The cysteine 703 to isoleucine or histidine mutation of the oxidosqualene-lanosterol cyclase from Saccharomyces cerevisiae generates an iridal-type triterpenoid

Article abstract of DOI:10.1016/j.biochi.2012.06.014

The Cys703 to Ile or His mutation within Saccharomyces cerevisiae oxidosqualene-lanosterol cyclase ERG7 (ERG7^C703I/H) generates an unusual truncated bicyclic rearranged intermediate, (8R,9R,10R)-polypoda-5,13E, 17E,21-tetraen-3β-ol, related to iridal-skeleton triterpenoid. Numerous oxidosqualene-cyclized truncated intermediates, including tricyclic, unrearranged tetracyclic with 17α/β exocyclic hydrocarbon side chain, rearranged tetracyclic, and chair-chair-chair tricyclic intermediates (compounds 3-9), were also isolated from the ERG7^C703X site-saturated mutations or the ERG7^F699T/C703I double mutation, indicating the functional role of the Cys703 residue in stabilizing the bicyclic C-8 cation and the rearranged intermediate or interacting with Phe699, and opened a new avenue of engineering ERG7 for producing biological active agents.

Full text of DOI:10.1016/j.biochi.2012.06.014

Biochimie 94 (2012) 2376e2381
Contents lists available at SciVerse ScienceDirect
Biochimie
journal homepage: www.elsevier.com/locate/biochi
Research paper
The cysteine 703 to isoleucine or histidine mutation of the
oxidosqualene-lanosterol cyclase from Saccharomyces cerevisiae
generates an iridal-type triterpenoid
Cheng-Hsiang Chang 1, Yi-Chi Chen 1, Sheng-Wei Tseng, Yuan-Ting Liu, Hao-Yu Wen, Wen-Hsuan Li,
*
Chiao-Ying Huang, Cheng-Yu Ko, Tsai-Ting Wang, Tung-Kung Wu
Department of Biological Science and Technology, National Chiao Tung University, 300 Hsin-Chu, Taiwan
a r t i c l e i n f o
a b s t r a c t
Article history:
The Cys703 to Ile or His mutation within Saccharomyces cerevisiae oxidosqualene-lanosterol cyclase ERG7
(ERG7C703I/H) generates an unusual truncated bicyclic rearranged intermediate, (8R,9R,10R)-polypoda-
Received 16 February 2012
Accepted 7 June 2012
Available online 23 June 2012
5,13E,17E,21-tetraen-3
b
-ol, related to iridal-skeleton triterpenoid. Numerous oxidosqualene-cyclized
exocyclic hydrocarbon
truncated intermediates, including tricyclic, unrearranged tetracyclic with 17a/b
side chain, rearranged tetracyclic, and chairechairechair tricyclic intermediates (compounds 3e9), were
also isolated from the ERG7C703X site-saturated mutations or the ERG7F699T/C703I double mutation, indi-
cating the functional role of the Cys703 residue in stabilizing the bicyclic C-8 cation and the rearranged
intermediate or interacting with Phe699, and opened a new avenue of engineering ERG7 for producing
biological active agents.
Keywords:
Oxidosqualene-lanosterol cyclase
Site-saturated mutagenesis
Iridal
Homology modeling
Marneral synthase
Ó 2012 Elsevier Masson SAS. All rights reserved.
1. Introduction
and characterized. Reported mutants traverse chaireboatechair to
chairechairechair substrate prefolded conformation and/or affect
Oxidosqualene-lanosterol cyclase (OSC or ERG7) cyclizes (3S)-
2,3-oxidosqualene (OS, 1) to lanosterol (2) in mammals and fungi
and some higher plants [1,2]. The intricate and yet highly stereo-
and regio-specific cationic mechanism, which comprises prefolded
substrate conformation, epoxide protonation and opening, four
consecutive ring annulations, 1,2-shifted hydride and methyl
groups migration, and final specific deprotonation, is of consider-
able interest both in considerations on the origin of sterol biosyn-
thesis and protein engineering for pharmaceutical applications
[3e11].
Protein engineering of enzyme active sites using site-saturated
mutagenesis offers promising abilities both for studying
structureefunction-mechanism relationships of proteins and for
creating new sequences with improved or novel properties
[12e14]. We previously utilized site-saturated mutagenesis to
elucidate several plasticity residues within Saccharomyces cer-
evisiae ERG7, which are critical for enzyme catalytic activity and/or
product specificity/diversity [15e22]. Numerous oxidosqualene-
cyclized truncated and rearranged products have been isolated
carbocation intermediate stabilization as well as enzyme plasticity
to generate single or multiple products.
On the other hand, biochemical characterizations of OSCs
purified from various sources have also significantly contributed
mechanistic insights into the cyclization/rearrangement cascade.
For example, protein purification and chemical inactivation of the
pig liver, bovine liver, S. cerevisiae, and Candida albicans OSC
enzymes with sulfhydryl group specific agents suggested that at
least two cysteine residues are critical for enzymatic activity, one
directly involved in catalysis, with a second being not directly
participating in the enzymatic activity, but may be involved in
maintenance of the proper enzymatic conformation [23e26].
However, no precise determination of the locations of putative
cysteine residues involved in the catalysis or conformation has
been reported.
Multiple sequences alignment results showed that Cys703 of
ERG7 is highly conserved among various OSCs from different
species (Fig. 1). The produced homology model of ERG7, based on
the X-ray structure of human OSC, revealed that Cys703 is a second-
tier residue located proximal to the previously identified first-tier
Phe699 residue with a distance of approximately 3.4 Å (Fig. 2)
[27,28]. The combination of these observations suggested that
Cys703 of ERG7 may be involved in catalysis directly or indirectly
through the interaction with proximal residues located within the
* Corresponding author. Tel.: þ886 3 5729287; fax: þ886 3 5725700.
E-mail address: tkwmll@mail.nctu.edu.tw (T.-K. Wu).
1
These authors contributed equally to this work.
0300-9084/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.biochi.2012.06.014

C.-H. Chang et al. / Biochimie 94 (2012) 2376e2381
2377
and a final extension at 68 ꢀC for 8 min. The mutations were confirmed
by DNA sequencing and subsequently electroporated into the yeast
strain TKW14 and selected for growth on SD þ Ade þ Lys þ His þ
Met þ Ura þ hemin þ G418 þ Erg plates. The plasmids were then
selected on SD þ Ade þ Lys þ His þ Met þ Ura þ hemin þ G418 þ 5-
FOA plates for complementation of cyclase activity as described
previously. Transformants were grown in SD þ Ade þ Lys þ His þ
Met þ Ura þ hemin þ G418 þ Erg medium for non-saponifiable lipid
(NSL) extraction and column chromatography. The NSL extract was
fractionated by silica gel column chromatography and assayed by gas
chromatographyemass spectrometry (GCeMS) to examine triterpe-
noid products with a molecular mass of m/z ¼ 426 as described
previously.
Fig. 1. Multiple sequence alignment of oxidosqualene-lanosterol cyclase (OSC). The
sequences shown are Homo sapiens OSC (HsaOSC), Mus musculus OSC (MmuOSC),
Rattus norvegicus OSC (RnoOSC), Saccharomyces cerevisiae OSC (SceOSC), Candida
albicans OSC (CalOSC), Pneumocystis carinii OSC (PcaOSC), Cephalosporium caerulens
(CcaOSC), and Alicyclobacillus acidocaldarius SHC (AacSHC), respectively. The consensus
sequences are boxed in black. These highlighted amino acid residues indicate the
similarity of OSCs among different species.
2.2. Chemical shifts of (8R,9R,10R)-polypoda-5,13E,17E,
21-tetraen-3
b-ol
Chemical 1H NMR (600 MHz, CD2Cl2) spectra showed distinct
protein’s active site cavity surface, particularly Phe699. To further
substantiate the functional role of the Cys703 in ERG7 catalysis and
its interaction with Phe699, we performed the site-saturated
mutagenesis of Cys703 and the Phe699/Cys703 double mutations,
and characterized the product profiles.
chemical shifts of four vinylic methyl signals ( 1.58, s, 3 Me, Me-27,
Me-28, and Me-29; 1.66, s, 1 Me, Me-30) with four sets of
undistinguishable trivalent methine protons ( 5.08, m, 3H, H-13,
H-17, and H-21; 5.43, t, 1H, H-6) and four saturated methyl groups
1.08, s, 1 Me; d, 1.09, s, 1 Me, Me-23, and Me-24; 0.84, s, 1 Me,
Me-25; and 0.78, d, 1 Me, Me-26), as well as one methine proton
attached to the carbon bearing a hydroxyl group at 3.42. Notably,
d
d
d
d
(d
d
d
2. Materials and methods
d
all proton signals on the 1H NMR are almost identical to the
previous assignment of one bicyclic alcohol obtained from the
SnCl4-catalized nonenzymatic cyclization of oxidosqualene, except
for some blur or ignored signals that were not previously identified
[29e31]. Moreover, the spectra characteristic of one 6/6-fused
bicyclic intermediate, which was isolated from a prokaryotic
squalene-hopene cyclase mutant, was similarly observed in that of
novel product from S. cerevisiae ERG7C703 mutants herein. The 13C
NMR (150 MHz, CD2Cl2) spectrum revealed the presence of three
set of tertiary-quaternary substituted double bonds (dc ¼ 124.25,
131.21; 124.42,134.92, and 125.30,134.47 ppm) and one set of weak
tertiary-quaternary substituted double bonds at dc ¼ 120.03 and
142.63 ppm, respectively.
2.1. Site-saturated mutagenesis and non-saponifiable lipid
extraction of ERG7C703
Mutagenesis of Cys703 in the S. cerevisiae ERG7 wild-type gene was
performed using the QuikChange site-directed mutagenesis kit (Stra-
tagene, La Jolla, CA). The degenerate mutagenic primers for Cys703
were the following, with substitutions underlined and silent mutation
italicized: ERG7C703X1: 50-CAACCACTCTNNNGCAATCGAATACCCAAG-
30 and ERG7C703X2: 50-CTTGGGTATTCGATTGCNNNAGAGTGGTTG-30.
The PCR reaction contains 0.8 mM each of dNTP, 100 ng of
pRS314ERG7WT plasmid as template, 1X Pfu polymerase buffer, 10
mL
of each primer (0.2 mM), 2.5 U of Pfu DNA polymerase and ddH2O to
the final volume of 20 m
L. The reaction mixture was denatured at 95 ꢀC
The HSQC spectrum showed that one olefinic methine protons
for 2 min, and then run for 25 cycles of denaturizing at 95 ꢀC for 30 s
each, annealing at 52 ꢀC for 1 min, polymerization at 68 ꢀC for 10 min,
(d
5.43 ppm) is attached to the carbon at 120.03 ppm whereas other
three methine protons at 5.08 are correlated to three tertiary
carbons at dc ¼ 124.25, 124.42, and 125.30 ppm, further suggesting
bicyclic skeleton with ring-fused tertiary-quaternary
d
a
a
substituted double bond and three exocyclic olefinic double bonds
at its hydrocarbon side chains. The HMQC spectrum also showed
that the methylene protons at d 2.05 (5-H), d 1.96 (4-H), d 1.88 (3-H)
are attached to the carbons at dc 21.99, 26.62, 26.78, and 39.73,
implying the correlation of methylene groups at exocyclic hydro-
carbon side chain. For the signals within the fused ring the methine
proton at
methine at 2.1 ppm (H-10) is attached to carbon at 39.98 ppm, and
methine proton at 3.42 is attached to the carbon at 76.56 ppm (C-
d 1.65 (H-8) is attached at the carbon at 32.81 ppm (C-8),
d
3), revealed by HSQC spectrum.
Moreover, the critical connectivity of hydrocarbon side chains
and relative double bond positions were observed from the HMBC
spectrum, and were described as following. (1) The chemical shift
of
d 5.08 revealed the connectivity with methyl groups at dc 15.88
(Me-27), dc 16.01 (Me-28), dc 17.66 (Me-29), and dc 25.67 (Me-30),
as well as six methylene signals at dc 21.99, 26.62, 26.78, and
39.73 ppm (C-11, C-12, C-15, C-16, C-19, and C-20), respectively;
The carbon at 124.25 ppm (C-21) is synchronously coupled to two
methyl protons at
tivity, whereas carbon at 124.42 ppm (C-17) and 125.29 ppm (C-13)
only showed 3J coupling to one methyl protons of
1.58 (Me-27 or
Me-28), respectively. Moreover, the carbon at 124.25 ppm (C-21),
d
1.58 (Me-29) and 1.66 (Me-30) by 3J connec-
Fig. 2. The superimposed homology model of S. cerevisiae ERG7C703I (gray) and
ERG7WT (blue) complexed with protosteryl C-17 cation (green). (For interpretation of
the references to colour in this figure legend, the reader is referred to the web version
of this article.)
d

2378
C.-H. Chang et al. / Biochimie 94 (2012) 2376e2381
and 124.42 ppm (C-17) coupled with methylene protons at
and 1.96 (H-15, H-16, H-19, or H-20) while 125.30 ppm (C-13)
further displayed a J connectivity at
tions in the HMBC spectrum clearly illustrated the exocyclic
hydrocarbon side chain’s connectivity. (2) The tertiary carbon at
32.81 ppm (C-8) and quaternary carbon at 35.96 ppm (C-9)
d
2.05
consistent with the genetic selection results. The ERG7C703D and
d
ERG7C703N viable mutants produced 2 as the sole product. Both
3
d
1.88 (H-11). These observa-
ERG7C703I and ERG7C703H produced
a
new yet unidentified
compound with a C30H50O formula, in conjunction with 2, (13 H)-
isomalabarica-14(26),17E,21-trien-3 -ol (3), (13 H)-isomalabarica-
14E,17E,21-dien-3 -ol (4), (13 H)-isomalabarica-14Z,17E,21-dien-
-ol (5), protosta-13(17),24-dien-3 -ol (6), (13 H)-malabarica-
14E,17E,21-trien-3 -ol (8), and 17 -protosta-20(22),24-dien-3 -ol
a
b
a
b
a
exhibited their connectivity to two same methyl groups (
d
0.78, Me-
3b
b
a
26 and 0.84, Me-25) whereas another tertiary carbon of C-10
d
b
a
b
(39.98 ppm) only correlated to one of these two methyl groups,
further suggesting the connectivity from C-8 to C-10 which is
scaffold next to the exocyclic hydrocarbon side chain in the B-ring
(9), by comparison with authentic standards by 1H and 13C NMR as
well as GCeMS [15e22]. The new compound was isolated and
further characterized with NMR and demonstrated to be
structure. (3) The methyl groups at A-ring (
d
1.08, s, Me-23;
d
1.09,
(8R,9R,10R)-polypoda-5,13E,17E,21-tetraen-3b-ol (10) (Fig. 3).
Me-24) showed 2J connectivity to one quaternary carbon at
42.10 ppm (C-4), 3J connectivity to methine carbon bearing
a hydroxyl group at 76.56 ppm (C-3), and 3J connectivity to
quaternary carbon at 142.63 ppm, that confirmed the position of
final deprotonation for the formation of a ring-fused tertiary-
quaternary substituted double bond.
Finally, the presence of NOEs among Me-25/H-8, the absence of
NOEs between Me-25/H-10, and the absence of 1He1H COSY
correlation between Me-25/Me-26 confirmed the stereochemistry
of the CeB 6-6 bicyclic nucleus. These findings unambiguously
established the unknown compound to be (8R,9R,10R)-polypoda-
Interestingly, compound 10 showed the same relative stereo-
chemistry as observed in 10-deoxy-17-hydroxyiridal, an iridal-
skeleton triterpenoid [29e32]. The ERG7C703X (X
¼
G/V/T)
mutants produced 2 and 3,while the ERG7C703S mutant produced 2,
and protosta-16,24-dien-3
-ol (7). The ERG7C703I mutant
3
b
produced diverse product profile 2, 3, 5, 6, 8, 9, and 10 in the
relative ratio of 20:22:11:10:15:18:4. The ERG7C703H mutant
produced diverse product profiles, including 2, 3, 4, 5, 6, 9, and 10,
in the ratio of 47:12:3:3:3:31:1. The non-viable genetic selection
and no product formation results of many ERG7C703X (X ¼ A/L/F/Y/
W/E/P/R/K/Q/M) mutants indicated a possible functional role of
Cys703 in ERG7 catalysis. The catalytic competency of ERG7C703X
(X ¼ G/V/T/I/S/H) mutants to lanosterol production indicated that
Cys703 is not directly involved in catalysis. Further identification of
diverse product profile produced by the ERG7C703S/I/H mutants may
suggest an indirect functional role of Cys703, presumably interacts
with spatially proximal residues such as Phe699.
5,13E,17E,21-tetraen-3b-ol, a chaireboat (CeB) 6-6 bicyclic product
with transesyn stereochemistry and D5,17,21 double bonds. Inter-
estingly, the steric isomer with chairechair conformation from the
SnCl4-catalized nonenzymatic cyclization of oxidosqualene
exhibited the similar but slightly different spectrometric data with
our unknown compound in the NMR spectrum [29e31].
To investigate the interaction between Cys703 and Phe699 of
ERG7, the ERG7F699A/C703I, ERG7F699T/C703I, and ERG7F699M/C703I
double mutations were performed. Previous studies uncovered
Phe699 as an essential residue involved in catalysis. Product profile
analysis of the ERG7F699X mutants also showed that the F699A
mutation did not produce any product, whereas the F699T, and
F699M mutations produced a single and a diverse product profile,
respectively [21]. In the present study, genetic selection and
product profile characterization results showed that the ERG7F699A/
3. Results and discussion
The ERG7C703X site-saturated and ERG7F699/C703 double muta-
tions were generated, using the QuikChange site-directed muta-
genesis, and transformed into a yeast TKW14 strain for genetic
selection and product characterization, as previously described
[16]. Following genetic selection with ergosterol complementation
experiments, the non saponifable lipid (NSL) from each mutant
were extracted and applied to AgNO3-impregnated silica gel
column for product profiles characterization, using GCeMS and
NMR (1H, 13C NMR, DEPT, 1He1H COSY, HMQC, HSQC, HMBC, and
NOE)spectroscopic techniques.
C703I
double mutant caused yeast non-viability and no product
formation, as expected. Alternatively, the ERG7F699T/C703I double
mutant altered product specificity from a single compound 6
(>99.8%) to multiple products 2, 6, and protosta-20(22),24-dien-
Table 1 shows the genetic selection results and the product
profiles of the ERG7C703X site-saturated and ERG7F699/C703 double
mutants with molecular mass of 426 Da. Neither lanosterol nor
truncated intermediates could be detected from the non-viable
3b
-ol (11) in the relative ratio of 15:67:13. The ERG7F699M/C703I
double mutant completely abolished both the enzymatic activity
and product formation. In parallel, steady-state kinetic analysis also
revealed that the ERG7F699T/C703I double mutant exhibited a 12-fold
decrease in the Vmax/KM value but only with a slight change in the
binding affinity for the substrate. Decline of the Vmax/KM value
could be attributed to the mutual interaction between Cys703 and
ERG7C703X (X
¼
A/L/F/Y/W/E/P/R/K/Q/M) mutants which is
Table 1
The product profiles of S. cerevisiae ERG7C703X site-saturated and ERG7F699A,T,M/C703I
double mutants.
AA substitution
C703
C703I
C703H
C703S
C703V
2
3
4
3
5
6
7
6
8
9
10
11
100
20
47
89
97
96
91
100
100
22
12
5
3
4
11
3
10
3
15
18
31
4
1
C703T
C703G
C703D
C703N
9
F699A/C703I
F699T/C703I
F699M/C703I
No product
15
No product
67
13
Fig. 3. Bond connectivity and stereochemistry established by HSQC/HMBC (bold bond,
━) and NOE interaction (curved arrows) of (8R,9R,10R)-polypoda-5,13E,17E,21-tetraen-
3b-ol.
aNeither cell viability nor product was characterized for the rest of the ERG7C703X
mutants.

C.-H. Chang et al. / Biochimie 94 (2012) 2376e2381
2379
Phe699 residues that introduction of the Cys703 mutation at the
pre-existed Phe699 mutant further affected enzymatic conforma-
tion and the interaction between Phe699 and the cationic inter-
mediate. This in turn affected the reaction rate and shifted the
product profile from a single product to diverse products and from
diverse products to no product formation.
deprotonation at C-9 or C-8 to form 2. Alternatively, the CeCeC
substrate conformer is cyclized to a CeC 6-6-5 tricyclic C-14
cation IIb, that is followed by direct abstraction of a proton from C-
15 to yield 8 as the end product.
Iridal-type compounds are biogenetically interesting mono-, bi-,
or spiro-cyclic triterpenoids discovered first by Marner et al. [32].
They exhibit a variety of biological activities, including cytotoxicity,
PKC activation, RasGRP pathway modulation, and ichthyotoxic
activity [33e38]. The biosynthesis of iridal-type triterpenoids has
been hypothesized to derive from oxidosqualene cyclization via
a B-ring boat intermediate [29e32]. Recently, Xiong et al. isolated
a marneral synthase (MRN1) from Arabidopsis thaliana that cata-
lyzes the formation of an iridal skeleton triterpenoid and validated
the abovementioned hypothesis [39]. In the present study, we
identified the ERG7C703I and ERG7C703H mutants that can catalyze
the biosynthesis of truncated bicyclic rearranged intermediate
related to iridal triterpenoids, further supporting the evolutionary
relationships between marneral synthase and oxidosqualene
cyclases. Interestingly, MRN1 contains methionine in place of
cysteine at the corresponding position. Previously, we identified
the adjacent first-tier residue Phe699 and showed its functional
role in affecting substrate folding, C- and D-ring annulation, and
exocyclic side chain stereochemistry [17,21]. Consistent with the
observation is the deletion of a single residue in this region of
squalene-hopene cyclase led to incomplete cyclization with some
stereochemical inversion [40]. Thus, the substitution of Cys703
with Ile or His may affect the interaction between Phe699 and OS to
form rings C and D, resulting in the isolation of intermediate
arrested at bicyclic stage for subsequent 1,2-shifts of hydride or
methyl group to form the C-5 cation Ia and subsequent abstraction
of a proton from C-6 to form compound 10.
A likely mechanism for the ERG7C703X mutants that leads to
diverse product profiles is shown in Scheme 1. The unusual struc-
tural features of 10 can be rationalized via a series of methyl and
hydride shifts starting from a bicyclic C-8 carbocation intermediate
(I, lanosterol numbering). In the ERG7C703I mutant, OS is folded in
either
a chaireboatechair (CeBeC) or a chairechairechair
(CeCeC) conformation. Most of the CeBeC substrate conformers
are cyclized to a CeBeC 6-6-5 tricyclic Markonikov C-14 cation IIa,
that is followed either by direct abstraction of a proton from C-26 or
C-15 to yield 3, 4, or 5 as end product, or by C-ring expansion and
subsequent D-ring annulation to generate the protosteryl C-20
cation (III) with different stereochemical control at the C-17 posi-
tion (IVa and IVb). Alternatively, minor amount of the CeBeC
substrate conformer undergoes cationic cyclization to the C-8
cation I and then consecutively migrates a hydride from H-9
, a methyl-group from Me-10 / Me-9 , and another hydride
from H-5 / H-10 to generate a bicyclic C-5 cation Ia, which then
loses a proton from C-6 to form compound 10. Notably, the cation
IVa with 17 -exocyclic hydrocarbon side chain undergoes
deprotonation from H-22 to produce 9; whereas the cation IVb
with 17 -exocyclic hydrocarbon side chain proceeds with
a backbone rearrangement of H-17 to H-20 to generate the
b / H-
8
b
b
b
a
a
a
a
a
b
a
a
protosteryl C-17 cation V. Elimination of a proton at C-13 and C-16
yielded 6 and 7, respectively. Subsequent skeletal rearrangements
of two hydride shifts (H-13
methyl groups (Me-14 / Me-13
generate the lanosteryl C-8/C-9 cation (VI), which undergoes
a
/ H-17
a
and H-9
b
/ H-8
b) and two
b
b
and Me-8
a
/ Me-14
a
)
The homology model of ERG7 also provides some insight into
the relationships between enzyme structure and product
Scheme 1. The proposed cyclization/rearrangement pathways of oxidosqualene within S. cerevisiae ERG7C703I mutant.

2380
C.-H. Chang et al. / Biochimie 94 (2012) 2376e2381
[2] K.U. Wendt, G.E. Schulz, E.J. Corey, D.R. Liu, Enzyme mechanisms for polycyclic
triterpene formation, Angew. Chem. Int. Ed. 39 (2000) 2812e2833.
[3] R. Xu, G.C. Fazio, S.P.T. Matsuda, On the origins of triterpenoid skeletal
diversity, Phytochemistry 65 (2004) 261e291.
[4] I. Abe, Enzymatic synthesis of cyclic triterpenes, Nat. Prod. Rep. 24 (2007)
1311e1331.
[5] C.J. Buntel, J.H. Griffin, Nucleotide and deduced amino acid sequences of the
oxidosqualene cyclase from Candida albicans, J. Am. Chem. Soc. 114 (1992)
9711e9713.
[6] Z. Shi, C.J. Buntel, J.H. Griffin, Isolation and characterization of the gene
encoding 2,3-oxidosqualene-lanosterol cyclase from Saccharomyces cerevisiae,
Proc. Natl. Acad. Sci. U. S. A. 91 (1994) 7370e7374.
[7] T. Hoshino, T. Sato, Squalene-hopenecyclase: catalytic mechanism and
substrate recognition, Chem. Commun. (2002) 291e301.
[8] B.A. Hess, Concomitant C-ring expansion and D-ring formation in lanosterol
biosynthesis from squalene without violation of Markovnikov’s rule, J. Am.
Chem. Soc. 124 (2002) 10286e10287.
[9] E.J. Corey, S.C. Virgil, H. Cheng, C.H. Baker, S.P.T. Matsuda, V. Singh,
S. Sarshar, New insights regarding the cyclization pathway for sterol
biosynthesis from (S)-2,3-oxidosqualene, J. Am. Chem. Soc. 117 (1995)
11819e11820.
[10] E.J. Corey, H. Cheng, C.H. Baker, S.P.T. Matsuda, D. Li, X. Song, Studies on the
substrate binding segments and catalytic action of lanosterol synthase.
Affinity labeling with carbocations derived from mechanism-based analogs of
2,3-oxidosqualene and site-directed mutagenesis probes, J. Am. Chem. Soc.
119 (1997) 1289e1296.
[11] S. Lodeiro, Q. Xiong, W.K. Wilson, M.D. Kolesnikova, C.S. Onak, S.P.T. Matsuda,
An oxidosqualene cyclase makes numerous products by diverse mechanisms:
a challenge to prevailing concepts of triterpene biosynthesis, J. Am. Chem. Soc.
129 (2007) 11213e11222.
[12] C. Jäckel, P. Kast, D. Hilvert, Protein design by directed evolution, Annu. Rev.
Biophys. 37 (2008) 153e173.
[13] J.A. Gerlt, P.C. Babbitt, Enzyme (re)design: lessons from natural evolution and
computation, Curr. Opin. Chem. Biol. 13 (2009) 10e18.
[14] P.A. Romero, F.H. Arnold, Exploring protein fitness landscapes by directed
evolution, Nat. Rev. Mol. Cell Biol. 10 (2009) 866e876.
[15] T.K. Wu, Y.T. Liu, C.H. Chang, Histidine residue at position 234 of
oxidosqualene-lanosterol cyclase from Saccharomyces cerevisiae simulta-
neously influences cyclization, rearrangement, and deprotonation reactions,
ChemBioChem 6 (2005) 1177e1181.
[16] T.K. Wu, Y.T. Liu, C.H. Chang, M.T. Yu, H.J. Wang, Site-saturated mutagenesis of
histidine 234 of Saccharomyces cerevisiae oxidosqualene-lanosterol cyclase
demonstrates dual functions in cyclization and rearrangement reactions,
J. Am. Chem. Soc. 128 (2006) 6414e6419.
specificity. The homology model showed that the Phe699 is the
first-tier residue with a distance of w4.8 Å to the C-17 protosteryl
cation, whereas the Cys703 is a second-tier residue located prox-
imal to the first-tier Phe699 residue with a distance of approxi-
mately 3.4 Å [17,21]. The functional role of Phe699 has been
suggested to be involved both in restricting the CeBeC confor-
mation and/or side chain rotation as well as in stabilizing the
protosteryl C-17 cation, through interaction with Tyr99, His234,
and Tyr707. Substitution of Cys703 with steric Phe, Tyr, or Trp
caused a steric hindrance in the active site cavity. This in turn
affected the folding of the substrate and resulted in non-viable
genetic selection and no product formation. Alternatively, the
Cys703Ile and Cys703His mutations might partially disrupt the
stabilization of Phe699 to the substrate folding or protosteryl C-17
cation, which resulted in accumulation of CeCeC or CeBeC
conformer-induced tricyclic, stereochemically inverted tetracyclic,
and truncated rearranged bicyclic products. Finally, the simulta-
neous mutation of Phe699 and Cys703 to Thr and Ile, respectively,
resulted in an increase of distance between amino acid position 699
and C-17 protosteryl cation of w0.7 Å, changing from 4.8 Å to about
5.5 Å. Consistent with the results is the observation of the decline of
the reaction rate and the shift of product profile from a single
product to diverse products and from diverse products to no
product formation. However, exact contributions of the mutations
on the product profile and proportions remain unclear and await
for X-ray structure determination of the mutated proteins.
In summary, this is the first example that a single amino acid
mutation within ERG7 can generate a bicyclic rearranged inter-
mediate related to the precursor of iridal triterpenoids. Further-
more, genetic selection and product characterization of the Cys703
site-saturated and Cys703/Phe699 double mutations indicated that
the functional roles of the ERG7C703 residue is involved in interac-
tions with Phe699 or other adjacent residues to affect proper
enzymatic conformation and subsequent regulation of product
profile. Therefore, our results exemplify the power of protein
engineering to generate a diverse product profile, which could be
customized via subtle changes to the interaction between neigh-
boring amino acid residues surrounding the active site cavity.
Although the yield for the iridal-type truncated intermediate is still
poor, the ERG7C703I mutant serves as a template for further product
specificity improvement. Further studies should focus on the
elucidation and engineering of new residues that directs iridal-type
structures with improved product specificity.
[17] T.K. Wu, H.Y. Wen, C.H. Chang, Y.T. Liu, Protein plasticity: a single amino acid
substitution in the Saccharomyces cerevisiae oxidosqualene-lanosterol cyclase
generates protosta-13(17),24-dien-3
b-ol, a rearrangement product, Org. Lett.
10 (2008) 2529e2532.
[18] T.K. Wu, T.T. Wang, C.H. Chang, Y.T. Liu, Importance of Saccharomyces cer-
evisiae oxidosqualene-lanosterol cyclase tyrosine 707 residue for chair-boat
bicyclic ring formation and deprotonation reactions, Org. Lett. 10 (2008)
4959e4962.
[19] T.K. Wu, C.H. Chang, Y.T. Liu, T.T. Wang, Saccharomyces cerevisiae
oxidosqualene-lanosterol cyclase:
a chemistryebiology interdisciplinary
study of the protein’s structureefunction-mechanism relationships, Chem.
Rec. 8 (2008) 302e325.
[20] T.K. Wu, W.H. Li, C.H. Chang, H.Y. Wen, Y.T. Liu, Y.C. Chang, Tyrosine 99 of
Saccharomyces cerevisiae oxidosqualene-lanosterol cyclase influences tricyclic
terpenoid moiety formation with differential stereochemical control, Eur. J.
Org. Chem. (2009) 5731e5737.
Acknowledgments
[21] T.K. Wu, C.H. Chang, H.Y. Wen, Y.T. Liu, W.H. Li, T.T. Wang, W.S. Shie, Alter-
ation of the substrate’s prefolded conformation and cyclization stereochem-
istry of oxidosqualene-lanosterol cyclase of Saccharomyces cerevisiae by
substitution at phenylalanine 699, Org. Lett. 12 (2010) 500e503.
[22] T.K. Wu, Y.C. Chang, Y.T. Liu, C.H. Chang, H.Y. Wen, W.H. Li, W.S. Shie,
Mutation of isoleucine 705 of the oxidosqualene-lanosterol cyclase from
We are grateful to Dr. John H. Griffin and Prof. Tahsin J. Chow for
helpful advice. We thank the Ministry of Education, Aiming for Top
University Plan (MOE ATU Plan), and the National Chiao Tung
University for financial support. This work was also supported in
part by the National Science Council of the Republic of China under
Contract No. NSC-99-2113-M-009-008-MY3, and NSC-99-2113-M-
009-004-MY2. We also thank the National Center for High-
Performance Computing for running Gaussian jobs.
Saccharomyces cerevisiae affects lanosterol’s C/D-ring cyclization and 17
a/b-
exocyclic side chain stereochemistry, Org. Biomol. Chem.
9 (2011)
1092e1097.
[23] A. Duriatti, F. Schuber, Partial purification of 2,3-oxidosqualene-lanosterol
cyclase from hog-liver. Evidence for a functional thiol residue, Biochem.
Biophys. Res. Commun. 151 (1988) 1378e1385.
[24] G. Balliano, G. Grosa, P. Milla, F. Viola, L. Cattel, 3-Carboxy-4-nitrophenyl-
dithio-1,10,2-trisnorsqualene: a site-directed inactivator of yeast oxidosqua-
lene cyclase, Lipids 28 (1993) 903e906.
Appendix A. Supplementary material
[25] L. Carrano, M. Noe, G. Grosa, P. Milla, M. Denaro, K. Islam, Solubilization and
identification of essential functional groups of Candida albicans oxidosqualene
cyclase, J. Med. Vet. Mycol. 33 (1995) 53e58.
[26] T.K. Wu, C.Y. Huang, C.Y. Ko, C.H. Chang, Y.J. Chen, H.K. Liao, Purification,
tandem mass characterization, and inhibition studies of oxidosqualene-
lanosterol cyclase enzyme from bovine liver, Arch. Biochem. Biophys. 421
(2004) 42e53.
Supplementary data related to this article can be found online at
http://dx.doi.org/10.1016/j.biochi.2012.06.014.
References
[27] R. Thoma, T. Schulz-Gasch, B. D’Arcy, J. Benz, J. Aebi, H. Dehmlow, M. Hennig,
M. Stihle, A. Ruf, Insight into steroid scaffold formation from the structure of
human oxidosqualene cyclase, Nature 432 (2004) 118e122.
[1] I. Abe, M. Rohmer, G.D. Prestwich, Enzymatic cycilzation of squalene and
oxidosqualene to sterols and triterpenes, Chem. Rev. 93 (1993) 2189e2206.

C.-H. Chang et al. / Biochimie 94 (2012) 2376e2381
2381
[28] K.U. Wendt, Enzyme mechanisms for triterpene cyclization: new pieces of the
puzzle, Angew. Chem. Int. Ed. 44 (2005) 3966e3971.
phorbol ester receptors with modest selectivity for the RasGRP receptor
subfamily, J. Med. Chem. 44 (2001) 3872e3880.
[29] K.B. Sharpless, E.E. van Tamelen, Terpene terminal epoxides. skeletal rear-
rangement accompanying bicyclization of squalene 2,3-oxide, J. Am. Chem.
Soc. 91 (1969) 1848e1849.
[36] H. Ito, S. Onoue, Y. Miyake, T. Yoshida, Iridal-type triterpenoids with
ichthyotoxic activity from Belamcanda chinensis, J. Nat. Prod. 62 (1999)
89e93.
[30] F.J. Marner, I. Longerich, Isolation and structure determination of new iridals
from Iris sibirica and Iris versicolor, Liebigs Ann. Chem. (1992) 169e172.
[31] F.J. Marner, T. Kasel, Biomimetic synthesis of the iridal skeleton, J. Nat. Prod.
58 (1995) 319e323.
[32] F.J. Marner, W. Krick, B. Gellrich, L. Jaenicke, W. Winter, Irigermanal and iri-
dogermanal: two new triterpenoids from rhizomes of Iris germanica L, J. Org.
Chem. 47 (1982) 2531e2536.
[33] J.P. Bonfils, F. Pinguet, S. Culine, Y. Sauvaire, Cytotoxicity of iridals, triterpenoids
fromIris, onhumantumorcelllinesA2780andK562, PlantaMed. 67(2001)79e81.
[34] K. Takahashi, S. Suzuki, Y. Hano, T. Nomura, Protein kinase C activation by
iridal type triterpenoids, Biol. Pharm. Bull. 25 (2002) 432e436.
[35] L. Shao, N.E. Lewin, P.S. Lorenzo, Z. Hu, I.J. Enyedy, S.H. Garfield, J.C. Stone,
F.J. Marner, P.M. Blumberg, S. Wang, Iridals are a novel class of ligands for
[37] A. Corbu, G. Gauron, J.M. Castro, M. Dakir, S. Arseniyadis, A domino-based
approach toward stereodefined heavily functionalized cyclohexanes:
synthesis of iridal’s core structure, Org. Lett. 9 (2007) 4745e4748.
[38] Z.J. Song, X.M. Xu, W.L. Deng, S.L. Peng, L.S. Ding, H.H. Xu, A new dimeric iridal
triterpenoid from Belamcanda chinensis with significant molluscicide activity,
Org. Lett. 13 (2011) 462e465.
[39] Q. Xiong, W.K. Wilson, S.P.T. Matsuda, An Arabidopsis oxidosqualene cyclase
catalyzes iridal skeleton formation by Grob fragmentation, Angew. Chem. Int.
Ed. Engl. 45 (2006) 1285e1288.
[40] T. Hoshino, K. Shimizu, T. Sato, Deletion of the Gly600 residue of Alicycloba-
cillus acidocaldarius squalene cyclase alters the substrate specificity into that
of the eukaryotic-type cyclase specific to (3S)-2,3-oxidosqualene, Angew.
Chem. Int. Ed. 43 (2004) 6700e6703.