An Alcohol Dehydrogenase from the Short-Chain Dehydrogenase/Reductase Family of Enzymes for the Lactonization of Hexane-1,6-diol

Article abstract of DOI:10.1002/cbic.201800533

Biocatalytic production of lactones, and in particular ?-caprolactone (CL), have gained increasing interest as a greener route to polymer building blocks, especially through the use of Baeyer–Villiger monooxygenases (BVMOs). Despite several advances in the field, BVMOs, however, still suffer several practical limitations. Alcohol dehydrogenase (ADH)-mediated lactonization of diols in turn has received far less attention and very few enzymes have been identified for the conversion of diols to lactones, with horse-liver ADH (HLADH) remaining the catalyst of choice. Screening of a diverse panel of ADHs, AaSDR-1, a member of the short-chain dehydrogenase/reductase family, was found to produce ?-caprolactone from hexane-1,6-diol. Moreover, cofactor regeneration by an NADH oxidase eliminated the requirement of co-substrates, yielding water as the sole by-product. Despite lower turnover frequencies as compared to HLADH, higher selectivity was found for the production of CL, with HLADH forming significant amounts of 6-hydroxyhexanoic acid and adipic acid through aldehyde dehydrogenation/oxidation of the gem-diol intermediates. Also, CL yield were shown to be dependent on buffer choice, as structural elucidation of a Tris adduct confirmed the buffer amine to react with aliphatic aldehydes forming a Schiff-base intermediate which through further ADH oxidation, forms a tricyclic acetal product.

Full text of DOI:10.1002/cbic.201800533

Accepted Article
Title: An alcohol dehydrogenase from the short-chain dehydrogenases/
reductases family of enzymes for the lactonization of 1,6-
hexanediol
Authors: Choaro David Dithugoe, Jacqueline van Marwijk, Martha
Sophia Smit, and Diederik Johannes Opperman
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: ChemBioChem 10.1002/cbic.201800533
Link to VoR: http://dx.doi.org/10.1002/cbic.201800533
A Journal of
www.chembiochem.org

ChemBioChem
10.1002/cbic.201800533
FULL PAPER
An alcohol dehydrogenase from the short-chain dehydrogenases/
reductases family of enzymes for the lactonization of 1,6-
hexanediol
[a]
[a]
[a]
[a]
Choaro D. Dithugoe, Jacqueline van Marwijk, Martha S. Smit, and Diederik J. Opperman*
Abstract: Biocatalytic production of lactones, and in particular ε-
caprolactone (CL), have gained increasing interest as a greener route
to polymer building blocks, especially through the use of Baeyer-
Villiger monooxygenases (BVMOs). Despite several advances in the
field, BVMOs, however, still suffer several practical limitations. Alcohol
dehydrogenase (ADH) mediated lactonization of diols in turn has
received far less attention and very few enzymes have been identified
for the conversion of diols to lactones, with horse-liver ADH (HLADH)
remaining the catalyst of choice. Screening of a diverse panel of
of ketones, is the oxidative lactonization of diols utilizing alcohol
dehydrogenases (ADHs; Scheme 1)[11,12] or the laccase/TEMPO
system[13]
.
O
OH
O
OH
OH
O
ADHs,
AaSDR-1,
a
member
of
the
short-chain
ADH
ADH
dehydrogenase/reductase family, was found to produce ε-
caprolactone from 1,6-hexanediol. Moreover, cofactor regeneration
by an NADH oxidase eliminated the requirement of co-substrates,
yielding water as the sole by-product. Despite lower turnover
frequencies as compared to HLADH, higher selectivity was found for
the production of CL, with HLADH forming significant amounts of 6-
hydroxyhexanoic acid and adipic acid through aldehyde
dehydrogenation/oxidation of the gem-diol intermediates. Also, CL
yield were shown to be dependent on buffer choice, as structural
elucidation of a Tris-adduct confirmed the buffer amine to react with
aliphatic aldehydes forming a Schiff-base intermediate which through
further ADH oxidation, forms a tricyclic acetal product.
NAD(P)+ NAD(P)H
NAD(P)+ NAD(P)H
OH
O
Scheme 1. ADH mediated oxidative lactonization of 1,6-hexanediol to ε-
caprolactone.
ADHs catalyzes the interconversion of primary and secondary
alcohols to aldehydes and ketones respectively. ADHs are
currently divided into three classes based on the protein’s
monomer
dehydrogenases/reductases
dehydrogenases/reductases
chain
length[14,15]
(MDR),
(SDR)
:
medium-chain
short-chain
long-chain
Introduction
and
dehydrogenases/reductases (LDR). ADHs from the MDR
superfamily of enzymes, which include horse-liver alcohol
dehydrogenase from Equus caballus (HLADH) and yeast ADHs
Lactones are important building blocks in organics synthesis,
especially of polymers such us polycaprolactones and other
polyesters as well as precursors for the synthesis of polyamides
(
YADH)
from
Saccharomyces
cerevisiae
(ScADH1),
such as nylon[1–6]
.
In recent years, Baeyer-Villiger
2+
predominantly contain Zn within the active site. In contrast, the
SDR family of enzymes doesn’t contain any metal ions in their
active sites.
monooxgyenases (BVMOs) have been studied intensively as a
green alternative to current organometallic catalysts for BV
oxidation of ketones[7–10]. The practical application of this class of
enzymes are however generally limited due to their requirement
for reduced cofactors (NADPH) and thus requirement of co-
substrates for cofactor regeneration, instability and substrate or
product inhibition. An alternative to BVMO catalyzed oxygenation
ADHs (often termed ketoreductases/KREDs) have been
extensively studied and used for the production of chiral alcohols
through (dynamics) kinetic resolution, enantioconvergence,
reactions[16,17]
deracemization and stereoinversion
.
The
lactonization of diols using ADHs have in turn received less
attention. The earliest examples of ADH-catalyzed lactonization
of diols came from the research group of J. Bryan Jones in the
late 1970s. Although the oxidation of hydroxy aldehydes to
lactones via their hemiacetals were already known at this stage,
it was their investigation into the regioselectivity and possible
enantioselectivity of HLADH that demonstrated the usefulness of
ADHs in the lactonization of diols[18–21]. Much of the early work on
ADH mediated lactonization of diols were focused on the
synthesis of chiral synthons. Asymmetric oxidation of pro-chiral
[a]
Mr. C.D. Dithugoe, Dr. J. van Marwijk, Prof. M.S. Smit, Dr. D.J.
Opperman
Department of Biotechnology
University of the Free State
205 Nelson Mandela Drive, Bloemfontein, 9300 (South Africa)
E-mail: opperdj@ufs.ac.za
Supporting information for this article is given via a link at the end of
the document.
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800533
FULL PAPER
diols were shown to be stereoselective and the oxidation of
various bishydroxy cyclopentyl and cyclohexyl substrates
demonstrated the selectivity of HLADH to yield pure regiomeric
lactones. The production of enantiopure lactones were also
demonstrated through the asymmetrization of various acyclic and
cyclic meso-diols[22–25]. This enantiotopic oxidation was again
shown to be absolute, resulting in enantiopure lactones from
Table 1. Panel of alcohol dehydrogenases (ADHs) evaluated for oxidative
lactonization of 1,6-hexanediol to ε-caprolactone
ADH
Origin
Co-factor
Type
preference
NAD+
HLADH
Equus caballus
MDR (Zn)
MDR (Zn)
MDR (Zn)
various exo- and endo-bridged bicyclic diols[26,27]
TADH[36]
Thermus sp. ATN1
NAD
+
.
One of the major drawbacks of the early HLADH catalysed
reactions were inefficient cofactor regeneration, leading to various
TeSADH[37]
Thermoanaerobacter
NADP+
ethanolicus 39E
research groups investigating bacteria and yeast species for in-
vivo lactonization of various diols[28–30]. With advances in cofactor
MLADH[38]
Micrococcus luteus
WIUJH20
NAD
+
MDR (Zn)
MDR (Zn)
SDR
regeneration strategies, the use of purified ADHs has again
emerged as a feasible reaction as it alleviates unwanted side
product formation. Hollmann and co-workers demonstrated the
use of a laccase-mediator system for cofactor regeneration of
ADH mediated lactonization of various diols, where only
molecular oxygen is required and water is produced as the sole
NAD+
CAD (cinnamyl-
alcohol DH)[
Eucalyptus gunnii
39]
calA (coniferyl-
alcohol DH)[
Pseudomonas
HR199
sp.
NADP+
NADP+
NADP+
NAD+
40]
by-product[11] as well as photocatalytic regeneration of NAD+[31]
.
AcADH[41]
Acinetobacter
calcoaceticus GK2
MDR (Zn)
SDR
More recently, a novel convergent cascade was developed,
combining both BVMO and ADH strategies with no requirement
AaSDR-1
Aedes aegypti
for external cofactor regeneration enzymes[32–35]
.
farnesol DH)[42]
(
Despite the numerous advances made within the field, the
repertoire of available ADHs for the efficient lactonization of diols
unfortunately remains limited, with HLADH remaining the ADH of
choice. Here we explore a diverse set of ADHs for the
lactonization of 1,6-hexanediol as an alternative to HLADH.
ScADH1
Saccharomyces
cerevisiae S288C
MDR (Zn)
NADP+
ScADH6
AdADH1
MDR (Zn)
Alcanivorax dieselolei
n.d.
MDR (Zn)[a]
B-5
Results and Discussion
AdADH2
AdADH3
n.d.
n.d.
MDR (Zn)[a]
MDR (Zn)[a]
ADH panel selection
n.d. not determined, [a] Predicted based on sequence similarity
In an effort to improve on the current state of the art in ADH-
mediated lactonization of 1,6-HD, we screened a panel of ADHs,
diverse in structure/mechanism, cofactor dependence and origin.
Apart from ADHs already available in our laboratory, ADHs
described in literature as having the ability to accept aliphatic
alkanols and/or alkanals were selected, as well as ADHs from
organisms known to metabolize linear alkanes. These were
compared with already known ADHs that efficiently catalyze the
formation of CL from 1,6-HD and thus 13 distinct ADHs were
included in this study (Table 1).
HLADH, TeSADH and TADH have previously been reported for
their use in the lactonization of 1,6-HD[33]. Apart from its well-
known acetaldehyde and ethanol conversion, one of the best
studied ADHs, ADH1 from Saccharomyces cerevisiae, have also
been reported to accept longer chain alkanols, such as hexanol[43]
.
ADH6 from the same organism has also been shown to be active
[44]
towards both hexanol and hexanal . The three ADHs from
Alcanivorax dieselolei (AdADH1-3) were included as this
organism have been reported to have an extensive metabolism
for alkane degradation[45]
.
Similarly, various strains of
Acenitobacter strains have been described for their n-alkane
metabolism and to contain ADHs with the ability to oxidize
medium-chain alkanols[46,47]. These ADHs together with cinnamyl-
alcohol dehydrogenase (CAD) are all Zn-dependent ADHs from
the MDR-family of ADHs. We therefore decided to also include
ADHs from the SDR-family. The farnesol dehydrogenase
(
AaSDR-1) from Aedes aegypti is known to accept isoprenoid and
aliphatic alcohols including octanol, 2-decanol and citronellol[42].
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800533
FULL PAPER
Cloning and heterologous expression
regenerate
both
NAD+
and
NADP+[49]
.
The
SmNOX2_V193R_V194H mutant (Mut10) was shown to accept
both NADH+ and NADP+ with kinetic studies revealing similar
affinities and catalytic efficiencies for both cofactors. Purified
ADHs and SmNOX2_V193R_V194H were therefore used for
biotransformations at concentrations of 0.2 and 0.1 mg mL-
respectively. Despite CL yields improving significantly,
surprisingly, the unknown compound remained as the major
product in all the reactions except for AaSDR-1 (Figure 1).
The ADHs from S. cerevisiae and A. dieselolei were PCR
amplified from genomic DNA and sub-cloned to the pET28-b(+)
vector for heterologous expression in E. coli. Our in-house library
of ADHs, and those commercially synthesized for this study, were
similarly cloned to pET28-b(+) to allow for the addition of an N-
terminal hexa-histidine tag. With the exception of CAD and TADH,
all the ADHs expressed in relatively high quantities as soluble
proteins with molecular weights as expected (Table S1 and Figure
S1 in Supporting Information).
1
2
1
1
0
6
2
8
4
Biotransformations
The panel of recombinant ADHs were evaluated for their ability to
convert 1,6-HD to CL using cell-free extracts (CFE). Initial
screening reactions gave low yields (less than 10% conversions)
of 1,6-HD to CL, even with HLADH (Figure S2 in Supporting
Information). Similar to reports by other research groups, neither
the hydroxy-aldehyde nor its cyclic hemiacetal were detected[11,48].
In an effort to improve the reactions, the CFEs were
0
0
8
6
4
2
0
8
6
4
2
0
2
1
1
1
1
1
8h
24h
8h
24h
8h
24h
8h
24h
8h
24h
+
supplemented with additional cofactor [1 mM NAD(P) ] and the
HLADH
TADH
Caprolactone
TeSADH
Unknown
AaSDR1
AdADH3
concentration of the starting material increased to 20 mM. An
improvement in CL production was observed (Figure S3 in
Supporting Information) but only 4 of the ADHs yielded CL. In both
sets of reactions an unknown product was formed either as a
byproduct as seen with HLADH and TADH, or as the sole product
with ScADH1 and AdADHs. Only trace amounts of products were
observed with MLADH, CAD and calA. MLADH is a known
secondary alcohol dehydrogenase that can in all probability not
accept terminal alcohols. CAD (cinnamyl alcohol dehydrogenase)
and calA (coniferyl-alcohol dehydrogenase) are known primary
alcohol dehydrogenases but typically function on substrates
containing an aryl group.
8
h
24h
8h
24h
8h
24h
8h
24h
8h
24h
HLADH
TADH
TeSADH
AaSDR1
AdADH3
Figure 1. Screening of 5 purified ADHs for the synthesis of ε-caprolactone
(
bottom graph) from 1,6-hexanediol (top graph). Reaction conditions: 1 mL
GC-MS analysis (Figures S6 in Supporting Information) revealed
the unknown compound to have a molecular weight of at least 198
g.mol-1 but a positive identification could not be made against the
-1
-1
reaction volumes containing 0.2 mg mL
SmNOX2_V193R_V194H, NAD(P) = 1 mM, 1,6-hexanediol = 20 mM, buffer:
50 mM Tris-HCl (pH 8), 30°C, 200 rpm.
ADH, 0.1 mg mL
+
available mass spectral libraries such as the NIST database. As
AdADH3 yielded this unknown compound as the major product,
with near complete conversion of the 1,6-HD, the reaction was
up-scaled for isolation and identification using NMR.
Unfortunately the molecular structure of the unknown compound
could still not be elucidated. As these reactions were performed
using CFE, we assumed that either side-reactions with E. coli host
enzymes or small molecular weight metabolites were taking place
with the hydroxy-aldehyde products from 1,6-HD. We therefore
turned to purified enzymes for the biotransformations, with the 4
most promising ADHs selected from the initial screening for CL
production (HLADH, TADH, TeSADH, and AaSDR-1) as well as
AdADH3.
Identification of byproduct formation
As NMR analysis revealed the unknown compound to contain 10-
carbons, we turned our attention to a possible side reaction with
the Tris-buffer. Indeed, low molecular weight aldehydes have
[
50]
been shown to react with Tris in aqueous solutions . The free
amine groups of Tris at pH 8 could potentially nucleophilically add
to the carbonyl carbon of aldehydes formed from 1,6-HD to give
the hemiaminal followed by dehydration to a Schiff base. The use
of Tris buffers in ADH-mediated lactonization reactions have been
extensively reported without the mention of such a Tris-adduct,
although these studies were conducted at different pH values. We
therefore performed the Tris-buffered reactions at different pH
values (6 – 8). At these lower pH values, the amount of unknown
byproduct is significantly reduced (pH 7), and no byproduct could
+
To avoid the use of stoichiometric amounts of NAD(P) cofactor,
we employed an NADH-oxidase for cofactor regeneration. The
water forming Streptococcus mutans NADH oxidase 2 (SmNOX2)
is specific for NADH. Petschacher and co-workers however
engineered SmNOX2 as a universal regeneration system to
be detected at pH 6 (Figure 2). As Tris has a pK
, we assume that at these lower pH values, the amine group is
protonated and thus not a nucleophile. The lowering of the pH
a
of approximately
8
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800533
FULL PAPER
could also promote the cyclization of the hydroxy-aldehyde to the
lactol, however no significant improved CL production was
observed at these lower pH values with HLADH and indeed
decreased CL production with AaSDR-1 (Figure S7 in Supporting
Information).
groups of Tris on the carbon of the C-N double bond. This product
can undergo a second round of alcohol oxidation by the ADH,
resulting in an acetal tricyclic product after reaction of the
aldehyde group with the remaining free hydroxyl groups (Scheme
2). Conceivably this can also occur from the dialdehyde forming
the Schiff-base and then cyclizing. The proposed tricyclic Tris-
adduct was compared with the NMR data, and indeed, all carbons
and protons could be assigned (Figure S8 in Supporting
Information). The Tris-adduct however decreased with time,
without any additional product identified with GC analysis,
suggesting possible further polymerization.
To further understand and confirm our theory, buffers without
primary amines (HEPES and Na-phosphate, pH 8) were tested in
the biotransformation reactions using HLADH. Gratifyingly, not
only did the unknown byproduct decrease, CL production
increased significantly, with more than 10 mM CL formed after 24
h in phosphate buffer. Since the purified enzymes were in Tris-
buffer, small amounts of Tris were still introduced into the reaction
mixtures. Purification of the enzymes (HLADH, AaSDR-1 and
SmNOX2_193R194H) in sodium phosphate buffer completely
eliminated the formation of the Tris-adduct as byproduct, with
yields increasing to approximately 12 mM CL (Figure 2).
OH
HO
HO
Tris Buffer
Schiff base
N
OH
ADH
OH
OH
OH
2
-H O
HO
+
OH
O
NH
2
HO
24
20
16
12
8
4
0
OH
OH
HN
O
H
-H O
2
O
ADH
O
O
H
N
OH
OH
N
O
O
acetal
product
OH
Scheme 2. Proposed reaction mechanism of ADH mediated Tris-adduct
formation.
8
h
24h 8h 24h 8h 24h 8h 24h 8h 24h 8h 24h
2
1
1
1
1
1
0
8
6
4
2
0
8
6
4
2
0
Tris pH 6 Tris pH 7 Tris pH 8 HEPES pH 8 Phosphate Phosphate
Caprolactone / Tr Ui snkno wp Hn 8 / Tris
pH 8
When AdADH3 was again tested in the phosphate buffer, only
trace amounts of unknown were detected, however, a depletion
of the 1,6-HD was still observed. We speculated that the poor
extraction of the hydroxy-aldehyde was below our detection limits
and therefore acidified the BRM before extraction. 6-Hydroxy
hexanoic acid (6-HHA) was identified as the sole detectable
product. As these reactions were performed using purified
enzymes, we attribute this aldehyde dehydrogenase/oxidase
activity to the aldehyde “dismutase” activity reported for various
ADHs. In the absence of cofactor regeneration, ADHs have been
shown to both reduce and oxidize various aldehydes to their
corresponding alcohols and acids respectively in equimolar
amounts. Oxidation of the aldehyde occur via a gem-diol formed
in aqueous environments. Wuensch and co-workers have
however demonstrated that when using catalytic amounts of
8
h
24h 8h 24h 8h 24h 8h 24h 8h 24h 8h 24h
Tris pH 6 Tris pH 7 Tris pH 8 HEPES pH 8 Phosphate Phosphate
/
Tris pH 8 / Tris pH 8
Figure 2. Effect of pH and buffer composition on the synthesis of ε-caprolactone
(
1
bottom graph) from 1,6-hexanediol (top graph) by HLADH. Reaction conditions:
cofactors, different ADHs can give significantly different ratios of
alcohol and acids during time-course reactions[51]. Since our
reactions contained excess amounts of SmNOX to ensure
cofactor regeneration was not limiting the reactions, only NAD(P)+
is available thus permitting only the gem-diol (aldehyde) oxidation
reaction.
mL reaction volumes containing 0.2 mg mL-1 ADH, 0.1 mg mL
-1
+
SmNOX2_V193R_V194H, NAD(P) = 1 mM, 1,6-hexanediol = 20 mM, buffers:
5
0 mM Tris-HCl (pH 6 - 8), 50 mM HEPES or NaP
enzyme preparation, 50 mM NaP (pH 8) from enzyme preparation also in same
NaP buffer, 30°C, 200 rpm.
i
(pH 8) / Tris introduced with
i
i
As the Schiff base adduct is reversible and our NMR analysis
suggested a cyclic or bicyclic compound with ten carbons and no
free aldehyde or hydroxyl groups, we suggest that the Tris-adduct
cyclizes through a nucleophilic attack of one of the free hydroxyl
AaSDR-1 as an alternative to HLADH
To date, HLADH has served as the prototype ADH for the
lactonization of 1,6-HD to CL. Despite HLADH giving rapid
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800533
FULL PAPER
conversion of the 1,6-HD to CL (complete conversion after 8 h)
using purified enzyme in the phosphate buffer, the yields of CL
was only ca. 60%. Acidification of the HLADH reaction before
extraction revealed large amounts of 6-HHA. Although CL have
been shown to undergo hydrolysis to 6-HHA, the rates exceeded
that observed by Kara and co-workers[33] as well as our tested
conditions (Fig 9A in Supporting Information). As our preliminary
screening had revealed AaSDR-1 as a promising alternative to
HLADH, we compared the two ADHs using time-course reactions.
With a higher concentration of ADHs used, HLADH showed
complete conversion of the 1,6HD within 4 h, however CL yields
within the BRM were only approximately 7.5 mM, with an almost
equimolar amount of 6-HHA observed. Further reaction showed
the slow hydrolysis of CL and the rapid oxidation of 6-HHA to
adipic acid (AA) by HLADH (Figure 3A). Despite the lower
reaction rates of AaSDR-1, the same amount of CL was observed
after 48 h at only ca. 40% conversion, with substantially lower 6-
HHA concentrations and no AA formation (Figure 3B). The initial
rate of 6-HHA production suggests that the HLADH has a higher
affinity for the gem-diol compared to AaSDR-1. Reaction of both
the HLADH and AaSDR-1 with CL as starting material further
revealed HLADH to further oxidize the 6-HHA hydrolysis product
to AA (Figure S9B and C in Supporting Information).
Yields can potentially be increased using similar strategies as
employed in BVMO oxidations for the production of CL: hydrolysis
of CL to 6-HHA by the lipase Cal-B from Candida antarctica[52] or
the polymerization of CL to oligo-ε-CL by Cal-A from the same
organism[53]
.
Substrate scope
Lastly, the AaSDR-1 was evaluated for its ability to convert
aliphatic diols other than 1,6-HD, as well as for its ability for
desymmetrization reactions using the prochiral diol, 3-methyl-1,5-
pentanediol. Unlike HLADH where the turnover frequencies
(TOFs) increase with decreasing chain length, AaSDR-1 showed
minimal conversion of 1,4-butanediol and 1,5-pentanediol (<0.3
mM product formation) after 24 h. Product formation from 3-
methyl-1,5-pentanediol was similarly low (ca. 1.3 mM after 24 h),
but chiral analysis showed the same enantiospecificity with the
(S)-enantiomer of 4-methyloxane-2-one formed as the major
product. During time resolved reactions the lactol of 3-methyl-1,5-
pentanediol were only observed with HLADH, suggesting the
second ADH-catalyzed step (acceptance of a secondary alcohol)
is not the rate-limiting step for AaSDR-1, which could further
explain the low Tris-adduct concentration seen in previous
experiments.
30
25
20
15
10
5
0
A
1
,6-Hexanediol
Caprolactone
-Hydroxyhexanoic acid
Adipic Acid
Conclusions
6
AaSDR-1
dehydrogenase/reductase family of enzymes able to lactonize
,6-hexanediol to ε-caprolactone (CL). Despite its lower TOFs
is
an
ADH
from
the
short-chain
1
compared to HLADH, higher selectivity was found for the
production of CL, with HLADH forming significant amounts of the
overoxidized products 6-hydroxyhexanoic acid and adipic acid.
As these rates exceed the hydrolysis rates of CL, their formation
is most likely due to the ability of HLADH to also readily accept
the geminal diols of the aldehyde intermediates as substrates.
+
0
10
20
30
40
50
Furthermore, as AaSDR-1 is NADP dependent, its use in
Time / h
convergent cascade strategies with NADPH dependent BVMOs
could facilitate and improve on these redox balanced, co-
substrate free reactions.
3
2
2
1
1
0
5
0
5
0
5
0
B
1
,6-Hexanediol
Caprolactone
-Hydroxyhexanoic acid
6
Lastly, choice of buffer should be carefully considered during
ADH-mediated lactonization reactions, as many commonly used
buffers such as Tris contain primary amines which under basic
conditions can rapidly react with the aldehyde intermediates,
yielding carbinolamines which dehydrate to a Schiff base. These
side-products can further polymerize, leading to significant losses
in product formation. During the preparation of this manuscript,
Kara and co-workers elegantly exploited this property for the
production of lactams[54]. Oxidative lactamization was achieved
through ADH catalyzed oxidation of amino alcohols with
subsequent intramolecular cyclization to the hemiaminal followed
by a second ADH oxidation to yield various lactams.
0
10
20
30
40
50
Time / h
Figure 3. Time-resolved conversion of 1,6-hexanediol by HLADH (A) and
AaSDR-1 (B). Reaction conditions: 1 mL reaction volumes containing 0.5 mg
-
1
-1
+
mL ADH, 0.25 mg mL SmNOX2_V193R_V194H, NAD(P) = 1 mM, 1,6-
hexanediol = 25 mM, buffer: 50 mM NaP (pH 8), 30°C, 200 rpm.
i
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800533
FULL PAPER
(50 mM Tris-HCl, 0.5 M NaCl and 20 mM Imidazole, pH 7.4) where after
Experimental Section
the clear lysates obtained from ultracentrifugation were loaded onto the
columns and allowed to bind. Unbound proteins were removed by washing
with 20 ml of equilibration buffer with bound proteins eluted using 3 ml
elution buffer (50 mM Tris-HCl, 0.5 M NaCl and 0.5 M Imidazole, pH 7.4).
Alternatively, large scale purifications were performed using 5 mL HisTrap
FF Ni-affinity columns (GE Healthcare) with elution using a linear gradient
of increasing imidazole concentration. The elution fraction were
concentrated to approximately 2 mL using 30 000 Da MWCO ultrafiltration
spin columns (Millipore) and desalted using gravity flow PD-10 columns
Growth of micro-organisms
Saccharomyces cerevisiae S288C was cultured in YPD broth (10 g L-1
yeast extract, 20 g L-1 peptone and 20 g L-1 dextrose) at 25°C with shaking.
Alcanivorax dieselolei B-5 was obtained from the Deutsche Sammlung von
Mikroorganismen und Zellkulturen (DSMZ) and cultured in Difco Marine
broth 2216 at 28°C with shaking. E. coli strains were routinely cultured in
LB medium (10 g L-1 tryptone, 5 g L-1 yeast extract and 5 g L-1 NaCl) at
(GE Healthcare) equilibrated with 10 mM Tris-HCl (pH 8) containing 20
37°C with shaking.
mM NaCl. Protein concentrations were determined using the
ThermoScientific Pierce bicinchoninic acid (BCA) Protein Assay Kit
according to the manufacturer’s instructions, with bovine serum albumin
Construction of expression plasmids
(BSA) as standard.
The ADHs from S. cerevisiae (ScADH1 and ScADH6) and A. dieselolei
(
(
AdADH1, AdADH2 and AdADH3) were PCR amplified from genomic DNA
gDNA). gDNA was isolated from S. cerevisiae according to the method by
Biotransformations
Harju and co-workers[55] and from A. dieselolei using the ZR
Crude cell free extracts (CFE) of the 13 ADHs, as well as CFE from E. coli
transformed with empty pET28b(+) as a negative control, were used for
biotransformation reactions. The 1 mL biotransformation reactions were
performed in capped 40 mL amber vials and contained 1,6-hexanediol (10
mM), 500 µL CFE and 500 µL of 50 mM Tris-HCl buffer (pH 8). Vials were
incubated at 30ºC with shaking at 200 rpm. Samples were taken at 8 h and
Fungal/Bacterial DNA MicroPrep kit (Zymoresearch). PCR reaction
mixtures (50 µl) consisted of 1X KOD Hot Start Polymerase buffer, MgSO
4
(1.5 mM), deoxynucleoside triphosphates (0.2 mM each), KOD Hot Start
DNA polymerase (1 U), gDNA (25 ng), and both the forward and the
reverse primers (Table S2 in Supporting Information; 0.3 µM each). The
PCR was performed with an initial denaturing step at 95°C for 2 min,
followed by 30 cycles of denaturing at 95°C (20 s), annealing at 53-62°C
24 h to evaluate production of ε-caprolactone. Samples were extracted
with a 1:1 ratio of ethyl acetate containing 2 mM 1-undecanol as internal
standard. For quantification of 6-hydroxyhexanoic acid and adipic acid, the
BRM was acidified using 50 µL 5M HCl before extracting with a 1:1 ratio
of ethyl acetate containing 2 mM 1-undecanol as internal standard.
(Table S1; 10 s), and elongation at 70°C (40 s), with a final extension at
70°C for 10 min. The amplicons were excised from an agarose gel and
purified using a GeneJET Gel Extraction Kit (Thermo Scientific). Products
were phosphorylated and ligated into pSMART (Lucigen) and transformed
into E. coli TOP10 competent cells (Invitrogen). Positive transformants
were selected for on LB-agar plates containing 30 µg.ml–1 kanamycin.
Plasmid DNA was isolated using a GeneJET Plasmid miniprep Kit (Thermo
Scientific) and inserts verified through DNA sequencing. For expression,
the ORFs were sub-cloned in pET28b(+) (Novagen) using NdeI and XhoI.
TADH from Thermus sp. ATN1 was kindly provided by Dr. Frank Hollmann
Products were analysed and quantified against authentic standards using
GC-FID with separation on a FactorFour VF-5ms column (60 m x 0.32 mm
x
0.28 µm, Agilent). 6-Hydroxyhexanoic acid concentrations were
estimated based on the response from adipic acid, and the unknown
adduct based on that of ε-caprolactone. Samples with unknown products
were subjected to GC-MS analysis using the same separation. Chiral
analysis were performed on a Chiraldex G-TA column (30 m x 0.25 mm x
(TUDelft, Netherlands) in pET22 where after it was sub-cloned to
pET28b(+) using NdeI and NcoI. All other ADHs were synthesized without
codon optimization by GenScript with sub-cloning to pET28b(+) using NdeI
and XhoI (EcoRI for MLADH). TeSADH was provided in pUC57 and sub-
cloned to pET28b(+) using NdeI and AvrII (NheI compatible sticky ends on
plasmid). SmNOX2_V193R_V194H was likewise synthesized and cloned
into pET28b(+) by GenScript using NdeI and XhoI.
0.12µm, Astec) using the reported products from HLADH for identification.
Biotransformations using purified enzymes were similarly performed with
reaction mixtures typically containing (unless otherwise stated) 20 mM 1,6-
hexanediol, 0.2 mg mL-1 ADH and 0.1 mg mL-1 SmNOX2_V193R_V194H
supplemented with 1 mM NAD+ (HLADH and TADH) or 1 mM NADP+
(
5
TeSADH and AaSDR-1). In later reactions, the buffer was changed from
0 mM Tris-HCl to 50 mM sodium phosphate buffer (pH 8).
Heterologous expression and Cell Free Extract (CFE) preparation
The pET28 constructs were transformed into E. coli BL21-Gold(DE3) cells
NMR conformation of Tris-adduct
(Agilent Technologies), and expression performed in auto-induction media
ZYP5052) containing 30 mg L-1 kanamycin at 25°C for 24 h (200 rpm).
(
Conversion of 1,6-hexanediol by purified AdADH3 was monitored by GC-
Cells were harvested by centrifugation (7 000 xg, 10 min, 4⁰C) and
suspended in 50 mM Tris-HCl buffer pH 8 (1 g wet weight per 5 mL buffer).
Cells were broken using a Constant cell disruption system (Constant
Systems) at 30 kPsi. CFEs were obtained by removing unbroken cells and
debris through centrifugation (7000 xg, 10 min, 4⁰C). Protein expression
was evaluated on SDS-PAGE along with PageRuler Prestained Protein
Ladder (ThermoScientific) and stained with Coomassie brilliant blue R-250.
(MS) until complete conversion was observed. Samples were extracted
using ethyl acetate and dried under vacuum where after it was dissolved
in deuterated acetone or chloroform for NMR analysis. NMR spectra were
recorded on an Avance II Bruker (600 MHz) spectrometer.
Acknowledgements
Protein purification
The authors gratefully acknowledge Mr Sarel Marais for GC-
FID/MS analysis and Dr Linette Twigge (Department of Chemistry,
UFS) for NMR analysis. This research was supported through a
Central Research Fund grant from the Faculty of Natural and
Agricultural Sciences (UFS). Mr C.D. Dithugoe thank the National
CFEs of the E. coli expressed HLADH, TADH, TeSADH, AaSDR-1,
AdADH3 and SmNOX2_V193R_V194H proteins were subjected to
ultracentrifugation (100 000 xg, 90 min, 4°C). Proteins were purified using
Ni-affinity chromatography using gravity-flow Ni-NTA Superflow columns
(Qiagen). The columns were equilibrated using 10 mL equilibration buffer
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800533
FULL PAPER
2
521–2526.
Research Foundation (NRF, South Africa) for financial support
during his M.Sc.
[
[
27]
28]
J. B. Jones, C. J. Francis, Can. J. Chem. 1984, 62, 2578–2582.
R. N. Patel, M. Liu, A. Banerjee, J. K. Thottathil, J. Kloss, L. J.
Szarka, Enzyme Microb. Technol. 1992, 14, 778–784.
F. Boratyński, E. Szczepańska, J. Pannek, T. Olejniczak, Catalysts
2015, 5, 2068–2084.
[29]
Keywords: ε-caprolactone • diols • lactones • alcohol
[
[
30]
31]
H. Ohta, H. Tetsukawa, Agric. Biol. Chem. 1981, 45, 1895–1896.
M. Rauch, S. Schmidt, I. W. C. E. Arends, K. Oppelt, S. Kara, F.
Hollmann, Green Chem. 2017, 19, 376–379.
dehydrogenase • short-chain dehydrogenase/reductase
[
1]
2]
M. A. Hillmyer, W. B. Tolman, Acc. Chem. Res. 2014, 47, 2390–
396.
Y. Jiang, K. Loos, Polymers (Basel). 2016, 8, DOI
0.3390/polym8070243.
C. Zhang, Biodegrad. Polyesters 2015, 1–24.
G. A. R. Nobes, R. J. Kazlauskas, R. H. Marchessault,
Macromolecules 1996, 29, 4829–4833.
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
A. Bornadel, R. Hatti-Kaul, F. Hollmann, S. Kara, Tetrahedron 2016,
72, 7222–7228.
A. Bornadel, R. Hatti-Kaul, F. Hollmann, S. Kara, ChemCatChem
2015, 7, 2442–2445.
L. Huang, E. Romero, A. K. Ressmann, F. Rudroff, F. Hollmann, M.
W. Fraaije, S. Kara, Adv. Synth. Catal. 2017, 359, 2142–2148.
C. Scherkus, S. Schmidt, U. T. Bornscheuer, H. Gröger, S. Kara, A.
Liese, Biotechnol. Bioeng. 2017, 114, 1215–1221.
V. Höllrigl, F. Hollmann, A. C. Kleeb, K. Buehler, A. Schmid, Appl.
Microbiol. Biotechnol. 2008, 81, 263–73.
D. S. Burdette, C. Vieille, J. G. Zeikus, Biochem. J. 1996, 316, 115–
122.
J.-K. Huang, J. K. Park, B. R. Dhungana, N. D. Youngblut, C.-T. Lin,
L. Wen, FASEB J. 2010, 24, 835.5-835.5.
J. Grima-Pettenati, C. Feuillet, D. Goffner, G. Borderies, A. Boudet,
Plant Mol. Biol. 1993, 21, 1085–1095.
J. Overhage, A. Steinbüchel, H. Priefert, Appl. Environ. Microbiol.
2002, 68, 4315–4321.
2
[
1
[
[
3]
4]
[
[
[
[
[
[
5]
J.-P. Lange, J. Z. Vestering, R. J. Haan, Chem. Commun. 2007,
3
488.
T. Moore, R. Adhikari, P. Gunatillake, Biomaterials 2005, 26, 3771–
782.
6]
3
7]
K. Balke, M. Kadow, H. Mallin, S. Sass, U. T. Bornscheuer, S. Saß,
U. T. Bornscheuer, Org. Biomol. Chem. 2012, 10, 6249–6265.
Z.-G. Zhang, L. P. Parra, M. T. Reetz, Chemistry (Easton). 2012,
8]
1
8, 10160–72.
H. Leisch, K. Morley, P. C. K. Lau, Chem. Rev. 2011, 111, 4165–
22.
9]
2
10]
M. Bučko, P. Gemeiner, A. Schenkmayerová, T. Krajčovič, F.
Rudroff, M. D. Mihovilovič, Appl. Microbiol. Biotechnol. 2016, 100,
M. Balseiro-Romero, P. Gkorezis, P. S. Kidd, J. Van Hamme, N.
Weyens, C. Monterroso, J. Vangronsveld, Int. J. Phytoremediation
2017, 19, 955–963.
6585–6599.
[11]
S. Kara, D. Spickermann, J. H. Schrittwieser, A. Weckbecker, C.
Leggewie, I. W. C. E. Arends, F. Hollmann, ACS Catal. 2013, 3,
[42]
J. G. Mayoral, M. Nouzova, A. Navare, F. G. Noriega, Proc. Natl.
Acad. Sci. 2009, 106, 21091–21096.
2436–2439.
[
12]
13]
A. Díaz-Rodríguez, J. Iglesias-Fernández, C. Rovira, V. Gotor-
Fernández, ChemCatChem 2014, 6, 977–980.
A. Díaz-Rodríguez, I. Lavandera, S. Kanbak-Aksu, R. A. Sheldon,
V. Gotor, V. Gotor-Fernández, Adv. Synth. Catal. 2012, 354, 3405–
[43]
[44]
W. Schöpp, H. Aurich, Biochem. J. 1976, 157, 15–22.
C. LARROY, M. R. FERNÁNDEZ, E. GONZÁLEZ, X. PARÉS, J. A.
BIOSCA, Biochem. J. 2002, 361, 163–172.
W. Wang, Z. Shao, Nat. Commun. 2014, 5, 5755.
a Tani, Y. Sakai, T. Ishige, N. Kato, Appl. Environ. Microbiol. 2000,
66, 5231–5.
M. R. Wales, C. A. Fewson, Microbiology 1994, 140, 173–183.
F. Boratyński, G. Kiełbowicz, C. Wawrzeńczyk, J. Mol. Catal. B
Enzym. 2010, 65, 30–36.
B. Petschacher, N. Staunig, M. Müller, M. Schürmann, D. Mink, S.
De Wildeman, K. Gruber, A. Glieder, Comput. Struct. Biotechnol. J.
2014, 9, e201402005.
W. A. Bubb, H. A. Berthon, P. W. Kuchel, Bioorg. Chem. 1995, 23,
119–130.
C. Wuensch, H. Lechner, S. M. Glueck, K. Zangger, M. Hall, K.
Faber, ChemCatChem 2013, 5, 1744–1748.
C. Scherkus, S. Schmidt, U. T. Bornscheuer, H. Gröger, S. Kara, A.
Liese, ChemCatChem 2016, 8, 3446–3452.
[
[45]
[46]
3408.
[14]
H. Jörnvall, J. Hedlund, T. Bergman, Y. Kallberg, E. Cederlund, B.
Persson, Chem. Biol. Interact. 2013, 202, 91–96.
B. Persson, Y. Kallberg, Chem. Biol. Interact. 2013, 202, 111–115.
H. Gröger, W. Hummel, S. Borchert, M. Kraußer, in Enzym. Catal.
Org. Synth., Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim,
Germany, 2012, pp. 1035–1110.
F. Hollmann, K. Bühler, B. Bühler, in Enzym. Catal. Org. Synth.,
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany, 2012,
pp. 1325–1437.
A. J. Irwin, J. B. Jones, J. Am. Chem. Soc. 1977, 99, 556–561.
J. B. Jones, H. B. Goodbrand, Can. J. Chem. 1977, 55, 2685–2691.
J. B. Jones, K. P. Lok, Can. J. Chem. 1979, 57, 1025–1032.
A. J. Irwin, J. B. Jones, J. Am. Chem. Soc. 1977, 99, 1625–1630.
I. J. Jakovac, G. Ng, K. P. Lok, J. B. Jones, J. Chem. Soc. Chem.
Commun. 1980, 515.
[47]
[48]
[
[
15]
16]
[49]
[17]
[50]
[51]
[52]
[53]
[18]
[19]
[20]
[21]
[22]
S. Schmidt, C. Scherkus, J. Muschiol, U. Menyes, T. Winkler, W.
Hummel, H. Gröger, A. Liese, H.-G. Herz, U. T. Bornscheuer,
Angew. Chemie Int. Ed. 2015, 54, 2784–2787.
L. Huang, G. V. Sayoga, F. Hollmann, S. Kara, ACS Catal. 2018, 8,
8680–8684.
[23]
[24]
[25]
[26]
I. J. Jakovac, H. B. Goodbrand, K. P. Lok, J. B. Jones, J. Am.
Chem. Soc. 1982, 104, 4659–4665.
A. J. Bridges, P. S. Raman, G. S. Y. Ng, J. B. Jones, J. Am. Chem.
Soc. 1984, 106, 1461–1467.
[54]
[55]
S. Harju, H. Fedosyuk, K. R. Peterson, BMC Biotechnol. 2004, 4, 8.
G. S.Y. Ng, L.-C. Yuan, I. J. Jakovac, J. Bryan Jones, Tetrahedron
1984, 40, 1235–1243.
K. P. Lok, I. J. Jakovac, J. B. Jones, J. Am. Chem. Soc. 1985, 107,
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800533
FULL PAPER
FULL PAPER
A short-chain
Choaro D. Dithugoe, Jacqueline van
Marwijk, Martha S. Smit, and Diederik J.
Opperman*
dehydrogenase/reductase catalyses
the oxidative lactonization of 1,6-
hexanediol to ε-caprolactone, with
only water as by-product when using
NADH-oxidase for co-factor
regeneration. ε-Caprolactone yields
are significantly affected by buffers
containing primary amines, resulting in
irreversible adducts with aldehyde
intermediates.
Page No. – Page No.
An alcohol dehydrogenase from the
short-chain dehydrogenases/
reductases family of enzymes for the
lactonization of 1,6-hexanediol
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.