Remarkable synergistic effect between MonBI and MonBII on epoxide opening reaction in ionophore polyether monensin biosynthesis

Article abstract of DOI:10.1016/j.tetlet.2011.07.145

Enzymatic epoxide-opening cascade is one of the key biosynthetic processes for constructing structurally diverse polyethers. Here we report the first in vitro analysis of the cyclization catalyzed by two epoxide hydrolases, MonBI and MonBII involved in mo

Full text of DOI:10.1016/j.tetlet.2011.07.145

Tetrahedron Letters 52 (2011) 5277–5280
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Remarkable synergistic effect between MonBI and MonBII on epoxide
opening reaction in ionophore polyether monensin biosynthesis
Kyohei Sato a, Atsushi Minami a, Toyoyuki Ose b, Hiroki Oguri a, Hideaki Oikawa a,
⇑
a Division of Chemistry, Graduate School of Science, Hokkaido University, Sapporo 060-0810, Japan
b Faculty of Pharmaceutical Sciences, Hokkaido University, Sapporo 060-0810, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Enzymatic epoxide-opening cascade is one of the key biosynthetic processes for constructing structurally
diverse polyethers. Here we report the first in vitro analysis of the cyclization catalyzed by two epoxide
hydrolases, MonBI and MonBII involved in monensin biosynthesis, using simple epoxy-alcohols and the
unprecedented synergistic effect on the epoxide-opening activity between these epoxide hydrolases.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 5 July 2011
Revised 26 July 2011
Accepted 29 July 2011
Available online 6 August 2011
Keywords:
Polyether
Monensin
Epoxide-opening cascades
Epoxide hydrolase
Ionophore polyethers,1 one of the major groups of natural poly-
ketides, have an unique polyether skeleton composed of small
numbers of tetrahydrofuran (THF) and tetrahydropyran (THP) rings
connected via a carbon–carbon bond. Although the number of ether
rings resulting from different cyclization modes is reflected the
structural diversity of this type of natural product, as represented
by the Cane–Celmer–Westley unified biosynthetic scheme,2 it is
suggested that polyether skeletons are constructed by a common
strategy involving nucleophilic epoxide-opening cascades of the
polyepoxides after enzymatic asymmetric epoxidation of the linear
polyolefins. This simple yet versatile strategy for the biosynthesis of
complex polyether metabolites prompted chemists to synthesize
ionophore antibiotics using epoxide-opening cascades.3,4 The
successful biomimetic synthesis of those polyether systems
supports the biosynthetic proposal.
To investigate the detailed biosynthesis for polyether construc-
tion, four polyether biosynthetic gene clusters involved in monen-
sin (1) (mon),5 nanchangmycin (nan),6 nigericin (nig),7 and
lasalocid (lsd, las) biosynthesis8 and the gene cluster involved in
tetronomycin (tet) biosynthesis,9 a related natural product, have
so far been characterized and it revealed that the putative key genes
for polyether construction, the epoxidase gene and epoxide hydro-
lase gene, are conserved in these biosynthetic gene clusters. Gene
disruption experiments of the epoxidase gene (monCI) and epoxide
hydrolase gene (monBI and monBII) in monensin biosynthesis
demonstrated that, as expected in the Cane–Celmer–Westley
model, epoxidation of the (E,E,E)-triene precursor and epoxide-
opening reaction of the corresponding epoxide is catalyzed by Mon-
CI, and MonBI and/or MonBII, respectively.10,11 However, the de-
tailed catalytic mechanism of epoxidase and epoxide hydrolase
(EH) at the enzymatic level remained unclear.
Recently, we have conducted functional analysis of epoxide
hydrolase Lsd19 involved in lasalocid A biosynthesis as a model
system of enzymatic epoxide-opening cascades. In 2008, we iden-
tified the function of Lsd19 in catalysis of the epoxide-opening
reactions with an energetically favored 5-exo cyclization and an
energetically disfavored 6-endo cyclization to afford a polyether
skeleton of lasalocid from the putative biosynthetic intermediate,
bisepoxyprelasalocid.12 Following this first in vitro characteriza-
tion, we investigated substrate specificity of Lsd19 using structur-
ally diverse bisepoxides.13 As reported for limonene epoxide
hydrolase and other EHs,14 Lsd19 also accepted various structur-
ally simplified bisepoxides. In addition, specificity toward the ste-
reochemistry of bisepoxide moiety revealed that the second
epoxide-opening reaction can be altered by the stereochemistry
at the terminal epoxide. The relaxed substrate tolerance of Lsd19
allowed us to investigate the detailed catalytic mechanism using
these substrate analogs. A site-directed mutagenesis study using
these simplified analogs revealed the conserved acidic amino acid
pair, located at the N- and C-terminal domain, as key catalytic res-
idues for the epoxide-opening reaction.15 Further domain dissec-
tion analysis clearly showed that the epoxide-opening cascades
involving 5-exo and 6-endo cyclization are separately catalyzed
by two catalytic domains, named Lsd19A (N-terminal region) and
Lsd19B (C-terminal region). These results demonstrated a simple
⇑
Corresponding author.
E-mail address: hoik@sci.hokudai.ac.jp (H. Oikawa).
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.07.145

5278
K. Sato et al. / Tetrahedron Letters 52 (2011) 5277–5280
strategy for polyether construction in which EH domain indepen-
dently catalyzes the single epoxide-opening reaction in a stepwise
manner. However, in monensin biosynthesis, the hypothesis that a
single EH domain/subunit catalyzes a single epoxide-opening reac-
tion cannot be applied because the number of domains/subunits
(MonBI and MonBII) is not the same as that of the epoxide-opening
reactions (triepoxide intermediate). This probably reflects previous
gene disruption experiments of MonBI and/or MonBII, which did
not yield the expected partially cyclized intermediates but essen-
tially the same products. We herein report the enzymatic epox-
ide-opening reaction with MonBI and MonBII using structurally
simple substrate analogs and the unprecedented synergistic effect
of MonBI and MonBII.
Epoxide-opening cascades in monensin biosynthesis are pro-
posed as follows (Scheme 1): (1) EH(s) selects a single hemiacetal
from the equilibrium mixture which undergoes 5-exo cyclization to
give the spiroketal intermediate 4; (2) EH(s) catalyzes the second
and third 5-exo cyclization to afford the cyclic ether intermediates
5 and 6. Given the broad substrate tolerance of Lsd19 including
epoxide stereochemistry, we designed a simplified epoxy-alcohol
(8a–d) to examine the enzymatic activity of MonBI and MonBII
(Scheme 2). The model substrate 8 is an analog of the epoxy-alco-
hol part at C12–C19 which may also mimic the C8–C13 and C16–
C21 part when a series of acyl groups ranging from C2 to C10 chain
length is employed.
Monoepoxy-alcohols (8a–d) were synthesized as described in
Scheme 2. It commenced with the protection of commercially
available (S)-methyl 3-hydroxy-2-methylpropanoate as silyl ether
followed by reduction with DIBAL to afford 11. After Swern oxida-
tion, aldehyde was treated with vinyl lithium reagent to give the
allyl alcohol 12 as a diastereomeric mixture. Synthesis of the de-
sired E-trisubstituted olefin 13 was achieved by the ortho ester
Claisen rearrangement.16 Treatment of methyl ester 13 with MeM-
gBr and subsequent removal of silyl group afforded the common
diol intermediate 14. Finally, condensations of various acyl groups
followed by Shi’s asymmetric epoxidation17 afforded a series of the
desired monoepoxy-alcohol (8a–d). The diastereoselectivity of 8a–
d was determined as 10:1 to 20:1 (R/S stereochemistry) based on
1H NMR analysis.
A synthetic monBI gene with codons optimized for expression in
Escherichia coli18 and monBII gene were separately cloned and ex-
pressed in E. coli BL21-Gold (DE3) as an N-terminal His6-tagged
protein and individual recombinant proteins were purified by Ni-
NTA column chromatography to a state of homogeneity as judged
by SDS–PAGE (Fig. S1).19 In addition, as reported in our previous
paper, alignment of MonBI, MonBII, and Lsd19 revealed that the
acidic amino acid pair corresponding to the catalytic residue for
the Lsd19 reaction (Lsd19A; D38-E65, Lsd19B; D170-E197) is con-
served in MonBI (D37 and E64) and MonBII (D38 and E65). There-
fore, alanine mutants were also constructed and purified by the
same procedure for wild type enzymes (Fig. S1).19
In our study with Lsd19, we found that various epoxy-alcohols
were prone to cyclization affording a 5-exo cyclization product
according to Baldwin’s rule.20 Therefore, prior to the detailed func-
tional analysis of MonBI and MonBII, the feasibility of 5-exo cycli-
zation of synthetic monoepoxy-alcohol (8a–d) was examined in
several pH ranges. LC–MS analysis of the reaction product revealed
that, in all cases, no cyclized product (9a–d) was detected in the pH
range of 6.5–8.0 even after 6 h incubation at 37 °C. This facilitated
the distinction of MonBI and/or MonBII catalyzing 5-exo cyclization
from the nonenzymatic reaction. Based on this result, the following
reaction conditions for the functional analysis of MonBI and Mon-
BII were established: 20 mM MOPS (pH 7.0), 10% glycerol, 8
lM
monoepoxy-alcohol, 8
l
M MonBI, and/or MonBII.21 Then, monoep-
oxy-alcohols (8a–d) were separately subjected to the MonBI or
MonBII reaction, and subsequent LC–MS analysis revealed a new
peak only in the MonBII reaction (Figs. 1, S2). Since the retention
time of the product is identical to those of authentic samples 9a–
d prepared by the acid treatment, the MonBII reaction product
was confirmed as a 5-exo cyclization product (THF-product). C10
analog represented the highest activity in the MonBII reaction
(C2: 0%, C6: 12%, C7: 13%, C10: 30%) (Fig. S2C) In contrast, MonBII
mutant (BII-D38A and BII-E65A) showed no catalytic activity to-
ward the C10 analog. More interestingly, the epoxide opening
activity was dramatically enhanced with the addition of MonBI
to the MonBII reaction mixture (Fig. 1). In this MonBII//MonBI sys-
tem (described in the order of EH//partner EH), C6 analog showed
the highest activity among those tested analogs (C2: 2%, C6: 95%,
C7: 91%, C10: 90%) (Figs. 1d, S2). This enhancement of cyclization
activity was also observed using MonBI-E64A as a partner EH
(Fig. 1e). In addition, the activity gradually increased depending
on the MonBI-E64A concentration (Fig. S3) Thus the observed
intriguing synergistic effect on the epoxide-opening reaction cata-
lyzed by MonBII in the presence of partner EH prompted us to
examine the MonBI catalyzing reaction again in the presence of
MonBII-D38A mutant. As expected, enhancement of MonBI cata-
lytic activity was observed in this MonBI//MonBII-D38A system
(C2: 0%, C6: 22%, C7: 19% and C10: 9%) (Figs. 1f, S2). In contrast,
a small but detectable amount of THF-product was observed in
the MonBI-E64A//MonBII-D38A system using C6 analog (4%)
(Fig. 1g).
The discovery of specific thioesterases catalyzing hydrolysis
of several PKS-bound polyether intermediates in the biosynthesis
of monensin and nanchangmycin showed that actual substrates
Scheme 1. Enzymatic epoxide-opening cascades in monensin biosynthesis.

K. Sato et al. / Tetrahedron Letters 52 (2011) 5277–5280
5279
O
Scheme 2. Reagents and conditions: (a) TBDPSCl, imidazole, THF, quant.; (b) DIBAL, Et2O, 96%; (c) (COCl)2, DMSO, Et3N, CH2Cl2; (d) 2-bromo propene, t-BuLi, Et2O, two steps
60%; (e) CH3CH(OMe)3, 140 °C, 70%; (f) MeMgBr, THF, 63%; (g) TBAF, THF, 83%; (h) fatty acid, DCC, DMAP, CH2Cl2, 70–99%; (i) A, OxoneR, K2CO3, Bu4NHSO4, CH3CN–
CH2(OMe)2–H2O, 0 °C, 51–69%.
(A)
O
O
MonBI; D37-E64
MonBII; D38-E65
MonBI
MonBII
-
H
R
O
E
O
E
O
R
O
O
O
O
synergistic
effect
OH
-
H
D
O
D
OH
(B)
MonBI
inactive
MonBI-E64A
MonBII-D38A
inactive
active
MonBII
weakly active
inactive
MonBII-D38A
MonBII-E65A
MonBI
MonBII
inactive
Figure 2. (A) Epoxide opening cascades catalyzed by MonBI and MonBII. (B)
Summary of synergistic effect.
intermediates 3 and/or 4 in the active site. This hypothesis is sup-
ported by the structural model of MonBII based on the Lsd19 crystal
structure that MonBII has an active site to accept a relatively long
polyketide chain compared with Lsd19A (unpublished result).
In the previous gene disruption experiments on either monBI or
monBII or both, all mutants afforded mixtures of the same interme-
diates including 9-epimonensin instead of natural-type polyether
product. The intriguing synergistic effect between MonBI and Mon-
BII for their cyclization activity might explain the puzzling obser-
vation as described above because disruption of each gene causes
overall decrease of epoxide opening activity (Fig. 2). The fact that
both monBI and monBII homolog are highly conserved either as a
separated EH system (NigBI//NigBII) in a nigericin biosynthetic
gene cluster or a fused EH system [Lsd19 (Lsd19A//Lsd19B) and
NanI (NanIA//NanIB)] in lasalocid, and a nanchangmycin biosyn-
thetic gene cluster implies that the synergistic effect is a common
feature of the enzymatic epoxide-opening cascades in polyether
biosynthesis. Although DNA methyltransferase shows a similar
synergistic function,23 to our knowledge, the effect on epoxide-
opening activity is unprecedented and probably a characteristic
for the biosynthetic construction of the polyether system.
Figure 1. LC–MS profiles of the reaction products with (a) control experiment, (b)
MonBI reaction, (c) MonBII reaction, (d) MonBII//MonBI reaction, (e) MonBII//
MonBI-E64A reaction, (f) MonBI//MonBII-D38A reaction, (g) MonBI-E64A//MonBII-
D38A reaction.
of EHs are PKS-bound polyepoxides, which are structurally complex
and thus difficult to synthesize.22 The in vitro analyses of MonBI
and MonBII described above are the first direct evidence that both
act as an epoxide hydrolase to catalyze 5-exo cyclization. This
clearly indicates that the use of simple epoxy-alcohol analogs offers
a practical approach to the analysis of EHs involved in ionophore
polyether biosynthesis. Loss of the cyclization activity in MonBII-
D38A, MonBII-E65A, and MonBI-E64A//MonBII-D38A reactions
indicates that each acidic amino acid pair is the catalytic residue
for the epoxide-opening reaction (Fig. 2). This is in accordance with
our previous results on Lsd19 and reveals that the highly conserved
motifs are key catalytic residues of various EHs found in the poly-
ether biosynthetic gene cluster. Clear preference of MonBII toward
C6–C10 analogs showed that MonBII recognizes acyl chain moiety
with medium chain length as an imitator of the putative

5280
K. Sato et al. / Tetrahedron Letters 52 (2011) 5277–5280
8. (a) Migita, A.; Watanabe, M.; Hirose, Y.; Watanabe, K.; Tokiwano, T.; Kinashi,
In summary, we successfully observed enzymatic activities of
H.; Oikawa, H. Biosci., Biotechnol., Biochem. 2009, 73, 169–176; (b) Smith, L.;
Hong, H.; Spencer, J. B.; Leadlay, P. F. ChemBioChem 2008, 9, 2967–2975.
9. Demydchuk, Y.; Sun, Y.; Hong, H.; Staunton, J.; Spencer, J. B.; Leadlay, P. F.
ChemBioChem 2008, 9, 1136–1145.
10. Bhatt, A.; Stark, C. B. W.; Harvey, B. M.; Gallimore, A. R.; Demydchuk, Y. A.;
Spencer, J. B.; Staunton, J.; Leadlay, P. F. Angew. Chem., Int. Ed. 2005, 44, 7075–
7078.
11. Gallimore, A. R.; Stark, C. B. W.; Bhatt, A.; Harvey, B. M.; Demydchuk, Y.;
Bolanos-Garcia, V.; Fowler, D. J.; Staunton, J.; Leadlay, P. F.; Spencer, J. B. Chem.
Biol. 2006, 13, 453–460.
12. Shichijo, Y.; Migita, A.; Oguri, H.; Watanabe, M.; Tokiwano, T.; Watanabe, K.;
Oikawa, H. J. Am. Chem. Soc. 2008, 130, 12230–12231.
13. Matsuura, Y.; Shichijo, Y.; Minami, A.; Migita, A.; Oguri, H.; Watanabe, M.;
Tokiwano, T.; Watanabe, K.; Oikawa, H. Org. Lett. 2010, 12, 2226–2229.
14. (a) van der Werf, M. J.; Orru, R. V. A.; Overkamp, K. M.; Swarts, H. J.; Osprian, I.;
Steinreiber, A.; de Bont, J. A. M.; Faber, K. Appl. Microbiol. Biotechnol. 1999, 52,
380–385; (b) Ueberbacher, B. T.; Oberdorfer, G.; Gruber, K.; Faber, K.
ChemBioChem 2009, 10, 1697–1704.
the epoxide hydrolase MonBI and MonBII in monensin biosynthe-
sis using simple monoepoxy-alcohol as an imitator of structurally
complex intermediate. Furthermore, the unexpected synergistic
effect found in this study advances our understanding of the enzy-
matic epoxide-opening cascades for polyether construction in the
biosynthesis of ionophore antibiotics. Our approach using simple
substrate analogs is proven to be effective for the functional anal-
ysis of EHs because the bottleneck of the biochemical analysis of
EHs exists in the synthesis of the putative polyepoxide substrates.
To examine the cooperative effect of polyether EHs, we are cur-
rently studying further various combinations of other EHs such
as Lsd19 and its equivalents.
Acknowledgments
15. Minami, A.; Migita, A.; Inada, D.; Hotta, K.; Watanabe, K.; Oguri, H.; Oikawa, H.
Org. Lett. 2011, 13, 1638–1641.
16. Patel, D. V.; VanMiddlesworth, F.; Donaubauer, J.; Gannett, P.; Sih, C. J. J. Am.
Chem. Soc. 1986, 108, 4603–4614.
17. Wang, Z. X.; Tu, Y.; Frohn, M.; Zhang, J. R.; Shi, Y. J. Am. Chem. Soc. 1997, 119,
11224–11235.
This work was supported by the Grants-in-Aid for Scientific Re-
search [22108002 to H.O. and 21710213 to A.M.] from the Japan
Society for the Promotion of Science (JSPS) and partially supported
by the Naito Foundation.
18. A synthetic monBI gene was purchased from GeneArt AG.
19. The subcloned monBI and monBII gene were ligated into the pColdI (Takara) to
give pColdI-monBI and pColdI-monBII. Each plasmids were separately
transformed into E. coli BL21-Gold (DE3) (Stratagene) for overexpression.
Purification of each enzyme was carried out according to Ref. 12 Mutations
were introduced into appropriate plasmids by polymerase chain reaction using
respective primers as described in the Table S1. Phosphorylation of 50-OH
termini and ligation were carried out according to Ref. 15.Transformation and
purification was applied the same procedure as described above.
20. Baldwin, J. E. J. C. S. Chem. Commun. 1976, 734–736.
Supplementary data
Supplementary data (LC–MS profiles) associated with this arti-
cle can be found, in the online version, at doi:10.1016/
j.tetlet.2011.07.145.
21. Typical conditions are as follows; a reaction mixture (50
lL of MOPS buffer, pH
References and notes
7.0) containing 8 M of the substrate, 8 M of enzyme(s), and 10% of glycerol
l
l
was incubated at 37 °C for 6 h. The reaction was quenched by the addition of
methanol, and the resultant mixture was vortexed and centrifuged at
12,000 rpm. The supernatant was directly analyzed by LC–MS.
1. Dutton, C. J.; Banks, B. J.; Cooper, C. B. Nat. Prod. Rep. 1995, 12, 165–181.
2. Cane, D. E.; Celmer, W. D.; Westley, J. W. J. Am. Chem. Soc. 1983, 105, 3594–3600.
3. Faul, M. M.; Huff, B. E. Chem. Rev. 2000, 100, 2407–2473.
4. Nicolaou, K. C.; Frederick, M. O.; Aversa, R. J. Angew. Chem., Int. Ed. 2008, 47,
7182–7225.
5. Oliynyk, M.; Stark, C. B. W.; Bhatt, A.; Jones, M. A.; Hughes-Thomas, Z. A.;
Wilkinson, C.; Oliynyk, Z.; Demydchuk, Y.; Staunton, J.; Leadlay, P. F. Mol.
Microbiol. 2003, 49, 1179–1190.
22. (a) Harvey, B. N.; Hong, H.; Jones, M. A.; Hughes-Thomas, Z. A.; Goss, R. M.;
Heathcote, M. L.; Bolanos-Garcia, V. M.; Kroutil, W.; Staunton, J.; Leadlay, P. F.;
Spencer, J. B. ChemBioChem 2006, 7, 1435–1442; (b) Liu, T.; You, D.; Valenzano,
C.; Sun, Y.; Li, J.; Yu, Q.; Zhou, X.; Cane, D. E.; Deng, Z. Chem. Biol. 2006, 13, 945–
955; (c) Liu, T.; Lin, X.; Zhou, X.; Deng, Z.; Cane, D. E. Chem. Biol. 2008, 15, 449–
458.
23. Li, J.-Y.; Pu, M.-T.; Hirasawa, R.; Li, B.-Z.; Huang, Y.-N.; Zeng, R.; Jing, N.-H.;
Chen, T.; Li, E.; Sasaki, H.; Xu, G.-L. Mol. Cell. Biol. 2007, 27, 8748–8759.
6. Sun, Y.; Zhou, X.; Dong, H.; Tu, G.; Wang, M.; Wang, B.; Deng, Z. Chem. Biol.
2003, 10, 431–441.
7. Harvey, B. M.; Mironenko, T.; Sun, Y.; Hong, H.; Deng, Z.; Leadlay, P. F.;
Weissman, K. J.; Haydock, S. F. Chem. Biol. 2007, 14, 703–714.