Rerouting and Improving Dauc-8-en-11-ol Synthase from Streptomyces venezuelae to a High Yielding Biocatalyst

Article abstract of DOI:10.1002/chem.202100962

The dauc-8-en-11-ol synthase from Streptomyces venezuelae was investigated for its catalytic activity towards alternative terpene precursors, specifically designed to enable new cyclisation pathways. Exchange of aromatic amino acid residues at the enzyme surface by site-directed mutagenesis led to a 4-fold increase of the yield in preparative scale incubations, which likely results from an increased enzyme stability instead of improved enzyme kinetics.

Full text of DOI:10.1002/chem.202100962

Full Paper
doi.org/10.1002/chem.202100962
Chemistry—A European Journal
Rerouting and Improving Dauc-8-en-11-ol Synthase from
Streptomyces venezuelae to a High Yielding Biocatalyst
Lukas Lauterbach+,[a] Anwei Hou+,[a] and Jeroen S. Dickschat*[a]
Dedicated to Prof. Henning Hopf on the occasion of his 80th birthday
Abstract: The dauc-8-en-11-ol synthase from Streptomyces
venezuelae was investigated for its catalytic activity towards
alternative terpene precursors, specifically designed to enable
new cyclisation pathways. Exchange of aromatic amino acid
residues at the enzyme surface by site-directed mutagenesis
led to a 4-fold increase of the yield in preparative scale
incubations, which likely results from an increased enzyme
stability instead of improved enzyme kinetics.
Introduction
cyclisation. This can open new reaction pathways, because in
the methylated substrates cationic charges can be stabilised at
carbons whose analogous positions in canonical substrates
would yield secondary cations. Two natural examples include
the methylation of GPP to 2-methyl-GPP (1) followed by its
cyclisation to 2-methylisoborneol (2) by 2-methylisoborneol
synthase (2-MIBS, Scheme 1A),[5–7] and the methylation-induced
cyclisation of FPP to pre-sodorifen diphosphate (3) by SodC and
further cyclisation to sodorifen (4) by the sodorifen cyclase
SodD (Scheme 1B).[8,9] In a recent study, Kirschning and co-
workers have converted non-natural FPP analogues with altered
methylation patterns such as 5, leading to new reaction
pathways. Using the presilphiperfolan-8β-ol synthase Bot2 from
Botrytis cinerea,[10] “iso-germacrene A” (6) was obtained that can,
like germacrene A, react by Cope rearrangement to “iso-β-
elemene” (7, Scheme 1C).[11]
Terpenes represent one of the structurally most diverse classes
of natural products,[1,2] although they all originate basically from
only two isomeric precursor molecules. Wallach’s visionary
proposal of
a general terpene ensemble by head-to-tail
connection of isoprene units helped tremendously in their
structure elucidation especially at his time.[3] The biochemical
background for the biogenetic isoprene rule rests in the
terpene monomers dimethylallyl (DMAPP) and isopentenyl
diphosphate (IPP), reactive isoprene derivatives, that are the
precursors of the oligomeric terpene precursors geranyl (GPP),
farnesyl (FPP), geranylgeranyl (GGPP) and geranylfarnesyl
diphosphate (GFPP) by prenyltransferases.[4] Their cyclisations
by terpene synthases proceed by abstraction of the diphos-
phate group, yielding a reactive allyl cation that can enter a
cascade reaction involving typical carbocation chemistry with
cyclisations by intramolecular attack of an olefin to a cationic
centre, hydride or proton migrations, Wagner-Meerwein rear-
rangements (WMR), eventually water quenching (in case of
terpene alcohols) and terminal deprotonation. The terpene
cyclisation cascades usually proceed along a series of stabilised
tertiary or allylic cations, demonstrating the importance of the
methyl branches in the oligoprenyl diphosphate precursors for
efficient enzymatic turnover. Today more and more examples
are known that escape the logic of Wallach’s isoprene rule, for
example by methylation of the terpene precursor prior to
[a] L. Lauterbach,+ Dr. A. Hou,+ Prof. Dr. J. S. Dickschat
Kekulé-Institut für Organische Chemie und Biochemie
Rheinische Friedrich-Wilhelms Universität Bonn
Gerhard-Domagk-Str. 1, 53121 Bonn (Germany)
E-mail: dickschat@uni-bonn.de
[+] These authors contributed equally to this work.
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/chem.202100962
© 2021 The Authors. Chemistry - A European Journal published by Wiley-
VCH GmbH. This is an open access article under the terms of the Creative
Commons Attribution License, which permits use, distribution and re-
production in any medium, provided the original work is properly cited.
Scheme 1. Biosynthesis of A) 2-methylisoborneol (2), and B) sodorifen (4).
Black dots indicate Me groups from methylations by SAM. C) Enzymatic
conversion of FPP analogue 5 with inverted isoprene units (bold) by Bot2.
Chem. Eur. J. 2021, 27, 1–8
1
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100962
Chemistry—A European Journal
In an attempt to systematically expand this work we now
report on the conversion of further FPP analogues with altered
methylation pattern that were designed to change their
cyclisation modes in terpene synthase catalysed reactions.
Dauc-8-en-11-ol synthase from Streptomyces venezuelae ATCC
10712 (DcS, introduced as “isodauc-8-en-11-ol synthase, IdS” in
our previous study, cf. Supporting Information Figure S1)[12] was
selected as a catalyst, because this enzyme has a high catalytic
efficiency in the conversion of FPP. The yield in preparative
scale incubations of the wildtype enzyme was increased in this
study to more than 4-fold through site-directed mutagenesis
and the newly obtained enzyme variant was used for the
enzymatic preparation of non-natural sesquiterpene analogues.
Results and Discussion
The cyclisation mechanism towards dauc-8-en-11-ol (8) starts
with isomerisation of FPP to nerolidyl diphosphate (NPP) and
proceeds with a likely concerted 1,7-6,10-cyclisation to A,
avoiding an intermediate secondary cation, followed by the
addition of water (Scheme 2A). Substrate analogues with a
methylation pattern different to that in FPP could lead to
specifically altered cyclisation modes, e.g. 10-methyl-FPP (9)
may not (only) react through the NPP analogue 10 in a 1,7-6,10-
cyclisation, but could (also) react by 1,7-6,11-cyclisation to B in
which the positive charge would be stabilised as tertiary cation
(Scheme 2B). On the contrary, the situation for 13-desmethyl-
FPP (11) would be less clear and thus particularly interesting to
investigate, as the natural 1,7-6,10- and the alternative 1,7-6,11-
cyclisation mode for DcS catalysis would both lead to a
secondary cation through collapse of the transition state shown
in C; however, this hypothetical secondary cation may be a
transient species attacked in a concerted reaction by the active
site water. Also of interest is 13-desmethyl-10-methyl-FPP (13)
for which a 1,7-6,11-cyclisation to the tertiary cation D may be
preferred over the 1,7-6,10-cyclisation to a secondary cation.
Shifting the 10,11-double bond of FPP to an 11,12-double bond
as in 15 could allow for a 1,7-6,12-cyclisation to E, which could
also be realised by ketone 16 that may cyclise to F.
Scheme 2. Dauc-8-en-11-ol synthase DcS. A) Cyclisation mechanism from
FPP to 8. B) Rationale for the design of substrate analogues based on
hypothetical alternative cyclisation modes. Structural modifications in
comparison to FPP are highlighted by red boxes.
a methyl group at C10 as in 9 allows this reaction to proceed
more efficiently towards formation of the methylated widdrane
19. The formation of 19 can be explained from the NPP
analogue 10 adopting a conformation (endo-10) that reflects
that of NPP in the cyclisation to 8. However, the configuration
at C4 of 20 is only explainable from exo-10. It is possible that
the additional methyl group increases the substrate’s steric bulk
which may enforce a conformational flip to exo-10 in the side
reaction to 20, while the main product 19 still follows the
ordinary pathway through endo-10. But 19 can also be
explained from exo-10, which would mean that the main and
the side product follow a common reaction trajectory, only in
this case a non-concerted mechanism via the previously
proposed cation B should be considered in which the cation
can be attacked by water from the Si face. This face is oriented
towards the C6=C7 double bond and is not directly accessible
in exo-10, thus a conformational rearrangement in B is required.
The absolute configurations of 19 and 20 were investigated
through stereoselective deuteration, introducing artificial ster-
eogenic centres at deuterated carbons of known configuration.
To investigate these hypotheses all FPP analogues were
synthesised for enzyme incubations with DcS. The conversion of
9 (for synthesis cf. Supporting Information Scheme S1, Figur-
es S2–S4) with DcS resulted in the formation of two C16 alcohols
(Supporting Information Figure S5). Both compounds were
isolated and their structures were elucidated by NMR (Support-
ing Information Tables S1 and S2, Figures S6–S20) as the main
product 3-methylwiddr-8-en-3-ol (19) and the minor product 4-
epi-4-methyldauc-8-en-11-ol (20, Scheme 3A). Widdranes are
rarely observed in Nature and only realised by
a few
representatives including widdrol (21) and its epoxide 22 from
Widdringtonia juniperoides[13] and ent-widdradiene (23) from
Cupressus macrocarpa (Scheme 3B).[14] The widdrane skeleton is
likely difficult to access by FPP cyclisation as it requires transient
cationic charges at the secondary carbons C6 and C10 (a type II
cyclisation with protonation at C10 could be a reasonable
alternative to explain widdrane biosynthesis). The installation of
Chem. Eur. J. 2021, 27, 1–8
www.chemeurj.org
2
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100962
Chemistry—A European Journal
The FPP isomer 13-desmethyl-10-methyl-FPP (13) can be
prepared enzymatically from tiglyl diphosphate (25, synthesised
as shown in Supporting Information Scheme S3, Figures S38–
S40) and IPP through FPPS catalysis. This reaction proceeds
with similarly high efficiency as for the conversion of the native
substrates DMAPP and IPP into FPP, which was demonstrated
by treatment of the products with calf intestinal phosphatase
(CIP, Scheme 5) and GC/MS analysis of the product 26 in
comparison to farnesol (Supporting Information Figure S41).
Further conversion with DcS surprisingly resulted in the
formation of the FPP-derived main product 8, besides smaller
amounts of other products not observed with FPP alone
(Supporting Information Figure S42). A deep investigation of
this reaction demonstrated that DcS coeluted with a very small
amount of E. coli IDI[18] from the Ni2+-NTA column (Supporting
Information Figure S43), requiring further enzyme purification
by FPLC to remove it and to suppress the conversion of IPP into
8 (Supporting Information Figure S44). After optimisation of the
reaction conditions, DcS showed only a poor conversion of 13
into a complex product mixture.
Scheme 3. A) Cyclisation mechanism for the conversion of 9 with DcS. B)
Structures of known widdranes.
Compounds 15 and 16 were synthesised through
a
common approach (Supporting Information Schemes S4 and
S5, Figures S45–S50). The incubation of 15 with DcS resulted in
the formation of three hydrocarbons (27–29, Supporting
Information Figure S51) which were isolated and subjected to
NMR for structure elucidation (Supporting Information Ta-
bles S5–S7, Figures S52–S73). Interestingly, none of the com-
pounds contained the two annulated seven-membered rings as
hypothesised for E in Scheme 2. Instead, 27 (tenuifola-2,10-
diene) and 28 (tenuifola-2,11-diene) showed a rare bicyclic
carbon skeleton that is only represented in Nature by the
structures of tenuifolene (30) and ar-tenuifolene (31), two
sesquiterpenes of unknown absolute configuration from the
African sandalwood Osyris tenuifolia,[19] while 29 (iso-β-sesqui-
phellandrene) is similar to the widespread natural product β-
sesquiphellandrene (32) that was first isolated from ginger
(Schemes 6A and B).[20] The formation of 29 proceeds through
isomerisation to the NPP isomer 17, followed by 1,6-cyclisation
to cation G, a 1,3-hydride shift to I and deprotonation from C15.
Compounds 27 and 28 require a 1,6-7,12-cyclisation to the
tertiary cation H and deprotonation from C10 or C12. Based on
DFT calculations Hong and Tantillo have suggested a mecha-
nism for the formation of natural 30 through the bisabolyl
cation (J) involving a 1,5-proton shift to K, another 1,7-proton
Additional 13C-labellings at these stereogenic centres allow for a
sensitive product analysis by HSQC. Conversion of 2-methyl-
DMAPP (S4, Supporting Information Scheme S1, Figures S21–
S23) and (R)- or (S)-(1-13C,1-2H)IPP[15] with FPPS from Streptomy-
ces coelicolor A3[16] and DcS (Supporting Information Figure S24)
established the absolute configurations of 19 and 20 as shown
in Scheme 3, that were confirmed by analogous use of (E)- and
(Z)-(4-13C,4-2H)IPP (Supporting Information Figure S25).[17]
Conversion of 11 (for synthesis cf. Supporting Information
Scheme S2, Figures S26–S28) with DcS yielded one main and
two side products (Supporting Information Figure S29). The
major compound was isolated and its structure elucidated by
NMR spectroscopy (Supporting Information Table S4, Figur-
es S30–S37), resulting in the structure of nor-widdr-8-en-4-ol
(24, Scheme 4). This experiment demonstrated that the inter-
mediate NPP analogue 12 preferentially reacts in a 1,7-6,11-
cyclisation with formation of a 7–6 bicyclic skeleton. The
absolute configuration was tentatively assigned as shown in
Scheme 4 based on the assumption that 12 adopts a similar
conformation as exo-10 in the cyclisation to 20, and on its
optical rotation (½a�2D0 = +28.4 (c 0.12, C6D6)) that presents the
same sign as for the compounds 8 (½a�2D2 = +19.4 (c 0.505,
C6D6)),[12] 19 (½a�D20 = +194.0 (c 0.1, C6D6)), and 20 (½a�2D0 = +4.8
(c 0.13, C6D6)).
shift to
G and cyclisation to H (Scheme 6C), the same
intermediates as proposed here for the cyclisation of 15, and
final deprotonation.[21] The natural products 30 and 31 then
Scheme 5. Conversion of 25 into the FPP isomer 13 and dephosphorylation
with CIP to 26.
Scheme 4. Cyclisation mechanism for the conversion of 11 with DcS.
Chem. Eur. J. 2021, 27, 1–8
www.chemeurj.org
3
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100962
Chemistry—A European Journal
Scheme 7. Conversion of methyl ketone 16 by DcS. A) Proposed catalytic
mechanism, B) structures of all four stereoisomers of α-bisabolol.
particular problem, because in all five compounds the neigh-
bouring stereogenic centres at C6 and C7 are connected by a
single bond around which free rotation is possible, which
hampers NOE based assignments. Assuming R configuration for
the NPP analogues 17 and 18, and a similar conformational fold
with C1 in the backside, C6=C7 in the middle and C11=C12 in
the front, as required for the natural intermediate (R)-NPP to
explain the stereochemistry of 8 (Scheme 2A), the stereo-
structures as shown in Schemes 6 and 7 should be expected.
To address this problem experimentally, the GPP analogues
36 and 37 were synthesised (Supporting Information
Scheme S6, Figures S90–S95) and incubated with FPPS, DcS and
(R)- or (S)-(1-13C,1-2H)IPP, leading to stereoselectively deuterated
and 13C-labelled FPP analogues 15 and 16 (Scheme 8). The syn-
allylic transposition of diphosphate results in stereospecifically
labelled 17 and 18, followed by anti-SN2’ attack at C1 in the
cyclisation to G and G’ that exhibit a stereogenic centre at
deuterated C1 of known configuration.[26] Analysis of the
products by HSQC spectroscopy (Supporting Information
Figures S96–S100) in conjunction with the NOESY interaction
between the 1-pro-R hydrogen and H6 revealed 6R configu-
ration for 27, 28 and 34. For 29 and 33 the specific migration of
the 1-pro-S hydrogen to C7 was observed which is in line with
6R configuration also for 29 and 6S configuration for 33 (this is
the same stereochemistry at C6, but there is a change in the
priorities of substituents). Similar experiments with (E)- and (Z)-
(4-13C,4-2H)IPP confirmed all assignments for the stereochemis-
try at C6 (Supporting Information Figures S101–S105).
Scheme 6. Enzymatic conversion of 15 with DcS. A) Cyclisation mechanism
for the conversion of 15. B) Known natural products that are structurally
similar to 27–29. C) Cyclisation mechanism suggested by Hong and
Tantillo[21] from the bisabolyl cation J to G as a proposed intermediate
towards tenuifolene (30).
require downstream oxidations. Thus, the enzyme reaction with
15 gives access to the biosynthetic precursor of tenuifolenes.
Conversion of 16 with DcS led to the isolation of two
compounds (Supporting Information Figures S74 and S75).
NMR-based structure elucidation gave access to the structures
of ketone 33 (Supporting Information Table S8, Figures S76–
S82) and hydroxy ketone 34 (Supporting Information Table S9,
Figures S83–S89), showing that the carbonyl group in 16 does
not participate in cyclisations. The cyclisation of 16 proceeds by
isomerisation to the NPP analogue 18, followed by 1,6-
cyclisation to G’ (Scheme 7A). A subsequent 1,3-hydride shift to
I’ and deprotonation yield 33, while the attack of water to G’
leads to 34. This alcohol is structurally similar to α-bisabolol for
which all four stereoisomers are known from Nature (Sche-
me 7B). The stereoisomer (6S,7S)-35 occurs in chamomile
oil,[22,23] (6S,7R)-35 (“anymol”) is known from Myoporum
crassifolium,[24] (6R,7R)-35 occurs in Populus balsamifera[24] and
(6R,7S)-35 in Rosa rugosa.[25]
Comparison of the 13C-NMR data of 29 and 33 to those
reported for synthetic ent-32 and 6-epi-32[27] suggested that
both compounds 29 and 33 have 7R configuration (Supporting
Information Figure S106), leading to the assignment of the
structures of (6S,7R)-29 and (6S,7R)-33. A similar comparison of
the 13C-NMR data of 34 to those of (6S,7S)-35 and (6S,7R)-35[28]
Assignment of the relative and absolute configurations of
the obtained compounds 27–29, 33 and 34 proved to be a
Chem. Eur. J. 2021, 27, 1–8
www.chemeurj.org
4
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100962
Chemistry—A European Journal
domain of geosmin synthase from Streptomyces coelicolor
complexed with three Mg2+ and alendronate as template (PDB:
5DZ2).[37] The model shows that the aromatic residues, H58,
H197 and Y241 are most likely located at the surface of the
protein (Figure 1).
In order to expand previous mutational work, these residues
were selected for site-directed mutagenesis, yielding the H58F,
H197F and Y241F enzyme variants as soluble proteins (Support-
ing Information Figure S115). Their catalytic activity in the
conversion of FPP was investigated in small scale incubation
experiments, showing a 2- to 3-fold production of 8 compared
to wildtype (WT) DcS in all three cases (Figure 2). This
serendipitous finding prompted us to investigate whether the
catalytic activity can be further increased for any of the three
Scheme 8. Determination of the configuration at C6 of 27–29, 33 and 34 by
stereoselective labelling experiments.
suggested the structure of (6R,7S)-34. This result was confirmed
by a synthetic transformation of (6S,7S)-35 into (6S,7S)-34
(Supporting Information Scheme S7) that showed different NMR
data to those of 34 obtained from 16 with DcS (Supporting
Information Figures S107–S113). For compounds 27 and 28
based on the assumed substrate fold in the DcS active site 7R
configuration is tentatively assigned, leading to (6R,7R)-27 and
(6R,7R)-28. Comparison of the optical rotation of 29 (½a�2D0 = +
3.3, c 0.03, CH2Cl2) and 33 (½a�2D0 = +21.4, c 0.21, CH2Cl2) to
reported data for 32 (½a�2D9 =À 3.99, in substance;[20] [α]D =À 7.48,
c 0.82, CHCl3)[29] were in line with their pseudoenantiomeric
relationship. Along similar lines, the optical rotation of 34 (
½a�2D0 = +51.7, c 0.12, CH2Cl2) in comparison to that of (6R,7S)-35
(½a�2D3 = +63, c 0.29, MeOH)[25] confirmed their stereochemically
coinciding skeletons.
Figure 1. DcS homology model generated by Swiss modelling. Amino acid
residues mutated in this study are highlighted.
It is well known that bacterial type I terpene synthases
contain several highly conserved motifs that are important for
catalytic activity. This includes the aspartate-rich motif DDXX(X)
D,[30] for DcS with
a
missing third Asp (92DDYFA), the
pyrophosphate sensor 190R,[31] the NSE triad 236NDVASYERE, and
the RY pair (Supporting Information Figure S114).[32] These
motifs are involved in binding of the Mg2+ cofactor and in
substrate recognition. In addition, active site aromatic residues
stabilise cationic intermediates along the terpene cyclisation
cascade,[33] and their mutation often causes adverse
effects.[31,34–36] Enzyme modelling of DcS using the Swiss Model
server returned the best quality model based on the N-terminal
Figure 2. Production of 8 by WT DcS and its enzyme variants obtained by
site-directed mutagenesis. The bars indicate average relative amounts of 8
from triplicates analysed by GC/MS.
Chem. Eur. J. 2021, 27, 1–8
www.chemeurj.org
5
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100962
Chemistry—A European Journal
possible combinations of double mutations or for the triple
mutant. While for the H58F/H197F enzyme variant and the
triple mutant the activity was in the range of WT DcS and the
H58F/Y241F variant did not show any positive effect in
comparison to the single mutants, the H197F/Y241F double
mutant exhibited the 4-fold production of 8. Large scale
incubations of FPP (80 mg) with H197F/Y241F increased the
isolated yield of 8 from 11% to 46%, showing that the higher
catalytic efficiency is preparatively very useful. Also the
incubation of 9 with H197F/Y241F increased the yields for 19
from 4% to 11% and for 20 from 15% to 42%.
these cases the additional methyl groups were not placed in
positions that lead to an altered reactivity of the obtained
methylated oligoprenyl diphosphate derivatives.[40,41] Also func-
tional groups such as the keto group in the FPP analogue 16
are tolerated by DcS, similar to recent findings for other terpene
synthases,[42–45] yielding products 33 and 34 with functionalised
side chains that allow further chemical transformations. We
have also investigated the effect of mutating aromatic amino
acid residues that are, according to a homology model, located
on the surface of DcS. These mutations resulted in an enzyme
variant that gives significantly improved yields in preparative
scale incubations, which provides another nice example for the
importance of serendipity in science.
Kinetic measurements comparing the WT with the H197F/
Y241F mutant revealed an increase of the turnover rate kcat
from 35.6�2.9 sÀ 1 to 65.5�8.6 sÀ 1, but also an increased Km
from 2.90�0.02 mm to 5.95�0.02 mm, and thus only minor
differences regarding kcat/KM (1.2×104 sÀ 1 mÀ 1 for the WT and Acknowledgements
1.1×104 sÀ 1 mÀ 1 for H197F/Y241F, Supporting Information Ta-
ble S11, Figures S116 and S117). The wildtype enzyme showed
substrate inhibition at FPP concentrations above 1 mm, but the
H197/Y241F tolerated slightly higher substrate concentrations
up to 2 mm and can convert the substrate at a higher rate. This
is one explanation for the higher product formation in the large
This work was funded by the Deutsche Forschungsgemein-
schaft (DI1536/7-2). Open access funding enabled and organ-
ized by Projekt DEAL.
scale experiments, performed with a substrate concentration of Conflict of Interest
2 mm, at which the mutant performs significantly better than
the WT.
Enzyme stability issues may give a second explanation. It is
The authors declare no conflict of interest.
well know that the mutations on protein surfaces can have a
significant influence on protein stability.[38] This has also been
shown for a terpene synthase, tobacco 5-epi-aristolochene
synthase, for which the thermal protein stability was signifi-
cantly improved through site-directed mutagenesis of surface
residues,[39] but this enzyme variant showed a moderately
reduced catalytic performance in comparison to the wildtype.
For the DcS H197F/Y241F variant the increased enzyme stability
may be responsible for the higher yield in preparative scale
reactions in comparison to the wildtype. These reactions are
usually carried out over reaction times of 10–12 h in our
laboratory, often leading to enzyme denaturation and precip-
itation during the course of the incubation.
Keywords: biocatalyst · isotopic labelling · mutagenesis ·
terpene cyclisation
[1] J. S. Dickschat, Angew. Chem. Int. Ed. 2019, 58, 15964–15976; Angew.
Chem. 2019, 131, 16110–16123.
[2] D. W. Christianson, Chem. Rev. 2017, 117, 11570–11648.
[3] O. Wallach, Justus Liebigs Ann. Chem. 1885, 227, 277–302.
[4] L. Ruzicka, Experientia 1953, 9, 357–396.
[5] M. Komatsu, M. Tsuda, S. Omura, H. Oikawa, H. Ikeda, Proc. Nat. Acad.
Sci. 2008, 105, 7422–7427.
[6] C.-M. Wang, D. E. Cane, J. Am. Chem. Soc. 2008, 130, 8908–8909.
[7] J. S. Dickschat, T. Nawrath, V. Thiel, B. Kunze, R. Müller, S. Schulz, Angew.
Chem. Int. Ed. 2007, 46, 8287–8290; Angew. Chem. 2007, 119, 8436–
8439.
[8] S. von Reuß, D. Domik, M. C. Lemfack, N. Magnus, M. Kai, T. Weise, B.
Piechulla, J. Am. Chem. Soc. 2018, 140, 11855–11862.
[9] E. R. Duell, P. M. D’Agostino, N. Shapiro, T. Woyke, T. M. Fuchs, T. A. M.
Gulder, Microb. Cell Fact. 2019, 18, 32.
Conclusion
[10] C.-M. Wang, R. Hopson, X. Lin, D. E. Cane, J. Am. Chem. Soc. 2009, 131,
8360–8361.
[11] V. Harms, B. Schröder, C. Oberhauser, C. D. Tran, S. Winkler, G. Dräger, A.
Kirschning, Org. Lett. 2020, 22, 4360–4365.
[12] P. Rabe, J. Rinkel, T. A. Klapschinski, L. Barra, J. S. Dickschat, Org. Biomol.
Chem. 2016, 14, 158–164.
[13] C. Enzell, Acta Chem. Scand. 1962, 16, 1553–1568.
[14] L. G. Cool, Phytochemistry 2005, 66, 249–260.
[15] J. Rinkel, J. S. Dickschat, Org. Lett. 2019, 21, 2426–2429.
[16] P. Rabe, J. Rinkel, B. Nubbemeyer, T. G. Köllner, F. Chen, J. S. Dickschat,
Angew. Chem. Int. Ed. 2016, 55, 15420–15423; Angew. Chem. 2016, 128,
15646–15649.
In summary we have shown that DcS accepts non-natural FPP
analogues that were specifically designed to open new reaction
trajectories. 10-Methyl-FPP not only gave access to the meth-
ylated daucenol 20, but also to the methylated widdrenol 19,
similar to the findings for 13-desmethyl-FPP, which resulted in
the nor-widdrenol 21. The FPP isomer 15 gave access to
tenuifola-2,10-diene (27) and tenuifola-2,11-diene (28). Both the
widdrane and the tenuifolane skeletons are difficult to form
from natural FPP, as is evident from the low number of known
natural products from these classes, but can be reached with
the altered reactivity of FPP analogues with non-natural meth-
ylation patterns. We and others have recently shown that
additional methyl groups can also enzymatically be incorpo-
rated into terpenes from methylated IPP derivatives, but in
[17] L. Lauterbach, J. Rinkel, J. S. Dickschat, Angew. Chem. Int. Ed. 2018, 57,
8280–8283; Angew. Chem. 2018, 130, 8412–8415.
[18] F. M. Hahn, A. P. Hurlburt, C. D. Poulter, J. Bacteriol. 1999, 181, 4499–
4504.
[19] A. T. Kreipl, W. A. König, Phytochemistry 2004, 65, 2045–2049.
[20] D. W. Connell, M. D. Sutherland, Aust. J. Chem. 1966, 19, 283–288.
[21] Y. J. Hong, D. J. Tantillo, J. Am. Chem. Soc. 2015, 137, 4134–4140.
[22] M. A. Schwartz, G. C. Swanson, J. Org. Chem. 1979, 44, 953–958.
[23] T. Iwashita, T. Kusumi, H. Kakisawa, Chem. Lett. 1979, 8, 947–950.
Chem. Eur. J. 2021, 27, 1–8
www.chemeurj.org
6
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202100962
Chemistry—A European Journal
[24] E. Flaskamp, G. Nonnenmacher, O. Isaac, Z. Naturforsch. 1981, 36b, 114–
118.
[25] Y. Hashidoko, K. Endoh, T. Kudo, S. Tahara, Biosci. Biotechnol. Biochem.
[38] V. G. H. Eijsink, A. Bjork, S. Gaseidnes, R. Sirevag, B. Synstad, B.
van den Burg, G. Vriend, J. Biotechnol. 2004, 113, 105–120.
[39] J. E. Diaz, C.-S. Lin, K. Kunishiro, B. K. Feld, S. K. Avrantinis, J. Bronson, J.
Greaves, J. G. Saven, G. A. Weiss, Protein Sci. 2011, 20, 1597–1606.
[40] A. Hou, L. Lauterbach, J. S. Dickschat, Chem. Eur. J. 2020, 26, 2178–2182.
[41] L. Drummond, M. J. Kschowak, J. Breitenbach, H. Wolff, Y.-M. Shi, J.
Schrader, H. B. Bode, G. Sandmann, M. Buchhaupt, ACS Synth. Biol. 2019,
8, 1303–1313.
2000, 64, 907–910.
[26] J. Rinkel, P. Rabe, P. Garbeva, J. S. Dickschat, Angew. Chem. Int. Ed. 2016,
55, 13593–13596; Angew. Chem. 2016, 128, 13791–13794.
[27] W. Kreiser, F. Körner, Helv. Chim. Acta 1999, 82, 1427–1433.
[28] Y. Hashidoko, S. Tahara, J. Mizutani, Phytochemistry 1994, 35, 325–329.
[29] M. Tori, M. Aoki, Y. Asakawa, Phytochemistry 1994, 36, 73–76.
[30] C. M. Starks, K. Back, J. Chappell, J. P. Noel, Science 1997, 277, 1815–
1820.
[42] M. Demiray, X. Tang, T. Wirth, J. A. Faraldos, R. K. Allemann, Angew.
Chem. Int. Ed. 2017, 56, 4347–4350; Angew. Chem. 2017, 129, 4411–
4415.
[31] P. Baer, P. Rabe, K. Fischer, C. A. Citron, T. A. Klapschinski, M. Groll, J. S.
Dickschat, Angew. Chem. Int. Ed. 2014, 53, 7652–7656; Angew. Chem.
2014, 126, 7783–7787.
[32] E. Y. Shishova, L. Di Costanzo, D. E. Cane, D. W. Christianson, Biochemis-
try 2007, 46, 1941–1951.
[33] C. A. Lesburg, G. Zhai, D. E. Cane, Science 1997, 277, 1820–1824.
[34] M. Seemann, G. Z. Zhai, J. W. de Kraker, C. M. Paschall, D. W. Christian-
son, D. E. Cane, J. Am. Chem. Soc. 2002, 124, 7681–7689.
[35] J. A. Aaron, X. Lin, D. E. Cane, D. W. Christianson, Biochemistry 2010, 49,
1787–1797.
[43] F. Huynh, D. J. Grundy, R. L. Jenkins, D. J. Miller, R. K. Allemann,
ChemBioChem 2018, 19, 1834–1838.
[44] M. Demiray, D. J. Miller, R. K. Allemann, Beilstein J. Org. Chem. 2019, 15,
2184–2190.
[45] C. Oberhauser, V. Harms, K. Seidel, B. Schröder, K. Ekramzadeh, S. Beutel,
S. Winkler, L. Lauterbach, J. S. Dickschat, A. Kirschning, Angew. Chem.
Int. Ed. 2018, 57, 11802–11806; Angew. Chem. 2018, 130, 11976–11980.
[36] H. Xu, J. Rinkel, X. Chen, T. G. Köllner, F. Chen, J. S. Dickschat, Org.
Biomol. Chem. 2021, 19, 370–374.
[37] G. G. Harris, P. M. Lombardi, T. A. Pemberton, T. Matsui, T. M. Weiss, K. E.
Cole, M. Köksal, F. V. Murphy, L. S. Vedula, W. K. W. Chou, D. E. Cane,
Biochemistry 2015, 54, 7142–7155.
Manuscript received: February 25, 2021
Accepted manuscript online: March 26, 2021
Version of record online: ■■■, ■■■■
Chem. Eur. J. 2021, 27, 1–8
www.chemeurj.org
7
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
Daucenol Synthase (DcS) from Strep-
tomyces venezuelae was redirected
using substrate analogues with addi-
tional, shifted, or lacking methyl
L. Lauterbach, Dr. A. Hou, Prof. Dr. J. S.
Dickschat*
1 – 8
Rerouting and Improving Dauc-8-en-
11-ol Synthase from Streptomyces
venezuelae to a High Yielding Bioca-
talyst
groups to yield molecules from a
naturally difficult to access chemical
space, and improved in its catalytic ef-
ficiency through site-directed muta-
genesis of surface residues, turning
DcS into a high yielding biocatalyst.