytic mechanism for hydration of substrate 2a
supported by docking experiments based on the crystal structure of
ferulic acid decarboxylase from Enterobacter sp.[21] (see Figure 2, PDB-
Code: 3NX2). WAT1=water molecule 1, see Figure 2.
employing PAD_Lp at various concentrations of bicarbonate. For 1b
the ee values are plotted (b b). The control (a a) corre-
sponds to spontaneous (nonenzymatic) hydration in the absence of
^
*
biocatalyst.
depends on the bicarbonate concentration, and the conver-
sion reached its maximum at approximately 3m. Moreover,
the stereoselectivity of hydration (expressed as ee of 1b) also
strongly increased from near racemic (in the absence of
bicarbonate) to reach a maximum (ca. 62% ee) at a bicar-
bonate concentration of approximately 0.5m, followed by
a continuous decrease to a plateau of approximately 43% ee.
In contrast, the carboxylation product (p-coumaric acid) was
observed only as minor side product (ꢀ 5%) independent on
the bicarbonate concentration. These results strongly suggest
that bicarbonate is mechanistically involved in the enzymatic
hydration.
In contrast to the mechanism suggested by Gu et al.[21] we
recognized that the active-site pocket provides productive
substrate binding through formation of hydrogen bonds to
Glu72 and Arg49, supported by a hydrophobic environment
for the aromatic system (Ile29, Ile41, Trp70, Val78, Leu80, and
Ile93). Tyr19 lies within an appropriate distance for the
deprotonation of the p-hydroxy group of the substrate
(supported by Tyr21, Figure 2), thereby initiating an electron
flow along the p-hydroxy styrene structure. The shift of
electrons enhances the nucleophilicity of the Cb atom, which
[*] Also reported for vanillyl-alcohol oxidase[22] and 4-ethylphenol
methylenehydroxylase.[23]
To provide a rationale for the promiscuous catalytic
hydratase activity of phenolic acid decarboxylases, detailed
Angew. Chem. Int. Ed. 2013, 52, 2293 –2297
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2295

.
Angewandte
Communications
Received: October 1, 2012
Revised: December 10, 2012
Published online: January 17, 2013
Keywords: biotransformations · catalytic promiscuity ·
.
decarboxylases · enzyme catalysis · hydratases
[1] J. Jin, U. Hanefeld, Chem. Commun. 2011, 47, 2502 – 2510.
[2] H.-D. Hahn, G. Dꢂmbkes, N. Rupprich, H. Bahl, Butanols in
Ullmannꢀs Encyclopedia of Industrial Chemistry, Electronic
version, Wiley-VCH, Weinheim, 2010.
[3] E. Hartmann, D. J. Vyas, M. Oesterreich, Chem. Commun. 2011,
47, 7917 – 7932.
[4] S.-Q. Wang, Z.-W. Wang, L.-C. Yang, J.-I. Dong, C.-Q. Chi, D.-N.
Sui, Y.-Z. Wang, J.-G. Ren, M.-Y. Hung, Y.-Y. Jiang, J. Mol.
Catal. A 2007, 264, 60 – 65.
[5] A. J. Boersma, D. Coquiere, D. Geerdink, F. Rosati, B. L.
Feringa, G. Roelfes, Nat. Chem. 2010, 2, 991 – 995.
[6] J. P. Bennett, L. Bertin, B. Moulton, I. J. S. Fairlamb, A. M.
Brzozowski, N. J. Walton, G. Grogan, Biochem. J. 2008, 414,
281 – 289.
[7] a) F. tenBrink, B. Schink, P. M. H. Kroneck, J. Bacteriol. 2011,
193, 1229 – 1236; b) R.-Z. Liao, F. Himo, ACS Catal. 2011, 1,
937 – 944; c) R.-Z. Liao, J.-G. Yu, F. Himo, Proc. Natl. Acad. Sci.
USA 2010, 107, 22523 – 22527; d) G. B. Seiffert, G. M. Ullmann,
A. Messerschmidt, B. Schink, P. M. H. Kroneck, O. Einsle, Proc.
Natl. Acad. Sci. USA 2007, 104, 3073 – 3077.
[8] a) L. E. Bevers, M. W. H. Pinkse, P. D. E. M. Verhaert, W. R.
Hagen, J. Bacteriol. 2009, 191, 5010 – 5012; b) A. Volkov, A.
Liavonchanka, O. Kamneva, T. Fiedler, C. Goebel, B. Kreike-
meyer, I. Feussner, J. Biol. Chem. 2010, 285, 10353 – 10361.
[9] C. S. Turbek, D. A. Simth, C. L. Schardl, FEMS Microbiol. Lett.
1992, 94, 187 – 190.
Figure 2. Energy-minimized oxyanion quinone methide intermediate of
substrate 2a (blue) in the active site of ferulic acid decarboxylase from
Enterobacter. sp. (PDB-Code: 3NX2).[21] The active-site surface is
colored by hydrophobicity from red (hydrophobic) to blue (hydro-
philic). Coordinated water molecules (WAT) are shown as balls; WAT1
is ideally positioned to launch a nucleophilic front-side attack resulting
in the formation of (S)-alcohol 2b (see Scheme 1 and Table 1). Figure
was prepared by using PyMOL.[24]
performs a nucleophilic attack onto bicarbonate, which is
coordinated by Glu72 and Arg49 through hydrogen bonds
and salt bridges, respectively. The latter results in an oxy-
anion-like p-quinone methide intermediate, which is stabi-
[
]
*
lized by water molecules WAT1–WAT3 (Figure 2), which
were found in the original crystal structure.
Interestingly, the polar residues that interact with the
substrate and those forming the hydrophobic environment for
the aromatic substrate moiety are strictly conserved among
the enzymes used in this study (Figure S3 in the Supporting
Information), although the overall sequence identities are
only between 40 and 80%.
In the carboxylation reaction that is catalyzed by the
enzyme in nature, deprotonation at Cb (accompanied by
tautomerization) yields cinnamic acid derivatives as products.
In contrast, hydration takes place through nucleophilic 1,6-
addition of a water molecule (WAT1, Figure 2) across the
quinone methide intermediate, followed by collapse of the
oxyanion, thereby causing a heterolytic cleavage of the
[10] J. A. Maresca, J. E. Graham, A. D. Pryant, Photosynth. Res.
2008, 97, 121 – 140.
[11] D. Brodkorb, M. Gottschall, R. Marmulla, F. Lꢃddeke, J. Harder,
J. Biol. Chem. 2010, 285, 30436 – 30442.
[12] A. Liese, K. Seelbach, A. Buchholz, J. Haberland in Industrial
Biotransformations, 2nd ed. (Eds.: A. Liese, K. Seelbach, C.
Wandrey), Wiley-VCH, Weinheim, 2006, p. 465 and p. 488.
[13] M. Wubbolts in Enzyme Catalysis in Organic Synthesis, 2nd ed
(Eds.: K. Drauz, H. Waldmann), Wiley-VCH, Weinheim, 2002,
pp. 686 – 697.
[14] a) G. Agnihotri, H.-w. Liu, Bioorg. Med. Chem. 2003, 11, 9 – 20;
b) B. J. Bahnson, V. E. Anderson, G. A. Petsko, Biochemistry
2002, 41, 2621 – 2629; c) A. F. Bell, Y. Feng, H. A. Hofstein, S.
Parikh, J. Wu, M. J. Rudolph, C. Kisker, A. Whitty, P. J. Tonge,
Chem. Biol. 2002, 9, 1247 – 1255.
[15] a) P. M. Leonard, A. M. Brzozowski, A. Lebedev, C. M. Mar-
shall, D. J. Smith, C. S. Verma, N. J. Walton, G. Grogan, Acta
Crystallogr. Sect. D 2006, 62, 1494 – 1501; b) P. M. Leonard, C. M.
Marshall, E. J. Dodson, N. J. Walton, G. Grogan, Acta Crystal-
logr. Sect. D 2004, 60, 2343 – 2345; c) A. Mitra, Y. Kitamura, M. J.
Gasson, A. Narbad, A. J. Parr, J. Payne, M. J. C. Rhodes, C.
Sewter, N. J. Walton, Arch. Biochem. Biophys. 1999, 365, 10 – 16.
[16] J. Jin, A. J. J. Straathof, M. W. H. Pinske, U. Hanefeld, Appl.
Microbiol. Biotechnol. 2011, 89, 1831 – 1840.
À
Cb Ccarboxylate bond through protonation of Cb by WAT1.
The latter yields the hydration product and reforms bicar-
bonate. The preference for (S)-alcohols is explained by
a nucleophilic front-side attack, which is favored by the
highly coordinated water network (WAT1–3).
In summary, we have presented for the first time the
=
enzyme-catalyzed addition of water to the C C double bond
of hydroxystyrene-type substrates in an asymmetric fashion.
[17] J. Jin, P. C. Oskam, S. K. Karmee, A. J. J. Straathof, U. Hanefeld,
Chem. Commun. 2010, 46, 8588 – 8590.
[18] a) U. T. Bornscheuer, R. J. Kazlauskas, Angew. Chem. 2004, 116,
6156 – 6165; Angew. Chem. Int. Ed. 2004, 43, 6032 – 6040; b) M.
Svedendahl Humble, P. Berglund, Eur. J. Org. Chem. 2011,
3391 – 3401.
[19] C. Wuensch, S. M. Glueck, J. Gross, D. Koszelewski, M. Schober,
K. Faber, Org. Lett. 2012, 14, 1974 – 1977.
[*] The docking calculations were initially performed without water
molecules present. During energy minimization water molecules
were added again automatically.
2296 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 2293 –2297

Angewandte
Chemie
[20] H. Rodriguez, I. Angulo, B. de Las Rivas, N. Campillo, J. A.
Paez, R. Munoz, J. M. Mancheno, Proteins Struct. Funct. Bioinf.
2010, 78, 1662 – 1676.
[21] W. Gu, J. Yang, Z. Lou, L. Liang, Y. Sun, J. Huang, X. Li, Y. Cao,
Z. Meng, K.-Q. Zhang, PLoS ONE 2011, 6, e16262.
Mozzarelli, L. Olivi, A. Coda, W. J. H. Van Berkel, Structure
1997, 5, 907 – 920.
[23] a) D. J. Hopper, L. Cottrell, Appl. Environ. Microbiol. 2003, 69,
3650 – 3652; b) C. D. Reeve, M. A. Carver, D. J. Hopper, Bio-
chem. J. 1989, 263, 431 – 437.
[22] a) R. H. H. van den Heuvel, M. W. Fraaije, W. J. H. van Berkel,
FEBS Lett. 2000, 481, 109 – 112; b) A. Mattevi, M. W. Fraaije, A.
[24] E. Krieger, G. Koraimann, G. Vriend, Proteins Struct. Funct.
Bioinf. 2002, 47, 393 – 402.
Angew. Chem. Int. Ed. 2013, 52, 2293 –2297
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 2297