Selective Ring-Opening of Di-Substituted Epoxides Catalysed by Halohydrin Dehalogenases

Article abstract of DOI:10.1002/cctc.201900103

Halohydrin dehalogenases (HHDHs) are valuable biocatalysts for the synthesis of β-substituted alcohols based on their epoxide ring-opening activity with a number of small anionic nucleophiles. In an attempt to further broaden the scope of substrates accepted by these enzymes, a panel of 22 HHDHs was investigated in the conversion of aliphatic and aromatic vicinally di-substituted trans-epoxides using azide as nucleophile. The majority of these HHDHs was able to convert aliphatic methyl-substituted epoxide substrates to the corresponding azidoalcohols, in some cases even with absolute regioselectivity. HheG from Ilumatobacter coccineus exhibited also high activity towards sterically more demanding di-substituted epoxides. This further expands the range of β-substituted alcohols that are accessible by HHDH catalysis.

Full text of DOI:10.1002/cctc.201900103

Accepted Article
Title: Selective Ring-Opening of Di-substituted Epoxides Catalysed by
Halohydrin Dehalogenases
Authors: Elia Calderini, Julia Wessel, Philipp Süss, Patrick Schrepfer,
Rainer Wardenga, and Anett Schallmey
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201900103
Link to VoR: http://dx.doi.org/10.1002/cctc.201900103
A Journal of
www.chemcatchem.org

10.1002/cctc.201900103
ChemCatChem
FULL PAPER
Selective Ring-Opening of Di-substituted Epoxides Catalysed by
Halohydrin Dehalogenases
Elia Calderini[a], Julia Wessel[a], Philipp Süss[b], Patrick Schrepfer[a], Rainer Wardenga[b], Anett
Schallmey*[a]
Abstract: Halohydrin dehalogenases (HHDHs) are valuable
biocatalysts for the synthesis of β-substituted alcohols based on their
HHDHs was able to convert aliphatic methyl-substituted epoxide
substrates to the corresponding azidoalcohols, in some cases even
with absolute regioselectivity. HheG from Ilumatobacter coccineus
exhibited also high activity towards sterically more demanding di-
substituted epoxides. This further expands the range of -substituted
alcohols that are accessible by HHDH catalysis.
epoxide ring-opening activity with
a number of small anionic
nucleophiles. In an attempt to further broaden the scope of substrates
accepted by these enzymes, a panel of 22 HHDHs was investigated
in the conversion of aliphatic and aromatic vicinally di-substituted
trans-epoxides using azide as nucleophile. The majority of these
Introduction
which acts in a concerted manner to catalyse the epoxide
formation and concomitant halide release.[13,14] In the epoxide
ring-opening reaction, the nucleophilic attack typically occurs at
the sterically less-hindered carbon atom following an SN2
mechanism.[15] Until recently, only terminal epoxides had been
reported to be accepted as substrates by HHDHs. When studying
HheC from Agrobacterium tumefaciens, HheA2 from Arthrobacter
sp. AD2 and HheB2 from Mycobacterium sp. GP1 for their activity
on vicinally di-substituted as well as cyclic epoxides, Majeric
Elenkov et al. found that none of the tested HHDHs exhibited
activity towards these sterically more demanding epoxides.[16]
Only in one paper by Hasnaoui-Dijoux et al., it is briefly noted that
HheC was observed to catalyse the ring-opening of 2,3-
epoxyheptane with nitrite as nucleophile but without experimental
detail.[4]
We recently reported the identification of a large set of novel
HHDH enzymes based on a database mining approach using
HHDH-specific sequence motifs.[17] Together with the five
previously known HHDHs, they compose a broad family of HHDH
enzymes subdivided into six phylogenetic subtypes A through G.
A representative subset of 17 novel HHDHs, which span the entire
phylogenetic tree, was subsequently cloned and biocatalytically
characterised.[18] This way, HheG from Ilumatobacter coccineus
was found to convert cyclic epoxides such as cyclohexene oxide
and limonene oxide with synthetic useful activity.[19] This
represents the first example for the ring-opening of cyclic epoxide
substrates using a halohydrin dehalogenase.
Enzymes are attractive catalysts as they offer the possibility to
attain building blocks through highly regio-, stereo-, and
chemoselective transformations. In this sense, halohydrin
dehalogenases (HHDHs, also called haloalcohol dehalogenases
or halohydrin hydrogen-halide-lyases), initially discovered for their
ability to degrade halogenated compounds,[1] are very useful
biocatalysts that catalyse the reversible dehalogenation of vicinal
haloalcohols via formation of the corresponding epoxides.[2,3] For
a practical application of HHDHs, their promiscuous epoxide ring-
opening activity with a range of small anionic nucleophiles such
as azide, cyanide, nitrite, cyanate or thiocyanate is even more
attractive as it enables the regio- and stereoselective formation of
novel C-N, C-C, C-O and C-S bonds.[4,5] Hence, HHDH-catalysed
reactions give access to synthetically important ß-substituted
alcohols and chiral epoxides. Biocatalytic examples include the
synthesis of ethyl (R)-4-cyano-3-hydroxybutyrate,
a chiral
synthon for the production of statin side chains[6], enantiopure
epihalohydrins,[7,8] highly enantioenriched oxazolidinones[9] and
tertiary alcohols.[10,11]
HHDHs belong to the superfamily of short-chain
dehydrogenases/reductases (SDR). In contrast to other SDR
enzymes, their catalytic mechanism does not require any
cofactor.[12] In HHDHs, the cofactor binding pocket found in SDR
enzymes is replaced by a nucleophile binding pocket.[13] Moreover,
they feature a conserved catalytic triad composed of Ser-Tyr-Arg,
Intrigued by this finding, we aimed to study also other vicinally di-
substituted but non-cyclic epoxides as substrates in
bioconversions with halohydrin dehalogenases. Employing a
representative set of 19 new and three previously known HHDH
enzymes, we screened for activity towards five different racemic
di-substituted trans-epoxides with aliphatic or aromatic
substituents using azide as nucleophile (Scheme 1). The resulting
azidoalcohols are useful intermediates for the synthesis of
corresponding aliphatic and arylaliphatic aminoalcohols. This is
[a]
[b]
E. Calderini, Dr. J. Wessel, Dr. P. Schrepfer, Prof. Dr. A. Schallmey.
Institute for Biochemistry, Biotechnology and Bioinformatics
Technische Universität Braunschweig
Spielmannstr. 7, 38106 Braunschweig, Germany
E-mail: a.schallmey@tu-braunschweig.de
Dr. P. Süss, Dr. R. Wardenga
Enzymicals AG
Walther-Rathenau-Straße 49A, 17489 Greifswald, Germany
Supporting information for this article is available on the WWW
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900103
ChemCatChem
FULL PAPER
the first comprehensive study of the ring-opening of vicinally di-
substituted epoxides catalysed by a large set of HHDHs.
respective pairs of enantiomers can be attained (Scheme 2). Thus,
the regioselectivity of active HHDHs in the conversion of
substrates 1-5 was studied. Most enzymes formed both
regioisomeric products a and b in varying ratios (Table 1). HheA3,
HheC and all tested E-type enzymes exhibited high
regioselectivity for substrates 1 and 2, giving 2-azidohexan-3-ol
(6a) and 2-azidoheptan-3-ol (7a) almost exclusively. In contrast,
enzymes from the HHDH subfamily D as well as HheG and
HheG2 formed both regioisomers (6a and 6b, as well as 7a and
7b) in roughly equal amounts. The same was observed for
chemical background azidolysis of 1 and 2 (“Control” in Table 1).
Compared to the conversion of 1, the terminal hydroxyl group
present in 3 had a positive effect on HheG’s regioselectivity. The
enzyme produced 5-azidohexane-1,4-diol (8a) with high
preference (Table 1). None of the active enzymes seemed to
display high regioselectivity in the conversion of 4 as in all cases
both azidoalcohol regioisomers 9a and 9b were formed. This
might not be surprising considering that both substituents of the
epoxide ring are rather similar in size, differing only in one CH2
group.
Scheme 1. HHDH-catalysed azidolysis of racemic di-substituted trans-epoxides
(1-5) resulting in the formation of two possible regioisomeric azidoalcohols (6-
10).
Results and Discussion
In accordance with literature,[13,15] none of the tested enzymes
exhibited a preference for formation of regioisomer b, except in
the conversion of epoxide 5. Here, HheG and HheG2 gave
azidoalcohol 10b almost exclusively, whereas HheD3 and HheF
formed also small amounts of regioisomer 10a. Chemical
azidolysis of epoxide 5 afforded 10b almost exclusively as well. In
this specific case, nucleophilic attack on the benzylic carbon is
favoured due to the electronic resonance effect of the aromatic
substituent.
A subset of 22 HHDHs from our panel, containing enzymes from
all six phylogenetic HHDH subtypes, was tested in bioconversions
of racemic di-substituted trans-epoxides 1-5 using azide as
exemplary nucleophile (Scheme 1). As a result, 17 (including
HheC) out of 22 tested HHDHs displayed activity in the
conversion of the two aliphatic substrates trans-2,3-epoxyhexane
(1) and trans-2,3-epoxyheptane (2), affording the corresponding
azidoalcohols in varying yields (Table 1). In contrast, significantly
less HHDHs converted the sterically more demanding epoxides
trans-4,5-epoxyhexan-1-ol (3), trans-3,4-epoxyheptane (4) and
trans--methylstyrene oxide (5). HheG from Ilumatobacter
coccineus was active on all five epoxide substrates yielding
highest conversion for substrates 1, 3, 4 and 5 among all tested
HHDHs. Significant conversion of 5 by HheG was unexpected as
the enzyme was shown to exhibit only very little activity on styrene
oxide.[18] For HheA2 and HheB2, hardly any product formation in
the conversion of the tested vicinally di-substituted epoxides was
observed, which is in agreement with previous findings.[15]
Interestingly, addition of a terminal hydroxyl group, as present
in substrate 3, abolished activity of all tested HHDHs that
displayed activity on the corresponding non-hydroxylated epoxide
1, except for HheG. Though this hydroxyl group is at the opposite
end of the aliphatic molecule compared to the epoxide ring, it can
support unproductive binding of the substrate in the enzyme
active site (see also below). Surprisingly, only HheG, but not
HheG2, is able to convert epoxide 3, albeit both enzymes share
74% sequence identity on protein level.[19]
Scheme 2. Possible regio- and stereoisomers in the azidolysis of vicinally di-
substituted trans-epoxides 1-5. The site of attack (α: less sterically hindered,
shown in blue; β: sterically more hindered, shown in plum) defines the
regioisomers obtained, while the enantioselectivity of the enzyme and the SN2-
type mechanism define the absolute configuration found in each set of
regioisomers. Hence, a total of two regioisomers a and b, each with their
respective (R,S)- and (S,R)-enantiomers, can be obtained.
Comparing substrates 2 and 4, which differ in the position of their
epoxide ring, only a reduced number of HHDHs was still active on
trans-3,4-epoxyheptane (4). Enzymes HheA3, HheD, HheD2,
HheD3 and HheD5, as well as HheG2 gave low to moderate
conversions, whereas HheG converted 4 completely within 15 h
reaction time.
The vicinally di-substituted epoxide substrates 1-5 each
contain two stereocenters. Depending on the site of nucleophilic
attack and the stereoconfiguration of the substrate during HHDH-
catalysed epoxide ring-opening, two regioisomers and their
Furthermore, the enantioselectivity of all active enzymes in
the ring-opening of racemic epoxides 1 and 2 was analysed. For
this, product enantiomeric excesses (eeP) and corresponding
apparent E-values for each of the two possible product
regioisomers were determined. As a result, most azidoalcohols
were formed with low to moderate enantioselectivity (Table 2).
Several HHDHs, however, displayed good enantioselectivity
(E>15) in the formation of azidoalcohol regioisomers 6b and 7b.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900103
ChemCatChem
FULL PAPER
Table 1. Substrate conversions and ratio of formed regioisomeric products a:b obtained in biocatalytic reactions of substrates 1-5 with HHDHs using azide as
nucleophile. All biotransformations were performed in duplicate.
Conversion [%] (ratio of regioisomeric products a:b)
HHDH[a]
HheA2
HheA3
HheA5
HheB2
HheB3
HheB4
HheB5
HheB6
HheB7
HheC
1
2
3
4
5
3.1±0.2 (n.d. [b]
)
3.1±0.2 (76:24)
97±<0.1 (97:3)
6.1±<0.1 (98:2)
5.6±<0.1 (88:12)
13±0.7 (67:33)
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
n.d. [b]
)
)
)
<0.01 (n.d. [b]
)
0.8±<0.1 (n.d. [b]
)
52±<0.1 (92:8)
3.6±<0.1 (n.d. [b]
5.4±<0.1 (n.d. [b]
13±0.9 (73:27)
1.2±0.3 (n.d. [b]
)
0.9±0.1 (n.d. [b]
0.5±<0.1 (n.d. [b]
n.d. [b]
)
)
)
<0.01 (n.d. [b]
n.d. [b]
)
)
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
)
)
)
)
)
)
0.6±<0.1 (n.d. [b]
)
<0.1 (n.d. [b]
)
0.2±<0.1 (n.d. [b]
97±0.1 (87:13)
)
0.5±0.1 (n.d. [b]
0.5±0.1 (n.d. [b]
)
59±1.6 (88:12)
76±5.3 (90:10)
81±2.4 (90:10)
44±12.6 (97:3)
44±2.8 (50:50)
64±2.2 (50:50)
88±3.8 (50:50)
)
100±0.1 (89:11)
100±<0.1 (89:11)
100±0.4 (98:2)
94±1.3 (55:45)
90±4.9 (50:50)
100±<0.1 (57:43)
100±<0.1 (50:50)
100±<0.1 (99:1)
85±<0.1 (98:2)
92±<0.1 (99:1)
15±<0.1 (97:3)
100±<0.1 (99:1)
70±2.1 (83:17)
100±0.1 (60:40)
96±0.9 (50:50)
1.6±0.6 (50:50)
0.5±<0.1 (n.d. [b]
0.6±<0.1 (n.d. [b]
)
)
0.6±0.3 (n.d. [b]
)
HheD
8.6 ± 0.2 (55:45)
2.6±0.1 (69:31)
25±2.2 (69:31)
4.5±0.6 (49:51)
0.5±0.2 (n.d. [b]
)
HheD2
HheD3
HheD5
HheE
0.6±<0.1 (n.d. [b]
1.1±0.3 (7:93)
0.7±<0.1 (n.d. [b]
)
)
100±<0.1 (64:36)
100±<0.1 (98:2)
12±<0.1 (91:9)
16±<0.1 (89:11)
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
<0.01 (n.d. [b]
)
)
)
)
)
)
0.4±0.2 (n.d. [b]
0.7±0.1 (n.d. [b]
0.5±0.1 (n.d. [b]
)
HheE2
HheE3
HheE4
HheE5
HheF
)
)
<0.1 (n.d. [b]
)
0.6±<0.1 (n.d. [b]
0.7±<0.1 (n.d. [b]
1.3±0.2 (13:87)
)
)
86±<0.1 (98:2)
56±0.7 (86:14)
100±<0.1 (50:50)
26±2 (50:50)
HheG
49±9.3 (92:8)
99±0.1 (51:49)
17±4.4 (76:24)
44±1.9 (0.4:99.6)
24±2.1 (0.4:99.6)
0.8±0.1 (1:99)
HheG2
Control
<0.01 (n.d. [b]
<0.01 (n.d. [b]
)
1.6±0.5 (50:50)
)
<0.01 (n.d. [b]
)
[a] Original strains and accession numbers of all enzymes are listed in Table S1 in the supporting information. [b] not determined
Hence, nucleophilic attack on the unfavoured carbon atom
occurred with higher enantioselectivity. Interestingly, most
azidoalcohol products in the conversion of 1 and 2 were
preferentially formed in 2S,3R-configuration. This indicates that
epoxides 1 and 2 with R,R-configuration were preferentially
attacked at the sterically less hindered carbon atom (C2),
whereas regioisomers 6b and 7b were preferentially formed by
nucleophilic attack of (S,S)-1 and (S,S)-2 at the sterically more
hindered carbon atom (C3). In contrast, nucleophilic attack of
azide at C2 in (S,S)-1 and (S,S)-2 yields azidoalcohols 6a and 7a
in 2R,3S-configuration, as found for HheA3, HheE2, HheF and
HheG2 with epoxide 1 and HheE4 and HheF with epoxide 2
(Table 2). A detailed analysis of the enzymes’ regiopreferences in
the azidolysis of R,R- and S,S-enantiomers of epoxides 1 and 2
is given in Table S3 in the supporting information. Interestingly,
D-type HHDHs displayed
a slightly higher preference for
nucleophilic attack of (S,S)-1 and (S,S)-2 at the sterically more
hindered carbon atom, whereas most other HHDHs exhibited a
clear preference for nucleophilic attack of the sterically less
hindered carbon atom independent of the substrate’s
stereoconfiguration. Additionally, the enantioselectivity of active
HHDHs in the ring-opening of rac-5 was determined and results
are given in Table S2 in the supporting information. No
enantioselectivity was observed in the chemical epoxide ring
opening of 1, 2 and 5 on preparative scale with azide as
nucleophile (data not shown).
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900103
ChemCatChem
FULL PAPER
Table 2. Conversions (C), enantiomeric excesses (eeP) and calculated apparent enantiomeric ratios (Eapp) of regioisomeric azidoalcohols 6a,b and 7a,b formed
in the HHDH-catalysed azidolysis of epoxides 1 and 2. The absolute configuration of the preferentially formed enantiomer of each regioisomer is given in
parentheses.
6a
eeP [%]
47
6b
eeP [%]
46
7a
eeP [%]
0.7
30
7b
eeP [%]
25.5
77
HHDH[a]
HheA3
HheB3
HheB5
HheB6
HheB7
HheC
C [%]
48
Eapp
C [%]
4.1
3.6
7.1
7.6
8.1
1.3
22
Eapp
C [%]
93
8.7
85
88
89
97
51
45
56
50
99
83
90
14
99
58
60
48
Eapp
C [%]
3.4
4.3
13
Eapp
4.2 (R,S)
2.2 (S,R)
2.1 (S,R)
2.6 (S,R)
2.9 (S,R)
2.6 (S,R)
1.9 (S,R)
1.7 (S,R)
7.7 (S,R)
6.5 (S,R)
1.1 (S,R)
6.4 (R,S)
1.1 (R,S)
n.d.[b]
2.8 (S,R)
5.0 (S,R)
14 (S,R)
18 (S,R)
30 (S,R)
n.d.[b]
1.0 (S,R)
1.9 (S,R)
1.6 (S,R)
1.4 (S,R)
1.4 (S,R)
1.1 (S,R)
4.0 (S,R)
3.7 (S,R)
7.4 (S,R)
1.3 (S,R)
1.1 (S,R)
1.2 (S,R)
1.0 (S,R)
2.0 (R,S)
1.1 (S,R)
1.9 (R,S)
4.4 (S,R)
1.2 (S,R)
1.7 (S,R)
7.7 (S,R)
39 (S,R)
28 (S,R)
32 (S,R)
19 (S,R)
14 (S,R)
8.1 (S,R)
9.0 (S,R)
1.2 (S,R)
n.d.[b]
9.7
52
35
66
25
86
24
95
69
24
89
16
11
93
73
24
93
17
11
94
n.d.[b]
86
4.6
60
2.3
43
90
43
34
HheD
22
28
17 (S,R)
17 (S,R)
3.4 (S,R)
3.2 (S,R)
n.d.[b]
87
HheD2
HheD3
HheD5
HheE
32
22
32
84
57
45
78
44
65
44
43
76
43
80
64
46
36
44
13
50
11
98
4.9
71
1.7
1.1
1.8
n.d.[b]
1.8
8.0
50
n.d.[b]
n.d.[b]
n.d. [b]
n.d.[b]
n.d.[b]
32
3.5
11
0.8
1.5
1.4
0.5
0.6
12
n.d.[b]
n.d.[b]
n.d.[b]
n.d.[b]
n.d.[b]
19
10
n.d.[b]
n.d.[b]
HheE2
HheE3
HheE4
HheE5
HheF
14
4.5
n.d.[b]
13
n.d.[b]
2.3
33
n.d.[b]
<0.1
84
n.d.[b]
n.d.[b]
2.2 (S,R)
5.4 (R,S)
1.6 (S,R)
3.6 (R,S)
n.d.[b]
3.5
32
n.d.[b]
48
54
2.0 (S,R)
1.5 (S,R)
5.6 (S,R)
1.5 (S,R)
17 (S,R)
1.2 (S,R)
HheG
50
16
15
63
40
89
13
54
13
67
48
10
HheG2
7.4
[a] Original strains and accession numbers of all enzymes are listed in Table S1 in the supporting information. [b] not determined
Among the tested halohydrin dehalogenases, HheG was the
only enzyme able to convert all five epoxide substrates. This
might be explained by the broad and open active site which is
present in HheG’s crystal structure.[19] In contrast, other HHDHs
with solved crystal structure, namely HheA/A2, HheB/B2 and
HheC,[13,20,21] possess active sites that are more buried within the
protein structure, with a rather narrow substrate channel leading
to the active site. Hence, the broad active site cleft of HheG likely
offers more space for the binding of bulky substrates such as the
herein tested vicinally di-substituted epoxides. This is also
supported by calculations of the active site volume of HheG,
which was found to be 2.7 times larger than the calculated active
site volume of HheC. On the other hand, epoxide substrates with
much bulkier substituents on both carbon atoms of the epoxide
ring are likely not accepted by HheG as it was previously shown
that trans-stilbene oxide was not converted by HheG or other
tested HHDHs.[18]
epoxides. This enantioselectivity is caused by non-productive
binding of the respective (S)-epoxides in the active site of
HheC.[15] Here, this HHDH seems significantly less
stereoselective in the conversion of 1 and 2 as products 6a and
7a have been obtained with rather low ee (34 % and 5 %,
respectively). On the other hand, HheC displayed very high
regioselectivity yielding azidoalcohol regioisomers 6a and 7a
almost exclusively. The regioselectivity of a HHDH during epoxide
ring opening of vicinally di-substituted epoxides is determined by
the exact positioning of epoxide substrate and nucleophile in the
active site of the enzyme. This relative positioning of epoxide ring
and nucleophile to each other will dictate which of the two carbon
atoms of the epoxide ring can be attacked by the nucleophile. To
investigate this in more detail for HheC, molecular docking of both
enantiomers of epoxides 1-5 into the enzyme’s active site was
performed. This revealed that both substrate enantiomers of 1
and 2 bind in similar productive orientations with the methyl-
substituted carbon atom in closer proximity to the nucleophile
binding pocket (Figure 1A and B). This is in line with our
HheC from A. tumefaciens was previously demonstrated to be
R-selective, especially in the conversion of terminal aromatic
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900103
ChemCatChem
FULL PAPER
experimental data that both epoxide enantiomers can be
converted by HheC, but the methyl-substituted carbon atom is
always attacked preferentially, resulting in the observed high
regio- but low enantioselectivity of this enzyme. In contrast,
docking of (R,R)- and (S,S)-4 yielded possible binding modes with
the epoxide oxygen being too far away from the serine of the
catalytic triad to facilitate catalysis (Figure 1D).
Similarly, for both enantiomers of epoxide 5 no binding mode
with the epoxide oxygen in hydrogen bonding distance to the
catalytic triad serine and tyrosine of HheC was observed (Figure
1E). This is again in agreement with our experimental data
obtained for HheC, which did not convert epoxides 4 and 5.
Docking of epoxide 3 revealed non-productive binding modes for
both substrate enantiomers with the terminal hydroxyl group
pointing towards the catalytic amino acid residues (Figure 1C).
This flipped binding mode of 3 found for HheC might also occur in
other HHDHs and could thus explain the lack of activity of most
HHDHs toward this epoxide. Since possible unproductive binding
modes could be obtained for 3, 4 and 5, all three epoxides might
act as inhibitors of HheC.
Figure 1. Docking results for HheC (PDB: 1ZMT) with substrates 1 (A), 2 (B), 3 (C), 4 (D) and 5 (E). Substrates with R,R-configuration are shown in light blue, while
substrates with S,S-configuration are shown in orange. Hydrogen bonds between epoxide oxygen and catalytic residues S132 and Y145 are represented as green
dotted lines. The water molecule present in the nucleophile binding pocket is shown as green sphere to indicate the position of the nucleophile.
For comparison, similar dockings of epoxides 1-5 were also
performed with HheG. Here, possible productive binding modes
of both enantiomers of 1 and 2 were obtained with the epoxide
oxygen in hydrogen bonding distance to the catalytic serine and
tyrosine (Figure 2). In contrast to HheC, however, both substrate
enantiomers display different orientations with the longer alkyl
side chain pointing in opposite directions. This is only possible
due to the much larger active site cavity of HheG compared to
HheC. With the other three substrates (3-5) and HheG, the
docking results are less informative as not all substrate
enantiomers yielded possible productive binding modes
(Supporting Information, S4). This is not surprising considering
that no substrate-bound structure of HheG is available yet. Hence,
the resulting substrate binding poses found for HheG may not
reflect the real binding modes. In case of HheC, molecular
dockings were performed using an epoxide-bound enzyme
structure (PDB: 1ZMT).[15] Therefore, the obtained docking results
will be more reliable. As no crystal structure of a member of HHDH
subfamily E has been determined so far, no structural insights into
the observed high regioselectivity could be gained for these
enzymes.
Selected regioselective HHDH-catalysed reactions were also
performed on preparative scale to demonstrate the synthetic
potential of this class of enzymes. For this, epoxide substrates 1
and 2 were each converted with a regioselective HHDH (HheE for
substrate 1 and HheE5 for substrate 2) by means of whole-cell
biocatalysis. Using each 200 mg (50 mM) of epoxide per reaction,
142 mg of regioisomer 6a and 151 mg of regioisomer 7a (50 %
and 48 % isolated yield, respectively) were obtained in pure form.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900103
ChemCatChem
FULL PAPER
Figure 2. Docking results for HheG (PDB: 5O30) with substrates 1 (A) and 2 (B). Substrates with R,R-configuration are shown in light blue, while substrates with
S,S-configuration are shown in orange. Hydrogen bonds between epoxide oxygen and catalytic residues S152 and Y165 are represented as green dotted lines.
The water molecule present in the nucleophile binding pocket is shown as green sphere to indicate the position of the nucleophile.
Trans-β-methylstyrene and trans-3-heptene were purchased from TCI
Deutschland GmbH (Eschborn, Germany). Trans-2-hexen-1-ol was
purchased from J&K Scientific (Lommel, Belgium). (1S,2S)-(-)-1-
phenylpropylene oxide and (1R,2R)-(+)-1-phenylpropylene oxide were
Conclusions
purchased from Sigma-Aldrich (Darmstadt, Germany).
A set of 22 halohydrin dehalogenases, representing all currently
known phylogenetic subtypes, was studied in the azidolysis of five
Racemic epoxides trans-2,3-epoxyhexane (1) and trans-2,3-
vicinally di-substituted epoxides. The majority of these enzymes
displayed significant activity towards simple aliphatic epoxides
that carried a methyl group at one of the carbon atoms of the
epoxide ring. Several HHDHs were even found to be highly
regioselective, facilitating the synthesis of regioisomerically pure
azidoalcohols on preparative scale. HheG from Ilumatobacter
coccineus converted also sterically more demanding non-terminal
epoxides with good to high activity. Overall, the observed regio-
and stereoselectivity of active enzymes towards the tested
vicinally di-substituted epoxides was found to be enzyme- and
substrate-dependent. Docking studies based on an epoxide-
bound structure of HheC revealed first structural insights into the
observed substrate and regio-selectivity of this enzyme.
With this, we could demonstrate that the substrate scope of
HHDHs is not limited to terminal epoxides. Instead, they can be
applied in the conversion of non-terminal epoxide substrates, thus,
expanding the range of accessible β-substituted alcohols. This
further broadens the biocatalytic applicability of HHDHs, in
particular of HheG, substantially. Further improvement of the
enzymes’ enantioselectivities by protein engineering will enable
the synthesis of optically pure products in the future.
epoxyheptane (2) were synthesised according to Sharma et al. using a
slightly modified protocol.[22] Epoxidations were carried out in CH2Cl2 (40
mL g-1 of alkene). m-Chloroperbenzoic acid (1.5 eq.) was added in small
portions over 15 min at room temperature (RT). After 1 h at RT, the mixture
was stirred on ice for 5 min and the white slurry was filtered. 5% w/v aq.
Na2SO3 (40 mL g-1 alkene) was added and stirred at RT for 15 min, then
the phases were separated. The aqueous layer was extracted two times
with CH2Cl2. The combined organic layers were washed with sat. aq.
NaHCO3 (1×) and brine (1×), dried over MgSO4 and filtered before solvent
removal by evaporation. The epoxides were purified by column
chromatography using a mixture of cyclohexane/ethyacetate, 95:5.
Racemic trans-4,5-epoxyhexan-1-ol (3), trans-3,4-epoxyheptane (4) and
trans--methylstyrene oxide (5) were synthesised by mixing the respective
alkene to a final concentration of 200 mM in water with 30% acetonitrile
and 10% of acetone. 3 eq. of oxone® were added portion-wise over 4
hours. The pH was kept at 8 by adding NaHCO3 to the solution.[23] The
reactions were followed by TLC. When all substrate was converted, the
reaction mixture was extracted three times with tert-butylmethylether
(TBME). The combined organic layers were washed with brine (1×), dried
over Na2SO4 and filtered before solvent removal by evaporation. The
epoxides
were
purified
by
column
chromatography
with
cyclohexane/isopropanol (95:5) for 3, cyclohexane/ethyl acetate (95:5) for
4 and cyclohexane/ethyl acetate (90:10) for 5. Formation and purity of the
epoxides was confirmed by NMR and resulting NMR data were consistent
with literature data.[24–26]
Experimental Section
Bacterial strains and plasmids
Chemicals
The strains Escherichia coli BL21(DE3) (Life Technologies, Darmstadt,
Germany), E. coli C43(DE3) (Lucigen Corporation, Middleton, WI, USA)
Trans-2-hexene was purchased from Acros Organics (Geel, Belgium),
trans-2-heptene was purchased from abcr GmbH (Karlsruhe, Germany).
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900103
ChemCatChem
FULL PAPER
and E. coli Top10 (Life Technologies) were used as hosts for heterologous
protein production. All HHDH genes except hheA2, hheB2 and hheC were
expressed from pET28a(+)-based vectors, utilizing a T7 promotor and
resulting in a N-terminal hexahistidin (His6) tag fusion.[17] Instead, vectors
pBAD-hheA2, pBAD-hheB2 and pBAD-hheC, utilizing an arabinose-
inducible promotor, were used for the expression of the HheA2, HheB2
and HheC genes, respectively, as described previously.[12,27]
phenylpropan-2-ol (10b) next to a small amount of (1S,2R)-2-azido-1-
phenylpropan-1-ol (10a). In contrast, ring-opening of the opposite
enantiomer (R,R)-5 generated (1S,2R)-1-azido-1-phenylpropan-2-ol (10b)
in excess and and a small amount of (1R,2S)-2-azido-1-phenylpropan-1-
ol (10a).
Preparative scale biotransformation
Expression and purification of HHDHs
Preparative-scale conversions of 1 and 2 were performed using whole
cells of E. coli harbouring either HheE, or HheE5. All biotransformations
were performed in 40 mL of 50 mM potassium phosphate buffer at pH 7.5
containing a cell density of OD600 = 40 (equivalent to 60 g wet cells per liter
reaction), 50 mM of either 1 or 2 and 100 mM sodium azide (NaN3). After
24 h at 25 °C, the reaction mixture was extracted with 40 mL TBME and
filtered through Celite®. The extracted organic layer was washed with brine
(1×) and MilliQ water (1×), dried over Na2SO4 and filtered before solvent
evaporation. 142 mg (50% yield) of 6a and 151 mg (48% yield) of 6b could
be recovered after column chromatography using cyclohexane/diethyl
ether, 90:10.
All HHDH enzymes except HheE4 and HheG2 were produced and purified
as reported previously.[12,18] Enzymes HheE4 and HheG2 were produced
in E. coli BL21(DE3). Respective overnight cultures were used to inoculate
(10% v/v) 500 mL TB medium (4 mL L-1 glycerol, 12 g L-1 peptone, 24 g L-
1
yeast extract, 0.17 M KH2PO4, 0.74 M K2HPO4) supplemented with 50
mg L-1 kanamycin and 0.2 mM IPTG. Expression cultures were grown at
22°C for 24 h. Cells were harvested by centrifugation (4,400 g, 20 min at
4°C) and cell pellets were stored at -20°C until further use. Both enzymes
were purified via affinity chromatography according to a previously
published protocol.[18] A list of all HHDHs used in this work with their
respective source organisms and accession numbers is given in Table S2
in the supporting information.
Chemical synthesis of authentic azidoalcohol standards
1
mmol of either trans-2,3-epoxyhexane (100 mg) or trans-2,3-
General biotransformation experiment
epoxyheptane (114 mg) were mixed with 202 mg of NaN3 (3.1 eq.) and
166 mg of NH4Cl (3.1 eq.) in 3.5 mL methanol and refluxed at 65°C until
no more substrate was visible on TLC (5-6 h). The reaction mixture was
then diluted with diethyl ether, washed with brine, the water phase was
then extracted twice with diethyl ether. The organic layers were combined
and dried over Na2SO4. The crude extracts were purified by column
chromatography (cyclohexane/diethyl ether, 90:10) yielding 52%
azidohexanol (6) and 48% azidoheptanol (7). [30]
Small-scale biotransformations (1 mL) were performed in 50 mM
potassium phosphate buffer at pH 7.5 containing 150 µg mL-1 HHDH
enzyme, 5 mM substrate (1-5) and 20 mM sodium azide (NaN3) at 25 °C.
After 15 h (substrates 1-4) or 4 h (substrate 5) of reaction, an aliquot of
400 µL was taken and extracted with the same volume of TBME containing
0.1 % v/v dodecane as internal standard. The resulting organic phase was
dried over anhydrous MgSO4 prior to injection on GC or GC-MS. All
biotransformations were carried out in duplicate. Chemical background
azidolysis was monitored in reactions using the same reaction conditions
but omitting HHDH enzyme.
Following a similar protocol, azidoalcohols 8, 9 and 10 were synthesised
by dissolving the corresponding epoxides 3-5 in MeOH with 5% H2O to a
final concentration of 200 mM, NaN3 (3.1 eq.) and NH4Cl (3.1 eq.) were
added and the reaction was stirred over night at 65 °C under reflux.
Methanol was evaporated, the crude extract was dissolved in TBME and
washed with the same volume of brine (1x). The water phase was
extracted three times with TBME. The combined organic layers were dried
over Na2SO4 and filtered before solvent removal by evaporation.
Diasteromeric mixtures of 8a/b, 9a/b and 10a/b were purified by column
chromatography using chloroform/acetone, 95:5; cyclohexane/ethyl
acetate, 95:5, and heptane/ethyl acetate, 70:30, respectively.
Determination of enantiomeric excesses
In order to distinguish trans-(2S,3S)-epoxyhexane and trans-(2R,3R)-
epoxyhexane as well as trans-(2S,3S)-epoxyheptane and trans-(2R,3R)-
epoxyheptane on chiral GC, trans-(2S,3S)-epoxyhexane (S,S-1) and
trans-(2S,3S)-epoxyheptane (S,S-2) were selectively synthesised
according to a published protocol using E. coli whole cells harbouring the
styrene monooxygenase (StyAB) from Rhodococcus sp. ST-10.[28,29]
Assignment of all azidoalcohol enantiomers on chiral GC was achieved by
further conversion of the obtained (S,S)-1 and (S,S)-2 using each a
regioselective and a non-regioselective HHDH. Using a regioselective
HHDH (HheE5 for 2 and HheE for 1), only (2R,3S)-6a and (2R,3S)-7a are
produced, while using non-regioselective enzymes (HheD5 for 2 and
HheG for 1), (2R,3S)-6a, (2S,3R)-6b, (2R,3S)-7a and (2S,3R)-7b are
obtained. Reactions were performed in 50 mM potassium phosphate
buffer at pH 7.5 using 5 mM (S,S)-1 or (S,S)-2, 150 µg mL-1 HHDH and 20
mM sodium azide at 25 °C. After 15 h, an aliquot of 400 µL was taken and
extracted with the same volume of TBME containing 0.1 % v/v dodecane
as internal standard. The resulting organic phase was dried over
anhydrous MgSO4 before injection on chiral GC. GC-MS analysis of the
same material was performed to assign the different regioisomers of 6 and
7.
Diastereomeric mixture of 2-azidohexan-3-ol (6a) and 3-azidohexan-
2-ol (6b): pale oil, 1H NMR (600 MHz, CDCl3) δ = 3.90 – 3.81 (m, 1H, 6b),
3.65 – 3.58 (m, 1H, 6a), 3.52 (dq, J=6.7, 3.9, 1H, 6a), 3.42 – 3.35 (m, 1H,
66b), 1.61 – 1.29 (m, 14H, 6a and 6b), 1.25 (d, J=6.7, 3H, 6a), 1.19 (d,
J=6.4, 3H, 6b), 0.96 (t, J=7.2, 3H, 6b), 0.94 (t, J=7.3, 3H, 6a). 13C NMR
(151 MHz, CDCl3) δ 73.55 (6a), 69.99 (6b), 67.92 (6b), 61.77 (6a), 34.57
(6a), 32.07 (6b), 19.65 (6b), 18.96 (6a), 18.13 (6b), 13.91 (6a), 13.80 (6b),
13.11 (6a). ESI-HRMS: [M+Na+] = 166.09524 m/z (calculated [M+Na+] =
166.09508 m/z).
Diastereomeric mixture of 2-azidohexan-3,6-diol (8a) and 3-
azidohexan-2,6-diol (8b): pale oil, 1H NMR (600 MHz, CDCl3) δ = 3.92 –
3.85 (m, 2H, 8b), 3.82 – 3.60 (m, 8a 3H, 8b 1H), 3.50 – 3.44 (m, 1H, 8a),
3.42 – 3.40 (m, J=4.98, 1H, 8b), 2.86 (s, 1H, 8a), 2.63 (s, 1H, 8b), 2.34 (s,
1H, 8a), 2.17 (s, 1H, 8b), 1.65 – 1.34 (m, 8H, 8a and 8b), 0.97 (t, J=7.0,
3H, 8a), 0.96 (t, J=6.9, 3H, 8b). 13C NMR (151 MHz, CDCl3) δ 73.80 (8a),
72.39 (8b), 67.04 (8b), 64.64 (8a), 63.33 (8a), 62.60 (8b), 36.00 (8b),
32.83 (8a), 19.74 (8a), 19.01 (8b), 14.13 (8b), 14.03 (8a). ESI-HRMS:
[M+Na+] = 182.09010 m/z (calculated [M+Na+] = 182.0905 m/z).
The assignment of enantiomers of azidoalcohols 10a and 10b on chiral
GC was performed in a similar way starting from commercially available
enantiopure epoxides (S,S)-5 and (R,R)-5. Due to the SN2-type
mechanism and the observed regioselectivity of the chemical azidolysis of
5, chemical conversion of (S,S)-5 lead to an excess of (1R,2S)-1-azido-1-
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900103
ChemCatChem
FULL PAPER
Diastereomeric mixture of 3-azidoheptan-4-ol (9a) and 4-azidoheptan-
3-ol (9b): pale oil, 1H NMR (600 MHz, CDCl3) δ 3.72 – 3.64 (m, 1H, 9a),
3.62 – 3.55 (m, 1H, 9b), 3.42 – 3.32 (m, 1H, 9b), 3.31 – 3.23 (m, 1H, 9a),
1.73 – 1.31 (m, 14H, 9a and 9b), 1.04 (t, J = 7.4 Hz, 3H, 9a), 1.00 (t, J =
7.4 Hz, 3H, 9b), 0.96 (t, J = 7.1 Hz, 3H, 9b), 0.95 (t, J = 7.2 Hz, 3H, 9a).
13C NMR (151 MHz, CDCl3) δ = 75.64 (9b), 73.64 (9a), 69.46 (9a), 67.26
(9b), 34.70 (9a), 31.74 (9b), 25.58 (9b), 22.90 (9a), 19.93 (9b), 19.20 (9a),
14.20 (9a), 14.10 (9b), 11.24 (9a), 10.43 (9b). ESI-HRMS: [M+Na+] =
180.11081 m/z (calculated [M+Na+] = 180.1112 m/z).
GC-MS analysis of azidoalcohol products was performed on a Shimadzu
GCMS-QP2010SE equipped with
a ZB-5MS GUARDIAN column
(Phenomenex) and helium as carrier gas.
NMR spectra were recorded on a Bruker Avance 600 MHz spectrometer
equipped with an inverse 1H/ 13C/ 15N/ 31P quadruple resonance cryoprobe
head and z field gradients. The sample was dissolved in deuterated
chloroform (CDCl3) and 1D proton, DEPT, COSY double quantum filter,
and HSQC experiments were performed at 25°C. The regioisomers of
chemically synthesised azidoalcohols 6, 8 and 9 were identified by NMR
spectroscopy employing COSY and HSQC techniques and further
compared to the assignment of regioisomers by GC and GC-MS.
Obtained NMR data for 7a/b[30] and 10a/b[31,32] were consistent with
literature data.
Analytical methods
Docking studies
Achiral GC analysis was performed on a GC-2010 plus gas chromatograph
(Shimadzu, Duisburg, Germany) equipped with an Optima 5ms column
(Macherey-Nagel, Düren, Germany) and FID detection using hydrogen as
carrier gas. Chiral GC analysis was performed on a GC-2010 plus gas
chromatograph (Shimadzu) equipped with two different chiral columns and
FID detection using hydrogen as carrier gas. A Lipodex E column
(Macherey-Nagel) was used for the separation of azidoalcohol products 6-
10, whereas a HYDRODEX γ-DiMOM column (Macherey-Nagel) was used
for the separation of epoxide substrates 1-5. Temperature programs and
retention times are listed in Table S23 in the supporting information.
Molecular docking of MM2-force field energy minimized (S,S)- and (R,R)-
1 to 5 into the active site of a monomeric representation of HheC (PDB-ID:
1ZMT) and HheG (PDB-ID: 5O30) was performed using the AutoDockVina
program environment of YASARA Structure.[34,35] All ten substrate
structures were docked into a simulation cell (X size = 25 Å, Y size = 25 Å,
Z size = 25 Å; angles: α = 90°, β = 90°, γ =90°) around the four residues
K91, L155, E197 and E216 for HheC and the corresponding residues D111,
I175, E214 and E233 in HheG. For each substrate, 999 docking runs were
performed with atoms and bonds of the corresponding substrates set as
rigid. The only exception is substrate 3 with HheG where a flexible ligand
was necessary to obtain one cluster with productive binding. Docking of
each substrate resulted in one or more clusters of similar binding modes
which were energy minimized using the YASARA2-force field after
inserting the water molecule of the nucleophile binding pocket as fixed
element. This water molecule is present in the respective crystal structures
and represents the position of the nucleophile in the binding pocket.
Figures of the structures with docked ligands were generated with PyMOL
Molecular Graphics System version 2.2.3 (Schrödinger, LLC, New York,
NY, USA).
Conversions of epoxides into the corresponding azidoalcohols were
determined by achiral GC based on relative peak areas. Ratios of
regioisomers 6a/b, 8a/b and 9a/b were calculated based on peak areas
derived from chiral GC, whereas ratios of regioisomers 7a/b, 10a/b, were
calculated based on peak areas derived from achiral GC. Enantiomeric
excesses (ee) of azidoalcohol products 6, 7 and 10 were calculated
according to Chen et al. based on chiral GC data.[33] Apparent
enantiomeric ratios (Eapp) of HHDHs for the formation of regioisomeric
products 6a and 6b, 7a and 7b as well as 10a and 10b were calculated
according to Chen et al. based on conversion (C) and product
enantiomeric excess (eeP).[33]
[6]
N.-W. Wan, Z.-Q. Liu, K. Huang, Z.-Y. Shen, F. Xue, Y.-G. Zheng, Y.-
C. Shen, RSC Adv. 2014, 4, 64027–64031.
Acknowledgements
[7]
H. M. S. Assis, A. T. Bull, D. J. Hardman, Enzyme Microb. Technol.
1998, 22, 545–551.
[8]
T. Nakamura, T. Nagasawa, F. Yu, I. Watanabe, H. Yamada,
Tetrahedron 1994, 50, 11821–11826.
M. M. Elenkov, L. Tang, A. Meetsma, B. Hauer, D. B. Janssen, Org.
Lett. 2008, 10, 2417–2420.
J. H. Lutje Spelberg, L. Tang, M. van Gelder, R. M. Kellogg, D. B.
Janssen, Tetrahedron Asymmetry 2002, 13, 1083–1089.
C. Molinaro, A.-A. Guilbault, B. Kosjek, Org. Lett. 2010, 12, 3772–
3775.
J. E. T. van Hylckama Vlieg, L. Tang, J. H. Lutje Spelberg, T. Smilda,
G. J. Poelarends, T. Bosma, A. E. J. van Merode, M. W. Fraaije, D. B.
Janssen, J. Bacteriol. 2001, 183, 5058–5066.
We thank Dr. Melinda Fekete from Enzymicals for helpful advice
regarding chemical syntheses. This work has received funding
from the European Union’s Horizon 2020 MSCA ITN-EID program
under grant agreement No 634200 (BIOCASCADES) and the
Horizon 2020 research and innovation programme under grant
agreement No 635595 (CARBAZYMES).
[9]
[10]
[11]
[12]
Keywords: Biocatalysis • Enzymes • Epoxide ring opening •
Halohydrin dehalogenase • Regioselectivity.
[13]
[14]
[15]
[16]
[17]
[18]
R. M. de Jong, J. J. W. Tiesinga, H. J. Rozeboom, K. H. Kalk, L. Tang,
D. B. Janssen,B. W. Dijkstra, EMBO J. 2003, 22, 4933–4944.
R. M. de Jong, B. W. Dijkstra, Curr. Opin. Struct. Biol. 2003, 13, 722–
730.
R. M. de Jong, J. J. W. Tiesinga, A. Villa, L. Tang, D. B. Janssen, B.
W. Dijkstra, J. Am. Chem. Soc. 2005, 127, 13338–13343.
M. M. Elenkov, B. Hauer, D. B. Janssen, Adv. Synth. Catal. 2006, 348,
579–585.
M. Schallmey, J. Koopmeiners, E. Wells, R. Wardenga, A. Schallmey,
Appl. Environ. Microbiol. 2014, 80, 7303–7315.
J. Koopmeiners, B. Halmschlag, M. Schallmey, A. Schallmey, Appl.
Microbiol. Biotechnol. 2016, 100, 7517–7527.
[1]
[2]
C. E. Castro, E. W. Bartnicki, Biochemistry 1968, 7, 3213–3218.
Z.-Y. You, Z.-Q. Liu, Y.-G. Zheng, Appl. Microbiol. Biotechnol. 2013,
97, 9–21.
[3]
[4]
[5]
A. Schallmey, M. Schallmey, Appl. Microbiol. Biotechnol. 2016, 100,
7827–7839.
G. Hasnaoui-Dijoux, M. Majerić Elenkov, J. H. Lutje Spelberg, B.
Hauer, D. B. Janssen, ChemBioChem 2008, 9, 1048–1051.
E. J. de Vries, D. B. Janssen, Curr. Opin. Biotechnol. 2003, 14, 414–
420.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900103
ChemCatChem
FULL PAPER
[19]
[20]
[21]
[22]
J. Koopmeiners, C. Diederich, J. Solarczek, H. Voß, J. Mayer, W.
Blankenfeldt, A. Schallmey, ACS Catal. 2017, 7, 6877–6886.
R. M. de Jong, K. H. Kalk, L. Tang, D. B. Janssen, B. W. Dijkstra, J.
Bacteriol. 2006, 188, 4051–4056.
F. Watanabe, F. Yu, A. Ohtaki, Y. Yamanaka, K. Noguchi, M. Yohda,
M. Odaka, Proteins Struct. Funct. Bioinforma. 2015, 83, 2230–2239.
R. Sharma, P. G. Bulger, M. McNevin, P. G. Dormer, R. G. Ball, E.
Streckfuss, J. F. Cuff, J. Yin, C. Chen, Org. Lett. 2009, 11, 3194–
3197.
[23]
[24]
[25]
[26]
[27]
S. E. Denmark, Z. Wu, C. M. Crudden, H. Matsuhashi, J. Org. Chem.
1997, 62, 8288–8289.
W. Kroutil, M. Mischitz, K. Faber, J. Chem. Soc. Perkin 1 1997, 0,
3629–3636.
E. G. Ankudey, H. F. Olivo, T. L. Peeples, Green Chem. 2006, 8, 923–
926.
J. Lindborg, A. Tanskanen, L. T. Kanerva, Biocatal. Biotransformation
2009, 27, 204–210.
R. M. Haak, C. Tarabiono, D. B. Janssen, A. J. Minnaard, J. G. de
Vries, B. L. Feringa, Org. Biomol. Chem. 2007, 5, 318–323.
H. Toda, R. Imae, N. Itoh, Adv. Synth. Catal. 2014, 356, 3443–3450.
H. Toda, R. Imae, N. Itoh, Tetrahedron Asymmetry 2012, 23, 1542–
1549.
[28]
[29]
[30]
G. Righi, S. Pietrantonio, C. Bonini, Tetrahedron 2001, 57, 10039–
10046.
[31]
[32]
B. Yang, Z. Lu, ACS Catal. 2017, 7, 8362–8365.
I. A. Sayyed, A. Sudalai, Tetrahedron Asymmetry 2004, 15, 3111–
3116.
[33]
[34]
[35]
C. S. Chen, Y. Fujimoto, G. Girdaukas, C. J. Sih, J. Am. Chem. Soc.
1982, 104, 7294–7299.
G. M. Morris, D. S. Goodsell, R. S. Halliday, R. Huey, W. E. Hart, R.
K. Belew, A. J. Olson, J. Comput. Chem. 1998, 19, 1639–1662.
O. Trott, A. J. Olson, J. Comput. Chem. 2010, 31, 455–461.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900103
ChemCatChem
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Elia Calderini, Julia Wessel, Patrick
Schrepfer, Philipp Süss, Rainer
Wardenga, Anett Schallmey*
Page No. – Page No.
Selective Ring-Opening of Di-
substituted Epoxides Catalysed by
Halohydrin Dehalogenases
More than you think: An unexpected large number of halohydrin dehalogenases
(HHDHs) was found to convert non-terminal epoxide substrates to the
corresponding azidoalcohols. Some enzymes displayed even absolute
regioselectivity, enabling the preparative-scale synthesis of regioisomerically pure
products. This further expands the range of β-substituted alcohols that are
accessible by HHDH catalysis.
This article is protected by copyright. All rights reserved.