Understanding and Engineering the Stereoselectivity of Humulene Synthase

Article abstract of DOI:10.1002/anie.202106718

The non-canonical terpene cyclase AsR6 is responsible for the formation of 2E,6E,9E-humulene during the biosynthesis of the tropolone sesquiterpenoid (TS) xenovulene A. The structures of unliganded AsR6 and of AsR6 in complex with an in crystallo cyclized reaction product and thiolodiphosphate reveal a new farnesyl diphosphate binding motif that comprises a unique binuclear Mg²⁺-cluster and an essential K289 residue that is conserved in all humulene synthases involved in TS formation. Structure-based site-directed mutagenesis of AsR6 and its homologue EupR3 identify a single residue, L285/M261, that controls the production of either 2E,6E,9E- or 2Z,6E,9E-humulene. A possible mechanism for the observed stereoselectivity was investigated using different isoprenoid precursors and results demonstrate that M261 has gatekeeping control over product formation.

Full text of DOI:10.1002/anie.202106718

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
International Edition Chemie
www.angewandte.org
Accepted Article
Title: Understanding and Engineering the Stereoselectivity of
Humulene Synthase
Authors: Carsten Schotte, Peer Lukat, Adrian Deuschmann, Wulf
Blankenfeldt, and Russell John Cox
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: Angew. Chem. Int. Ed. 10.1002/anie.202106718
Link to VoR: https://doi.org/10.1002/anie.202106718

Angewandte Chemie International Edition
10.1002/anie.202106718
COMMUNICATION
Understanding and Engineering the Stereoselectivity of Humulene
Synthase
[
a]
[b]
[a]
[b,c]
[a]
Carsten Schotte,* Peer Lukat, Adrian Deuschmann, Wulf Blankenfeldt and Russell J. Cox*
[
a]
C. Schotte, A. Deuschmann, Prof. Dr. R. J. Cox
Institute for Organic Chemistry and BMWZ
Leibniz Universität Hannover
Schneiderberg 38, 30167 Hannover, Germany
E-mail:carsten.schotte@oci.uni-hannover.de; russell.cox@oci.uni-hannover.de
Dr. P. Lukat, Prof. Dr. W. Blankenfeldt
[
[
b]
c]
Structure and Function of Proteins
Helmholtz Centre for Infection Research
Inhoffenstr. 7, 38124 Braunschweig, Germany
Prof. Dr. W. Blankenfeldt,
Institute for Biochemistry, Biotechnology and Bioinformatics,
Technische Universität Braunschweig,
Spielmannstr. 7, 38106 Braunschweig, Germany
Supporting information for this article is given via a link at the end of the document.
Abstract: The non-canonical terpene cyclase AsR6 is responsible for
the formation of 2E,6E,9E-humulene during the biosynthesis of the
tropolone sesquiterpenoid (TS) xenovulene A. The structures of
unliganded AsR6 and of AsR6 in complex with an in crystallo cyclized
A
Xenovulene-type TS
Neosetophome-type TS
reaction product and thiolodiphosphate reveal
a new farnesyl
2
O
H
O
2
E
2+
HO
O
3
diphosphate binding motif that comprises a unique binuclear Mg -
cluster and an essential K289 residue that is conserved in all
humulene synthases involved in TS formation. Structure-based site-
directed mutagenesis of AsR6 and its homologue EupR3 identify a
single residue, L285/M261, that controls the production of either
3
Z
H
HO
H
HO
O
H
O
xenovulene A 1
neosetophome B 2
2
E,6E,9E- or 2Z,6E,9E-humulene. A possible mechanism for the
H
H
O
O
2
2
observed stereoselectivity was investigated using different isoprenoid
precursors and results demonstrate that M261 has gatekeeping
control over product formation.
3
3
trans
cis
HO
H
HO
H
HO
O
HO
O
O
O
O
O
pycnidione 3
eupenifeldin 4
HO
HO
Tropolone sesquiterpenoids (TS) are privileged meroterpenoid
natural products of hybrid terpene/polyketide origin.
Representative examples (e.g. xenovulene A 1, neosetophome
B 2, pycnidione 3, eupenifeldin 4; Scheme 1A) possess potent
psychoactive, antitumor and antiarthritic properties.[1–5] All TS
feature a core 11-membered macrocycle that is derived from the
sesquiterpene humulene.[6,7] The biosynthesis of TS proceeds via
a core intermolecular hetero Diels-Alder reaction that connects
humulene with the polyketide-derived tropolone nucleus observed
in 1-4 (Electronic Supplementary Information [ESI] Figure S1-
S2).[6] TS can further be divided into two major subfamilies.[6]
Xenovulene-type TS are derived from 2E,6E,9E-humulene 5 and
the 2E-alkene configuration in 5 is mirrored in the final pathway
B
H
H
11
11
B:
B:
9
11
11
9
9
9
2
?
2
2
3
2
3
6
E
6
6
6
E
Z
Z
3
3
1
5
15
E,E-humulyl
cation 8
Z,E-humulyl
cation 10
2
Z-humulene 6
proposed
ɑ-humulene 5
1,11 ring closure
?
9
11
11
9
7
1
1,3-suprafacial
transposition
1
7
1
OPP
2,3-rotation
3
3
3
5
product (2E-alkene in 1; trans-fusion at the C-2/C-3 ring junction
in 3; Scheme 1AB).[8] Neosetophome-type TS display a 2Z-alkene
5
OPP
OPP
E,E-FPP 7
NPP 9
configuration in monosubstituted pathway products (e.g. 2) and
cis fusion at the C-2/C-3 ring junction in disubstituted TS such as
Scheme 1. A, representative examples of tropolone sesquiterpenoids; B,
proposed humulene formation during the biosynthesis of 1-4; dashed arrows =
putative steps in the formation of 6; B = general base.
4
(Scheme 1AB). 2Z,6E,9E-humulene 6 was proposed as the
precursor of 2 and 4, to explain the difference in stereochemistry,
but there is only limited evidence for the existence of 6 as a
biosynthetic intermediate.[6]
Sesquiterpenes such as 5 are biosynthesized by class I
terpene cyclases from the universal precursor 2E,6E-farnesyl
diphosphate 7 (FPP).[9] All known class I terpene cyclases share
1
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202106718
COMMUNICATION
a common ɑ-helical fold and two conserved aspartate-rich motifs
2E-humulene 5. The exact conformation of the ring-like structure
could not be determined unambiguously and the displayed
structure is in best agreement with the observed electron density
(ESI Fig. S45).
(
DDxxD and NSE/DTE; bold indicates metal binding) that
2+
coordinate a trinuclear Mg -cluster responsible for binding 7 and
triggering the loss of diphosphate (PP
).[9] Cleavage of PP
i
i
generates a reactive allylic cation that can undergo diverse
intramolecular cyclizations, rearrangements and final proton
loss.[9]
AsR6
FPP
EupR3
+ FPP
A
5 6
B
5 6
+
From a mechanistic viewpoint 2E-humulene synthases
belong to the simplest cyclases. 1,11-cyclization of FPP 7 yields
the E,E-humulyl cation 8 and deprotonation then affords 5 without
further rearrangement (Scheme 1B). The occurrence of Z-
configured alkenes in terpenes (e.g. 6) requires a formal
isomerization of the 2E-alkene in FPP 7, that is assumed to
15
16
17
15
16
17
proceed via a 1,3-suprafacial transposition of the PP
cisoid nerolidyl diphosphate 9b (NPP) after rotation of the C-2/C-
bond (Scheme 1B).[10–12] Cleavage of PP
and 1-11-cylization
i
in 7 to give
3
i
C
E
AsR6K289S
D
F
AsR6L285N,K289S
+ FPP
+
FPP
would yield the Z,E-humulyl cation 10 and deprotonation could
then afford 6. However, the structural factors that control the
transposition from 7 to 9 are poorly understood.
Two cryptic terpene cyclases (AsR6, PycR6) have recently
been identified as responsible for 2E-humulene 5 formation in TS
natural products.[8,13] They bear no significant sequence identity
to any known terpene synthase and the typical metal-binding
motifs involved in FPP 7 coordination are not identifiable.[8]
Homologous enzymes (EupE; EupfG and EupR3) have been
linked to the production of neosetophome B 2 and eupenifeldin 4,
but the terpene product of these enzymes was not directly
identified.[6,13,14]
1
5
16
17
15
16
17
AsR6L285M
EupR3M261L
+ FPP
+ FPP
Here, the E. coli codon-optimized genes asR6 and eupR3
were expressed in E. coli BL21 as polyhistidine-tagged enzymes.
Incubation of AsR6 with FPP 7 gives, as previously reported,[7]
1
5
16
17
time /min
15
16
17
time /min
2
E-humulene 5 (m/z 204). Production of 5 was confirmed by gas
chromatography mass spectrometry (GCMS) analysis (Figure 1A;
ESI Fig. S3).[8] Identical in vitro assays with EupR3 afforded a
single product with the same molecular mass (m/z 204) as 5, but
Figure 1. Total ion chromatograms of extracts from assays of wildtype enzymes
and mutants with FPP 7: A, incubation of AsR6 with 7; B, incubation of EupR3
with 7; C, incubation of AsR6K289S with 7; D, incubation of AsR6L285N
_
K289S with
a
different retention time (Figure 1B). A preparative-scale
7
; E, incubation of AsR6L285M with 7; F, incubation of EupR3M261L with 7.
biotransformation of FPP 7 with EupR3 allowed analysis of the
terpene product by nuclear magnetic resonance (NMR). Structure
elucidation confirmed the product as the expected 2Z-humulene
i
The SPP is coordinated by R240, K289, R350, Y351 and R354,
6
(ESI Fig. S4-S23; Table S1).[6] The different configurations of
the C-2/C-3 alkene in 5 and 6 were confirmed by NOESY NMR.
Key nOe correlations of H -15 to H-2 in 6, but from H -15 to H -1
whereas W138 forms an aromatic wall of the reaction chamber
and may be involved in stabilizing cationic intermediates (Figure
3
3
2
2
C). R350 and Y351 form a conserved RY-dimer that is widely
in 5, together with other correlations, confirm the expected alkene
configuration (ESI Fig. S24; Table S1+S2).
encountered in class I terpene cyclases.[20] Intriguingly, despite
the overall structural similarity to canonical terpene cyclases,
analysis of the electron density map identifies only two
coordinated MgA+B2+-ions (Figure 2C). Canonical terpene
cyclases typically harbour a trinuclear Mg2+ cluster.[21] Both Mg2+-
ions in AsR6 are coordinated by the carboxylate of D164 (situated
on an aspartate-rich motif on HL-4).
Next, we obtained the crystal structure of AsR6 in its open
conformation (2.0 Å, PDB: 7OC5, Table S7) and in complex with
the synthetic substrate analogue farnesyl-thiolodiphosphate
(
FSPP, 2.0 Å, PDB: 7OC4, Figure 2AB, Table S7).[15,16] AsR6
crystallized as a homodimer, which was confirmed by analysis
with PDBePISA[17] and is in accordance with multi angle light
scattering data ([MALS] ESI Fig. S25). Each polypeptide chain
folds into a single, purely ɑ-helical domain consisting of eleven
helices (HL) with HL-3 to HL-10 forming a barrel-like core with a
central hydrophobic cavity. The overall protein fold of AsR6 is
similar to other Type I sesquiterpene cyclases. DALI analysis[18]
identified the bacterial selinadiene synthase (PDB: 4OKM) as the
closest structural relative (12% sequence identity, RMSD 4.1 Å,
ESI Table S3 and Fig. S26).[18,19] To our surprise FSPP did not
crystallize intact. Analysis of the electron density identified a
Structure comparison with known fungal terpene cyclases
reveals identical aspartates e.g. in trichodiene synthase[ and
22]
aristolochene synthase,[ that coordinate the equivalent Mg
23]
2+
A
and Mg 2
+
ions respectively (ESI Fig. S27+28). However,
+-ion is
B
whereas in canonical terpene cyclases the third Mg 2
coordinated by a second metal binding motif (N224
selinadiene-synthase; Figure 2D),[19] in AsR6 the Mg
C
S
228
E232 in
2+-ion is
C
replaced by the e-ammonium of K289 that occupies the
corresponding position in the active site (Figure 2D). K289 is
located on the opposite side of the active site on HL-7 and
coordinates the diphosphate via its lysyl side chain.
cleaved thiolodiphosphate (SPP
i
) moiety and a ring-like structure
within the active site that best matches the reaction product
2
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202106718
COMMUNICATION
Figure 2. Crystal structure of AsR6: A, crystal structure of the AsR6 dimer; chain A shown as cartoon coloured from N- (blue) to C-terminus (red); chain B shown
in surface representation; ligands are represented as sticks/spheres; B, crystal structure of AsR6 in complex with in crystallo cyclized reaction product and
thiolodiphosphate; Zn2+-binding site highlighted in red; C, AsR6 Mg2+-binding site showing bound humulene (green); D, structural alignment of the diphosphate
binding site in AsR6 (Mg2+ = light green; residues = cyan) and selinadiene synthase (PDB: 4okm, Mg2+ = dark green; residues = grey, italic labels).
To validate this highly unusual diphosphate binding motif, we
mutated K289. Incubation of FPP 7 with K289S and analysis by
GCMS showed that formation of 2E-humulene 5 was abolished,
consistent with the proposed crucial role of K289 in
pyrophosphate coordination (Figure 1C). We then probed
whether the typically observed NSE-motif could be (re)introduced
product configurations we identified all residues in the active site
of AsR6 that are within 6 Å of the bound FSPP/humulene
substrate (ESI Fig. S31). Mapping of these residues to a global
sequence alignment of the humulene synthases involved in
formation of 5 (AsR6, PycR6) and 6 (EupR3, EupE) showed that
all active site residues are highly conserved in both types of
into AsR6, to reconstitute the canonical diphosphate binding motif. humulene synthases. Only one residue is consistently different in
Structural alignment of AsR6 with trichodiene and aristolochene
synthases reveals that the double mutant AsR6L285N_K289S would
fulfil this requirement, as the complementary E298 already
occupies the appropriate position in the active site (ESI Fig. S29).
However, analysis of the AsR6L285N_K289S double mutant revealed
that enzyme activity was not restored upon incubation with FPP 7
the two types of cyclases: humulene cyclases producing 5 have a
conserved L285 residue that is replaced by a methionine in
cyclases giving 6 (ESI Fig. S32).
L285 extends into the active site cavity of AsR6 with one of
ẟ
its C atoms in close proximity to the C-2/C-3 alkene (4.5 Å and
3.9 Å; Figure 3A). We reasoned that in the absence of other
conserved changes residues at this position are likely involved in
controlling product stereoselectivity.
(
Figure 1D).
To complete the structural characterization of AsR6 we
report coordination of a Zn2+ ion by C43, C85, C87 and H41. This
Site-directed mutagenesis afforded the cross-convergent
mutants AsR6L285M and EupR3M261L.[24] Incubation of the
AsR6L285M mutant with FPP 7 and analysis by GCMS identified
production of minor 2E-humulene 5 (12%; based on peak
integration) and major 2Z-humulene 6 (>85%; Figure 1E).
Analysis of the EupR3M261L mutant validated the role of this
residue.
binding motif is conserved in the AsR6-like humulene synthases
(
ESI Fig. S30), but absent in canonical class I terpene cyclases
and might be involved in protein stability but was not further
investigated as it is located away from the active site (Figure 2B).
Despite sharing ∼50% sequence identity, AsR6 and EupR3
each give rise to a single humulene isomer (5 or 6) in the presence
of FPP 7 and exhibit complete control of the stereochemical
outcome. To identify structural features that drive the different
3
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202106718
COMMUNICATION
single point mutant T296V abolished production of amorphadiene
from FPP 7 and production was only restored upon incubation
with (3R)-NPP 9.[25]
We probed the in vitro activity of wild-type enzymes and
mutants with synthetic (±) NPP 9. Incubation of 9 with AsR6
affords the native terpene product 5 as the major product (Figure
3
B). Minor formation of the Z-isomer 6 confirms that 9 is indeed a
precursor of Z-humulene 6 - but in AsR6 this trajectory is less
favorable. Incubation of EupR3 with 9 exclusively affords 2Z-
humulene 6 (Figure 3C). For AsR6 the transoid conformer of the
tertiary allylic isomer NPP, 9a, is the likely true substrate (Scheme
2
). EupR3 likely catalyzes the conversion of 9a into the cisoid
conformer 9b and then formation of 6 (Scheme 2). GCMS
analysis of the cross-convergent mutants revealed that AsR6L285M
produces 2Z-humulene 6 and only traces of 5 were observed
(
Figure 3D).
Finally, incubation of 9 with EupR3M261L reconstituted the
formerly abolished formation of 6. However, intriguingly the
experiment afforded a ∼1:1 mixture of 5 and 6 (Figure 3E).
Formation of both 5 and 6 suggests that the methionine M261 is
not the sole controlling factor during the transformation of FPP
into NPP. The exact catalytic functions of this residue (control of
the allylic isomerization of FPP into NPP vs controlling different
reaction trajectories) are currently under further investigation.
We also probed whether only one of the two NPP
enantiomers is converted by AsR6. The (3R)-enantiomer of NPP
9
is usually observed to be the intermediate during cyclisation
reactions.[26] E.g., Z-γ-bisabolene synthase (BbS) selectively
converts (3R)-NPP to Z-γ-bisabolene, while incubation with the
(
3S)-enantiomer affords the acyclic E-β-farnesene.[12] First, (±)-9
was incubated with AsR6. Then BbS was added in order to
convert any residual NPP to either Z-γ-bisabolene or E-β-
farnesene. GCMS analysis revealed formation of 2E-humulene
and no production of Z-γ-bisabolene when compared to a control
reaction lacking AsR6, suggesting that AsR6 converts both (3R)-
NPP and (3S)-NPP to 2E-humulene 5 (ESI Fig. S35-1&2).
However, more detailed studies are required to confirm this
unusual conversion of both enantiomers of (±)-9.
Figure 3. Investigations into the stereoselectivity of humulene formation: A,
position of key residue L285 in the active site; B, total ion chromatogram of
extracts from incubation of AsR6 with rac-NPP 9; C, incubation of EupR3 with
2
E
Z
2
9
; D, incubation of AsR6L285M with 9; E, incubation of EupR3M261L with 9.
11
11
5
6
B:
B:
Here, production of the native terpene product 6 is abolished
H
2
E
H
Z
2
and EupR3M261L exclusively produces 5 instead (Figure 1F).
Notably, the introduced M261L mutation leads to a complete
switch in the observed product stereochemistry and indicates that
the presence of methionine at this position is a prerequisite for 6
formation. Further site-directed mutagenesis experiments were
directed to investigate the influence of smaller amino acids on this
position (L285A and L285V) but did not affect product formation
11
11
8
10
AsR6
EupR3M261L
AsR6
EupR3M261L
EupR3
EupR3M261L
AsR6L285M
EupR3
EupR3M261L
AsR6L285M
EupR3
AsR6L285M
OPP
1
OPP
3
3
3
(
ESI Fig. S33).
As formation of the 2Z alkene requires
1
11
11
11
OPP
1
a
formal
E,E-FPP 7
transoid 9a
cisoid 9b
isomerization of the 2E alkene in FPP 7, presumably via the well-
established initial formation of NPP 9 (Scheme 1B), we reasoned
that M261 could be involved in the initial transformation of 7 to 9.
Alternatively, the residue could be involved in controlling the
possible carbocation trajectories that lead to the E,E- or Z,E-
humulyl cations 8 and 10 respectively. Cane et al. reported similar
effects of an individual amino acid residue on the stereochemical
outcome for amorpha-4,11-diene synthase (ESI Fig. S34).[25] The
Scheme 2. Putative biosynthetic pathways leading to 5 and 6 from the different
isoprenoid precursors; B = general base.
To validate the conversion of (3S)-NPP by humulene
synthases we synthesized the (3S)-NPP enantiomer (see ESI for
details). As expected, incubation of (3S)-NPP with AsR6, EupR3
4
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202106718
COMMUNICATION
[
[
13]
14]
C. Schotte, L. Li, D. Wibberg, J. Kalinowski, R. J. Cox, Angew.
Chemie - Int. Ed. 2020, 59, 23870–23878.
Y. Zhai, Y. Li, J. Zhang, Y. Zhang, F. Ren, X. Zhang, G. Liu, X. Liu,
Y. Che, Fungal Genet. Biol. 2019, 129, 7–15.
and AsR6L285M as well as EupR3M261L affords identical product
distributions as observed for incubation with rac-NPP (ESI Fig.
S36).
[
[
15]
16]
R. M. Phan, C. D. Poulter, J. Org. Chem. 2001, 66, 6705–6710.
G. Ramamoorthy, R. M. Phan, C. D. Poulter, J. Org. Chem. 2016,
In summary, we report the first crystal structure of a non-
canonical humulene synthase that shows no significant sequence
homology to other known class I terpene cyclases and humulene
synthases. A novel diphosphate binding motif was identified and
validated for class I terpene cyclases that comprises a binuclear
8
1, 5093–5100.
[
17]
E. Krissinel, K. Henrick, J. Mol. Biol. 2007, 372, 774–797.
[18]
19]
L. Holm, Protein Sci. 2020, 29, 128–140.
P. Baer, P. Rabe, K. Fischer, C. A. Citron, T. A. Klapschinski, M.
Groll, J. S. Dickschat, Angew. Chemie - Int. Ed. 2014, 53, 7652–
[
magnesium cluster and
a strictly conserved lysine K289.
7
656.
Binculear magnesium clusters have been identified in very few
other enzymes (e.g. UbiA-type prenyltransferases)[27] but these
show no significant homology to the fungal AsR6-like humulene
synthases reported here (Supporting Figure S52), and do not
feature the replacement of a magnesium ion by the e-ammonium
of a lysine residue. AsR6 can utilize FPP 7 and apparently both
enantiomers of NPP 9 as substrates. A single amino acid residue
drives the stereochemical outcome of 2E vs 2Z-humulene
formation. Our discoveries thus broaden the reservoir of fungal
non-canonical terpene cyclases and pave the way for further
engineering of new-to-nature tropolone sesquiterpenoids.
[20]
P. Rabe, T. Schmitz, J. S. Dickschat, Beilstein J. Org. Chem. 2016,
12, 1839–1850.
D. W. Christianson, Chem. Rev. 2017, 117, 11570–11648.
[
[
21]
22]
M. J. Rynkiewicz, D. E. Cane, D. W. Christianson, Proc. Natl. Acad.
Sci. 2001, 98, 13543–13548.
M. Chen, N. Al-Lami, M. Janvier, E. L. D’Antonio, J. A. Faraldos, D.
E. Cane, R. K. Allemann, D. W. Christianson, Biochemistry 2013,
[
23]
5
2, 5441–5453.
T. G. Köllner, J. Gershenzon, J. Degenhardt, Phytochemistry 2009,
0, 1139–1145.
[24]
25]
[26]
27]
7
[
Z. Li, R. Gao, Q. Hao, H. Zhao, L. Cheng, F. He, L. Liu, X. Liu, W. K.
W. Chou, H. Zhu, et al., Biochemistry 2016, 55, 6599–6604.
M. B. Quin, C. M. Flynn, C. Schmidt-Dannert, Nat. Prod. Rep. 2014,
3
1, 1449–1473.
[
W. Cheng, W. Li, Science (80-. ). 2014, 343, 878–881.
Acknowledgements
CS thanks Leibniz University for funding. DFG is thanked for
funding (INST 187/686-1). Prof. Jeroen Dickschat is thanked for
the gift of plasmid pYE-BbS and Vanessa Harms is thanked for
the gift of farnesyl pyrophosphate. Luca Codutti and Georg Krüger
are thanked for MALS measurements. Ute Widow is thanked for
technical assistance. We acknowledge DESY (Hamburg,
Germany), a member of the Helmholtz Association HGF, for the
provision of experimental facilities. Parts of this research were
carried out at beamline P11 at the PETRA III storage ring and we
would like to thank the beamline staff for assistance during data
collection. Beamtime was allocated for proposal Xh-20010031.
Keywords: biosynthesis • meroterpenoid • tropolone
sesquiterpenoid • enzyme engineering • humulene
[
[
[
[
[
[
1]
2]
3]
4]
5]
6]
A. M. Ainsworth, M. I. Chicarelli-Robinson, B. R. Copp, U. Fauth, P.
J. Hylands, J. A. Holloway, M. Latif, G. B. O´Beirne, N. Porter, D. V
Renno, et al., J. Antibiot. (Tokyo). 1995, 48, 568–573.
T. El-Elimat, H. A. Raja, S. Ayers, S. J. Kurina, J. E. Burdette, Z.
Mattes, R. Sabatelle, J. W. Bacon, A. H. Colby, M. W. Grinstaff, et
al., Org. Lett. 2019, 21, 529–534.
C.-J. Hsiao, S.-H. Hsiao, W.-L. Chen, J.-H. Guh, G. Hsiao, Y.-J.
Chan, T.-H. Lee, C.-L. Chung, Chem. Biol. Interact. 2012, 197, 23–
3
0.
C. Y. Bemis, C. N. Ungarean, A. S. Shved, C. S. Jamieson, T.
Hwang, K. S. Lee, K. N. Houk, D. Sarlah, J. Am. Chem. Soc 2021,
1
43, 6006–6017.
Z. Y. Al Subeh, N. Q. Chu, J. T. Korunes-Miller, L. L. Tsai, T. N.
Graf, Y. P. Hung, C. J. Pearce, M. W. Grinstaff, A. H. Colby, Y. L.
Colson, et al., J. Control. Release 2021, 331, 260–269.
Q. Chen, J. Gao, C. Jamieson, J. Liu, M. Ohashi, J. Bai, D. Yan, B.
Liu, Y. Che, Y. Wang, et al., J. Am. Chem. Soc. 2019, 141, 14052–
1
4056.
[
[
7]
8]
F. Yu, S. Okamto, K. Nakasone, K. Adachi, S. Matsuda, H. Harada,
N. Misawa, R. Utsumi, Planta 2008, 227, 1291–1299.
R. Schor, C. Schotte, D. Wibberg, J. Kalinowski, R. J. Cox, Nat.
Commun. 2018, 9, 1963.
[
[
[
9]
J. D. Rudolf, C. Y. Chang, Nat. Prod. Rep. 2020, 37, 425–463.
D. Arigoni, Pure Appl. Chem. 1975, 41, 219–245.
D. E. Cane, R. Iyengar, M. S. Shiao, J. Am. Chem. Soc. 1981, 103,
10]
11]
9
14–931.
J. Rinkel, J. S. Dickschat, Beilstein J. Org. Chem. 2019, 15, 789–
94.
[
12]
7
5
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202106718
COMMUNICATION
Entry for the Table of Contents
AsR6
Terpene Cyclase
OPP
1
residue
Mutation
FPP
2E-humulene
AsR6
AsR6L285M
OPP
NPP
both enantiomers
2
Z-humulene
AsR6 is a type I sesquiterpene cyclase with very low homology to other terpene cyclases. It has a unique bi-magnesium diphosphate
binding site and thus lacks the canonical terpene cyclase aspartate-rich signature sequences. One magnesium is replaced by the e-
ammonium ion of a conserved lysine. AsR6 can use FPP and both enantiomers of NPP as substrate. In addition, a single mutation
directs formation of 2E or 2Z humulene with very high selectivity.
@
carsten_schotte; @RussellCox_Chem
6
This article is protected by copyright. All rights reserved.