Importance of Saccharomyces cerevisiae oxidosqualene-lanosterol cyclase tyrosine 707 residue for chair-boat bicyclic ring formation and deprotonation reactions

Article abstract of DOI:10.1021/ol802036c

(Chemical Equation Presented) A contact mapping strategy was applied to identify putative amino acid residues that influence the oxidosqualene- lanosterol B-ring cyclization reaction. A bicyclic intermediate with two altered deprotonation products, in conjunction with lanosterol, were isolated from the ERG7^Y707X mutants, indicating that the Tyr707 residue may play a functional role in stabilizing the chair-boat bicyclic C-8 cation and the lanosteryl C-8/C-9 cation intermediates.

Full text of DOI:10.1021/ol802036c

ORGANIC
LETTERS
2008
Vol. 10, No. 21
4959-4962
Importance of Saccharomyces
cerevisiae Oxidosqualene-Lanosterol
Cyclase Tyrosine 707 Residue for
Chair-Boat Bicyclic Ring Formation and
Deprotonation Reactions
Tung-Kung Wu,* Tsai-Ting Wang, Cheng-Hsiang Chang, Yuan-Ting Liu, and
Wen-Shiang Shie
Department of Biological Science and Technology, National Chiao Tung UniVersity,
300, Hsin-Chu, Taiwan (Republic of China)
tkwmll@mail.nctu.edu.tw
Received September 1, 2008
ABSTRACT
A contact mapping strategy was applied to identify putative amino acid residues that influence the oxidosqualene-lanosterol B-ring cyclization
reaction. A bicyclic intermediate with two altered deprotonation products, in conjunction with lanosterol, were isolated from the ERG7Y707X
mutants, indicating that the Tyr707 residue may play a functional role in stabilizing the chair-boat bicyclic C-8 cation and the lanosteryl
C-8/C-9 cation intermediates.
Enzyme active sites provide highly optimized microenvi-
ronments for enzyme activity. Changes at the active site
center can have large effects on enzyme activity and provide
a promising way to access new reactive functionalities.1
However, in order to redesign proteins to acquire a new or
higher specificity and activity, it is a prerequisite to identify
important plasticity residues in the active site cavity with
the capability of producing diverse product profiles. In
parallel, the characterization of mutated amino acids that
successfully alter product specificity forms the basis of
understanding the contribution that specific amino acids make
to catalysis. We report herein the elucidation of plasticity
residues within Saccharomyces cereVisiae oxidosqualene-
lanosterol cyclase (ERG7), which represents the first step in
undertsanding the enzyme’s structure-function relationships
and opens up the possibility of generating new reactivities.
The oxidosqualene-lanosterol cyclase-templated cycliza-
tion/rearrangement of (3S)-2,3-oxidosqualene (OS, 1) to
lanosterol (LA, 2) proceeds through a series of discrete,
conformationally rigid, partially cyclized carbocation inter-
mediates.2-5 The reaction for LA formation is characterized
by a precise chair-boat (C-B) B-ring conformation, 6-6-
(1) (a) Penning, T. M.; Jez, J. M. Chem. ReV. 2001, 101, 3027–3046.
(b) Tao, H.; Cornish, V. W. Curr. Opin. Chem. Biol. 2002, 6, 858–864. (c)
Toscano, M. D.; Woycechowsky, K. J.; Hilvert, D. Angew. Chem., Int. Ed.
2007, 46, 3212–3236.
(2) (a) van Tamelen, E. E. J. Am. Chem. Soc. 1982, 104, 6480–6481.
(b) Poralla, K. Bioorg. Med. Chem. Lett. 1994, 4, 285–290. (c) Poralla, K.;
Hewelt, A.; Prestwich, G. D.; Abe, I.; Reipen, I.; Sprenger, G. Trends
Biochem. Sci. 1994, 19, 157–158
.
10.1021/ol802036c CCC: $40.75
Published on Web 10/09/2008
2008 American Chemical Society

6-5 tetracyclic protosteryl cation formation, hydride and
methyl group migration to the lanosteryl C-8/C-9 cation, and
highly specific deprotonation. The formation of pentacyclic
ꢀ-amyrin, catalyzed by ꢀ-amyrin synthase, proceeds similarly
with the exception of chair-chair (C-C) bicyclic conforma-
tion, and the cationic cyclization of the tetracyclic dammare-
nyl cation, which is followed by annulation of a fifth ring.
Among various triterpene biosyntheses, the concerted or
nonconcerted A/B ring cyclization mechanism and the C-B
versus C-C B-ring conformation have been of specific interest
to most researchers.3,6 In the oxidosqualene-lanosterol cy-
clase-catalyzed cyclization, although it was proposed that
expoxide opening was concerted with A-ring formation or
even B-ring formation, truncated monocyclic cyclization
products have been isolated from various S. cereVisiae ERG7
mutants.4a,5a-d,6a Furthermore, although B-ring boats are
well-represented in triterpenoids with three or more rings,
biosynthesis of bicyclic triterpene alcohol has only been
achieved with C-C trans-decalin derivatives.7 No known
bicyclic triterpene alcohols derived from a C-B bicyclic
cation have been reported. The energetically unfavorable boat
conformation in oxidosqualene-lanosterol cyclase and oxi-
dosqualene-cycloartenol synthase is probably achieved by
enzyme residues through a steric and/or an electronic effect.8
Recently, we have reported on the isolation of enzymatic
products derived from the carbocationic intermediates of
monocyclic, tricyclic, tetracyclic, truncated-rearranged, and
altered deprotonated products produced by single amino acid
replacement of Tyr99, Trp232, His2434, Phe445, Tyr510,
or Phe699 in S. cereVisiae ERG7.5 Previously, it was
predicted that the human oxidosqualene-lanosterol cyclase’s
(human OSC) Tyr98 residue (corresponding to Tyr99 in S.
cereVisiae oxidosqualene-lanosterol cyclase, ERG7) was
spatially positioned to enforce the energetically unfavorable
B-ring boat conformation of OS by pushing the methyl group
at C-8 below the molecular plane.8 However, our site-
saturated mutagenesis experiments established that the prob-
able function of ERG7 Tyr99 is to achieve both the C-B
tricyclic Markovnikov C-14 (lanosterol numbering) cation
stabilization and the regio- or stereochemical control of the
protons at the C-15 position for subsequent deprotonation,
but not B-ring formation.
We previously generated ERG7 wild-type and mutant
homology models, based on human OSC X-ray crystal
structure, to investigate the effects of amino acid substitutions
on the cyclization mechanism and product profiles.5 To
explore the plasticity residues involved in OS B-ring
cyclization, we applied a contact mapping strategy that aims
to probe amino acids within van der Waals radii distance of
the C-8 carbon of OS or LA, based on the assumption that
amino acids within the active site are most likely to contribute
to a catalytic outcome. Mutational studies of plasticity
residues may provide insight into the role of these residues
in the determination of the B-ring configuration or in
generating an evolved enzyme with new reactivities. Among
the various putative ERG7 C-8 contact mapping residues
under investigation, the Tyr707 residue is of particular
interest and was therefore subjected to genetic selection and
product characterization. In this study, we describe the
identification of the ERG7 Tyr707 residue as a crucial
component for B-ring stabilization and/or final deprotonation,
as well as the characterization of the mutational effects on
product specificity/diversity.
We first generated ERG7∆Y707 deletion and ERG7Y707X site-
saturated mutations and expressed them in a yeast HEM1
ERG7 double-knockout mutant TKW14, as previously de-
scribed.5 The genetic selection results of Tyr707X mutants
showed that most mutations at Tyr707 cannot abolish the
activity of ERG7 cyclase with the exception of the Tyr707Arg
mutation (Table 1), indicating that the Tyr707 mutations are
not detrimental to the catalytic activity of ERG7. However,
the loss of activity of ERG7∆707 deletion mutant suggested
that the existence of Tyr707 is essential for the catalytic
function of ERG7. Next, the TKW14[pERG7Y707X] mutant
strains were cultured and nonsaponifiable lipid (NSL) extracts
were isolated for product characterization. Silica gel column
chromatography coupled with thin layer chromatography
(TLC) and gas chromatography-mass spectrometry (GC-
MS) were used to analyze the triterpenoid products with a
molecular mass of 426 Da. The ERG7Y707X mutants showed
one to four distinct products. Three products were identified
as authentic LA (2), 9ꢀ-lanosta-7,24-dien-3ꢀ-ol (3), and
parkeol (4) by comparison with authentic standards using
1H and 13C NMR as well as GC-MS. A fourth distinct
compound, with a 1.5 min reduced retention time compared
to LA on the GC column, was isolated and further character-
(3) (a) Abe, I.; Rohmer, M.; Prestwich, G. D. Chem. ReV. 1993, 93,
2189–2206. (b) Wendt, K. U.; Schulz, G. E.; Corey, E. J.; Liu, D. R. Angew.
Chem., Int. Ed. 2000, 39, 2812–2833. (d) Wendt, K. U. Angew. Chem., Int.
Ed. 2005, 44, 3966–3971.
(4) (a) Corey, E. J.; Cheng, H.; Baker, C. H.; Matsuda, S. P. T.; Li, D.;
Song, X. J. Am. Chem. Soc. 1997, 119, 1289–1296. (b) Hart, E. A.; Hua,
L.; Darr, L. B.; Wilson, W. K.; Pang, J.; Matsuda, S. P. T. J. Am. Chem.
Soc. 1999, 121, 9887–9888. (c) Joubert, B. M.; Hua, L.; Matsuda, S. P. T.
Org. Lett. 2000, 2, 339–341. (d) Matsuda, S. P. T.; Darr, L. B.; Hart, E. A.;
Herrera, J. B. R.; McCann, K. E.; Meyer, M. M.; Pang, J.; Schepmann,
H. G.; Wilson, W. K. Org. Lett. 2000, 2, 2261–2263. (e) Herrera, J. B. R.;
Wilson, W. K.; Matsuda, S. P. T. J. Am. Chem. Soc. 2000, 122, 6765–
6766. (f) Meyer, M. M.; Xu, R.; Matsuda, S. P. T. Org. Lett. 2002, 4, 1395–
1398. (g) Xu, R.; Fazio, G. C.; Matsuda, S. P. T. Phytochemistry 2004, 65,
261–291. (h) Lodeiro, S.; Schulz-Gasch, T.; Matsuda, S. P. T. J. Am. Chem.
Soc. 2005, 127, 14132–14133. (i) Lodeiro, S.; Wilson, W. K.; Shan, H.;
Matsuda, S. P. T. Org. Lett. 2006, 8, 439–442. (j) Lodeiro, S.; Xiong, Q.;
Wilson, W. K.; Kolesnikova, M. D.; Onak, C. S.; Matsuda, S. P. J. Am.
Chem. Soc. 2007, 129, 11213–11222.
(5) (a) Wu, T. K.; Chang, C. H. ChemBioChem 2004, 5, 1712–1715.
(b) Wu, T. K.; Liu, Y. T.; Chang, C. H. ChemBioChem 2005, 6, 1177–
1181. (c) Wu, T. K.; Yu, M. T.; Liu, Y. T.; Chang, C. H.; Wang, H. J.;
Diau, E. W. G. Org. Lett. 2006, 8, 1319–1322. (d) Wu, T. K.; Liu, Y. T.;
Chang, C. H.; Yu, M. T.; Wang, H. J. J. Am. Chem. Soc. 2006, 128, 6414–
6419. (e) Wu, T. K.; Liu, Y. T.; Chiu, F. H.; Chang, C. H. Org. Lett. 2006,
8, 4691–4694. (f) Wu, T. K.; Wen, H. Y.; Chang, C. H.; Liu, Y. T. Org.
Lett. 2008, 10, 2529–2532. (g) Wu, T. K.; Chang, C. H.; Liu, Y. T.; Wang,
T. T. Chem. Rec. 2008, in press.
(6) (a) Gao, D.; Pan, Y. K.; Byun, K.; Gao, J. J. Am. Chem. Soc. 1998,
120, 4045–4046. (b) Hess, B. A., Jr.; Smentek, L. Org. Lett. 2004, 6, 1717–
1720
.
(7) (a) Boar, R. B.; Couchman, L. A.; Jaques, A. J.; Perkins, M. J. J. Am.
Chem. Soc. 1984, 106, 2476–2477. (b) Bennett, G. J.; Harrison, L. J.; Sia,
G. L.; Sim, K. Y. Phytochemistry 1993, 32, 1245–1251. (c) Nguyen,
L. H. D.; Harrison, L. J. Phytochemistry 1998, 50, 471–476. (d) Fu¨ll, C.;
Poralla, K. FEMS Microbial. Lett. 2000, 183, 221–224. (e) Fu¨ll, C. FEBS
Lett. 2001, 509, 361–364. (f) Sato, T.; Hoshino, T. Biosci. Biotechnol.
Biochem. 2001, 65, 2233–2242. (g) Hoshino, T.; Sato, T. Chem. Commun.
2002, 291–301.
(8) Thoma, R.; Schulz-Gasch, T.; D’Arcy, B.; Benz, J.; Aebi, J.;
Dehmlow, H.; Hennig, M.; Stihle, M.; Ruf, A. Nature 2004, 432, 118–
122.
4960
Org. Lett., Vol. 10, No. 21, 2008

the viable mutants, nine of them (Y707X, X ) H, Q, D, E,
S, T, C, G, and A) produced other products besides 2,
including two altered deprotonation products, 3 and 4, as
well as the novel bicyclic product, 5. In the ERG7Y707Q and
ERG7Y707H mutants, large amounts of 5 were generated as
the major product, while 2 was the major product in the other
viable mutations. The ERG7Y707L, ERG7Y707I, and ERG7Y707F
mutants produced only 2, 3, and a slight amount of 4. Six
mutations (Y707X, X ) V, P, M, N, W, K) produced only
2 in line with the wild-type ERG7. Interestingly, most
nonpolar amino acid substitutions formed 2 or other altered
deprotonation products but no truncated cyclization product.
The product profile of the ERG7Y707X mutants suggested
that OS is folded in a chair-boat-chair conformation and
initiates ring cyclization through opening of the oxirane ring
and subsequent A-ring annulation (Scheme 1). A bicyclic
Table 1. Product Profiles of S. cereVisiae TKW14 Expressing
the ERG7∆Y707 Deletion and ERG7Y707X Site-Saturated Mutants9
ERG7mut
5
2
3
4
Y707H
Y707Q
Y707E
Y707D
Y707T
Y707S
Y707C
Y707G
Y707A
Y707L
Y707I
Y707F
Y707V
Y707P
Y707M
Y707N
Y707W
Y707K
Y707R
∆707
83.6
82
12.3
21.8
9.8
21.3
6
28.3
18.9
4.7
4.3
6.2
5.5
13.7
6.4
70.8
67.3
65.1
42.9
47.1
48.2
56.9
85.3
87.9
89.2
100
100
100
100
100
100
10.5
7.6
3.3
13.5
20.5
42.6
16.3
19.9
14.7
12.1
7.7
11.6
15.3
4.3
7.2
4.3
3.1
Scheme 1. Proposed Cyclization/Rearrangement Mechanism
Occurring in the ERG7Y707X Site-Saturated Mutants
No new product
1
ized with NMR (1H, 13C NMR, DEPT, HSQC, HMBC, H
-1H COSY and NOE) and demonstrated to be (9R,10S)-
polypoda-8(26),13E,17E,21-tetraen-3ꢀ-ol (5) based on the
1
following data. The compound showed a distinct H NMR
chemical shift with three unseparated olefinic protons (δ
5.163, 3H) and two methylene protons (δ 4.576, δ 4.748),
as well as five methyl singlets (δ 0.802, 0.952, 1.023, 1.614,
1.720) and two unseparated methyl singlets (δ 1.644). The
13C NMR spectrum revealed the presence of one secondary-
quaternary (δ ) 109.42, 149.26 ppm) and three tertiary-
quaternary substituted double bonds (δ ) 124.96, 134.94;
124.45, 135.05; and 124.54, 131.34 ppm). The HSQC
spectrum showed that the olefinic methylene protons at δ
4.576 and 4.748 are attached to the carbon at 149.26 ppm.
The aforementioned NMR information indicated a bicyclic
triterpene alcohol with three double bonds on the side chain
as well as a double bond of two germinal protons. Further-
more, the presence of NOEs among Me-24/Me-25, Me-25/
H-26, Me-25/H-1, Me-25/H-9, Me-23/H-3, Me-23/H-2, and
Me-23/H-5, as well as the absence of NOEs among Me-23/
H-9, Me-23/ Me-25, Me-24/H-3, Me-24/H-5, and Me-25/
H-5, confirmed the structure to be 5, a C-B 6-6 bicyclic
product with a C-9R hydrocarbon side chain configuration
and ∆8(26),13,17,21 double bonds. This is the first reported
isolation of a bicyclic triterpene alcohol derived from a C-B
bicyclic cation, although a stereoisomer with a C-9ꢀ hydro-
carbon side chain configuration has previously been reported
as a natural product isolated from the bark of Cratoxylum
C-8 cation (lanosterol numbering) was formed without
disruption at the monocyclic C-10 cation position, and it was
followed by deprotonation of Me-26 to produce C-B
truncated 5. Further cyclization of the C- and D-ring
proceeded to produce a tetracyclic protosteryl C-20 cation.
Then skeletal rearrangement of the two hydrides and two
methyl groups, H-17R f 20R, H-13R f 17R, Me-14ꢀ f
13ꢀ, Me-8R f 14R, and/or one hydride shift from H-9ꢀ to
H-8ꢀ generated the lanosteryl C-8/C-9 cation, which under-
went deprotonation at C-8, C-7, and C-11 to yield 2, 3, and
4, respectively.
Tyr707 of ERG7 is a strictly conserved residue in both
squalene cyclases (SHCs) and oxidosqualene cyclases, and
corresponds to Tyr609 in SHC from Alicyclobacillus aci-
docaldarius and Tyr704 in human OSC. Tyr609 of SHC has
been previously suggested to assist the function of Phe365
(corresponds to Phe445 of ERG7), to stabilize the C-8 cation
via cation-π interaction, by both an electronic and steric
effect. Substitution of the Tyr609 of SHC with Phe, Ala,
Leu, Cys, and Ser caused a perturbation in the active site
cochinchinense.7b Comparison of the H and 13C NMR
1
spectra of these two stereoisomers further confirmed the
structural assignment.
The product profiles of each mutant are summarized in
Table 1. No products with m/z ) 426 were observed in the
nonviable mutants, ERG7Y707R and ERG7∆707. However, in
Org. Lett., Vol. 10, No. 21, 2008
4961

and lowered the π-electron density around the C-8 position,
thus quenching the cyclization at the bicyclic stage and
resulting in truncated bicyclic product formation.7d-g The
homology modeling of ERG7 revealed that Tyr707 is
spatially proximal to both the C-10 and C-8 positions of 2,
with the hydroxyl group of the phenolic side chain pointing
toward the B-ring capable of stabilizing the bicyclic cationic
intermediate (Figure 1). The phenolic oxygen of Tyr707 was
Interestingly, there were significant accumulations of 5 in
Tyr707His and Tyr707Gln mutations, whereas no truncated
or altered deprotonation product could be identified in the
Tyr707Asn mutation. A possible explanation for this behav-
ior is the similarity in H-bonding but with different orienta-
tion of His and Gln, thus deviating the electron density
distribution for truncated bicyclic product formation. For the
Tyr707Asn mutation, although a deleted methylene group
from the side chain of Gln may enlarge the active site cavity,
the generated space may be enough to insert a water
molecule, thus compensating for the lost interaction and
restoring the product profile. It is interesting to note that
Tyr707Ile, Tyr707Leu, and Tyr707Phe mutants make altered
deprotonation products. The mutations may destroy the
original H-bonding but retain the ordered water with a slight
shift in active site geomentry. This in turn could cause
relocation of the final acceptors and result in altered
deprotonation product formation. To our surprise, most
nonpolar amino acid substitutions formed 2 without any
truncated cyclization product. Perhaps the loss of H-bonding
in these mutants did not cause an obvious interference in
the active-site cavity with water retention, thus following
the normal pathway to 2 formation. However, the precise
molecular interactions between the enzymatic active site and
the specific product profile still awaits final X-ray structure
determination.
Figure 1. Homology model of wild-type ERG7 complexed with
lanosterol (green) or bicyclic C-8 cation (yellow). The distance of
Tyr707 to the C-8 position of lanosterol or bicyclic C-8 cation is
5.1 and 6.2 Å, respectively.
In summary, oxidosqualene cyclase mutants producing a
truncated C-B bicyclic cyclization intermediate were isolated
for the first time for LA biosynthesis. Formation of the
truncated bicyclic triterpene alcohol as well as altered
deprotonation products in the ERG7Y707X mutants indicated
that the Tyr707 residue plays a functional role in stabilizing
the bicyclic C-8 cation and final lanosteryl C-8/C-9 cation
intermediates. This result provides new structure-function
relationship information indicating how enzymes act to
control the cyclization/rearrangement mechanism. However,
the Tyr707 mutations did not affect the B-ring boat confor-
mation, indicating the importance of other residue(s) in
controlling C-B versus C-C substrate conformation. Further
studies should focus on characterizing the residues that
influence substrate C-B versus C-C conformation and
divergent evolution of new enzymatic functions.
found at a distance of 5.6 and 5.1 Å from C-10 and C-8 of
2, respectively. Interestingly, a distance of ∼6.2 Å was
observed between the phenolic oxygen of Tyr707 and the
bicyclic C-8 cation intermediate.
The product profile analysis of ERG7 mutants revealed
that the truncated bicyclic product accumulated with polar,
acidic, or smaller hydrophobic side chain substitutions.
Substitutions of Tyr707 with polar or acidic side chain groups
may reduce polarity and allow interaction with the bicyclic
C-8 cation or other residues nearby. Consistent with the
results is the observation that in human OSC, the Tyr704 is
H-bonded to the backbone carbonyl group of Trp581
(corresponding to Trp587 of ERG7) through a water bridge.
The reduced polarity may destroy the H-bonding of Tyr707
to the backbone carbonyl group of Trp587 or form a new
interaction with other residues, thus destabilizing the bicyclic
cation or disturbing the active site structure. Alternatively,
substitution of Tyr707 with small hydrophobic side chains
such as Gly or Ala, resulting in additional space in the cavity
and complete loss of polarity, may disturb the H-bonding
pattern with Trp587 and direct compound 5 formation.
Acknowledgment. The National Science Council of the
Republic of China and the MOE ATU Program funded this
research under Contract NSC-96-2113-M-009-003 and NSC-
96-2113-M-009-004.
Supporting Information Available: Details of experi-
mental procedures, GC-MS and NMR spectra, and and NMR
assignments of (9R,10S)-polypoda-8(26),13E,17E,21-tetraen-
3ꢀ-ol. This material is available free of charge via the Internet
at http://pubs.acs.org.
(9) The TKW14[pERG7Y707X] strain expressed ERG7Y707X as the only
oxidosqualene cyclase. The in vivo product profiles represent metabolic
equilibrium of lanosterol and related triterpenes under physiological
conditions, which may vary from that of in vitro assay.
OL802036C
4962
Org. Lett., Vol. 10, No. 21, 2008