β-Amyrin Biosynthesis: The Methyl-30 Group of (3S)-2,3-Oxidosqualene Is More Critical to Its Correct Folding to Generate the Pentacyclic Scaffold than the Methyl-24 Group

Article abstract of DOI:10.1002/chem.201502389

Oxidosqualene cyclases catalyze the transformation of oxidosqualene (1) into numerous cyclic triterpenes. Enzymatic reactions of 24-noroxidosqualene (8) and 30-noroxidosqualene (9) with Euphorbia tirucalli β-amyrin synthase were conducted to examine the role of the branched methyl groups of compound 1 in the β-amyrin biosynthesis. Substrate 8 almost exclusively afforded 30-nor-β-amyrin (>95.5%), which was produced through a normal cyclization pathway, along with minor products (<4.5%). However, a lack of the Me-30 group (analogue 9) resulted in significantly high production of premature cyclization products, including 6/6/6/5-fused tetracyclic and 6/6/6/6/5-fused pentacyclic skeletons (64.6%). In addition, the fully cyclized product (35.4%) having the 6/6/6/6/6-fused pentacycle was produced; however, the normally cyclized product, 29-nor-β-amyrin was present in only 18.6% of these products. The conversion yield of substrate 8 possessing a Z-Me group at the terminus was approximately twofold greater than that of compound 9 with an E-Me group. Thus, the Me-30 group is essential for the correct folding of a chair-chair-chair-boat-boat conformation of compound 1 for the production of the β-amyrin scaffold, whereas the Me-24 group exerts little influence on the normal polycyclization cascade. Here, we show that the Me-30 group plays critical roles in constructing the ordered architecture of a chair-chair-chair-boat-boat structure, in facilitating the ring-expansion reactions, and in performing the final deprotonation reaction at the correct position.

Full text of DOI:10.1002/chem.201502389

DOI: 10.1002/chem.201502389
Full Paper
&
Biosyntheses |Hot Paper|
b-Amyrin Biosynthesis: The Methyl-30 Group of (3S)-2,3-
Oxidosqualene Is More Critical to Its Correct Folding To Generate
the Pentacyclic Scaffold than the Methyl-24 Group
[a, b]
[a]
[a]
[a]
Tsutomu Hoshino,*
Ikki Kaneko
Yuri Miyahara, Masaki Hanaoka, Kazunari Takahashi, and
[a]
Abstract: Oxidosqualene cyclases catalyze the transforma-
tion of oxidosqualene (1) into numerous cyclic triterpenes.
Enzymatic reactions of 24-noroxidosqualene (8) and 30-nor-
oxidosqualene (9) with Euphorbia tirucalli b-amyrin synthase
were conducted to examine the role of the branched methyl
groups of compound 1 in the b-amyrin biosynthesis. Sub-
duced; however, the normally cyclized product, 29-nor-b-
amyrin was present in only 18.6% of these products. The
conversion yield of substrate 8 possessing a Z-Me group at
the terminus was approximately twofold greater than that
of compound 9 with an E-Me group. Thus, the Me-30 group
is essential for the correct folding of a chair–chair–chair–
boat–boat conformation of compound 1 for the production
of the b-amyrin scaffold, whereas the Me-24 group exerts
little influence on the normal polycyclization cascade. Here,
we show that the Me-30 group plays critical roles in con-
structing the ordered architecture of a chair–chair–chair–
boat–boat structure, in facilitating the ring-expansion reac-
tions, and in performing the final deprotonation reaction at
the correct position.
strate
8
almost exclusively afforded 30-nor-b-amyrin
(
>95.5%), which was produced through a normal cyclization
pathway, along with minor products (<4.5%). However,
a lack of the Me-30 group (analogue 9) resulted in signifi-
cantly high production of premature cyclization products, in-
cluding 6/6/6/5-fused tetracyclic and 6/6/6/6/5-fused penta-
cyclic skeletons (64.6%). In addition, the fully cyclized prod-
uct (35.4%) having the 6/6/6/6/6-fused pentacycle was pro-
[
4]
[5]
Introduction
well characterized and its X-ray crystal structure was solved.
However, few OSCs other than the human lanosterol synthase,
have been characterized. Very recently, we succeeded in the
complete purification of b-amyrin synthase from Euphorbia tir-
The polycyclization cascades of squalene and (3S)-2,3-oxido-
squalene (1) have been attractive to organic chemists and bio-
chemists for over 70 years since Ruzicka and coworkers pro-
[
6]
ucalli (EtAS) and in characterizing its enzymatic properties.
Oxidosqualene (1) is folded in a chair–chair–chair–boat–boat
[1]
posed the “biogenetic isoprene rule.” The reactions proceed
with complete regiospecificity and stereospecificity to yield
[
1,7]
conformation by b-amyrin synthase (Scheme 1).
Proton
[2]
sterols and triterpenes with remarkable structural diversity:
more than hundred different carbon frameworks that exhibit
attack on the epoxide ring initiates the polycyclization reaction
to yield the dammarenyl cation 3 with the 6/6/6/5-fused A/B/
C/D-tetracyclic ring system, which undergoes ring expansion
to create the 6/6/6/6-fused tetracyclic baccharenyl cation 4.
Further cyclization occurs to give the 6/6/6/6/5-fused pentacy-
clic lupanyl cation 5. A second ring enlargement of the five-
membered E-ring generates the 6/6/6/6/6-fused pentacyclic
oleanyl cation 6. 1,2-Hydride shifts (H-18a to C-19 cation and
H-13b to C-18) followed by deprotonation of the H-12a intro-
duce a double bond between the C-12 and C-13 atoms to fi-
nally generate the b-amyrin skeleton (Scheme 1). Notably, in-
termediates 4 and 6 are secondary cations, which are less
stable than the tertiary cations 3 and 5, indicating that the life-
times of the cations 4 and 6 are relatively shorter than those
of the intermediates 3 and 5. Functional analyses of the active
site residues of the lanosterol synthase from Saccharomyces
cerevisiae and the squalene-hopene cyclase (SHC) from Alicyclo-
bacillus acidocaldarius have been well documented by the
[3]
important biological activities can be produced. b-Amyrin (2)
consists of a pentacyclic scaffold with eight chiral centers and
is widely distributed among plants. The characterization of oxi-
dosqualene cyclases (OSCs) is incomplete because OSCs are
membrane-bound enzymes and are extremely difficult to
handle in vitro. In 2004, the human lanosterol synthase was
[
a] Prof. Dr. T. Hoshino, Y. Miyahara, M. Hanaoka, K. Takahashi, I. Kaneko
Graduate School of Science and Technology, Niigata University
Ikarashi 2-8050, Nishi-ku, Niigata 950-2181 (Japan)
Fax: (+81)25-262-6854
E-mail: hoshitsu@agr.niigata-u.ac.jp
[
b] Prof. Dr. T. Hoshino
Department of Applied Biological Chemistry
Faculty of Agriculture, Niigata University
Ikarashi 2-8050, Nishi-ku, Niigata 950-2181 (Japan)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201502389.
[
8]
[9]
group of Wu and by our group, respectively. Through muta-
Chem. Eur. J. 2015, 21, 15769 – 15784
15769
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
pound 1 influence the polycyclization cascade remains ambig-
uous. Recently, we reported the effect of the steric bulk at the
[
15e]
C-19 position of compound 1 on the cyclization cascade.
The replacement of the methyl group at the C-19 atom by an
ethyl group or a hydrogen atom gave the following three new
compounds: (17b-H, 20S)-20-ethyl-dammara-12,24-diene, the
b-amyrin homologue with an ethyl group at the C-17 atom,
and the novel tetracyclic scaffold with a 6/6/6/6-fused A/B/C/
D-ring system. The enzymatic reaction of 24,30-bisnoroxido-
squalene (7) (Figure 1) by the P. sativum b-amyrin synthase has
[
15c]
been reported;
however, little is known of the reactions of
the nor-analogues 8 and 9 which lack the methyl-24 and
methyl-30 groups, respectively. Here, we report the enzymatic
products of compounds 8 and 9 by using EtAS. Analogue 8 af-
forded nearly exclusively the b-amyrin homologue (i.e., 30-nor-
b-amyrin, 95.5%), whereas analogue 9 produced tetracyclic
cores (ꢀ50.5%) together with a small amount of the b-amyrin
homologue (i.e., 29-nor-b-amyrin, 18.6%). This finding demon-
strates that the Me-30 group of compound 1 is critical to its
accurate folding to generate the 6/6/6/6/6-fused pentacyclic
scaffold (i.e., the b-amyrin core) resulting from the chair–chair–
chair–boat–boat conformation and further indicates that the
binding of compound 9 with the EtAS was decreased due to
the lack of the Me-30 group, which significantly suppressed
the formation of the normal folding conformation; thus, the
polycyclization reaction terminated at the premature tetracy-
clic intermediate. The stereochemical fates of the Me-24 and
Me-30 groups of compound 1 during the b-amyrin biosynthe-
sis have been investigated by administering stable isotopically
labeled mevalonate and/or sodium acetate into the cell sus-
Scheme 1. Cyclization pathway from (3S)-2,3-oxidosqaulene (1) to b-amyrin
2). The Me-24 and Me-30 groups of compound 1 are regiospecifically con-
verted into the Me-30 and Me-29 groups of compound 2, respectively. The
filled circle shows the destiny of the C-23 atom of compound 1 during the
formation of compound 2; the C-23 atom of compound 1 is transformed
into C-20 atom of compound 2.
(
genesis experiments, we recently clarified that the cationic in-
termediates 4 and 5 are stabilized by the Phe728 residue
[10]
through cation–p interactions. In contrast, the p electrons of
the Phe464 residue are not important for the catalysis; instead,
the steric bulk of the aromatic ring is more critical for confer-
[11]
ring the normal folding conformation. However, knowledge
of the active-site residues of the b-amyrin synthase is still limit-
[6,10–12]
ed.
In addition to the mutagenesis experiments, the en-
zymatic cyclization products of substrate analogues have pro-
vided deep insight into how the polycyclization reaction is af-
fected by specific modifications, for example, alteration of the
folding conformation leading to a different stereochemistry,
a truncation of the ring-forming cascade, or a different cation-
quenching mode. A number of investigations regarding the
enzymatic reactions of the substrate analogues with lanosterol
[
7f]
pension culture of Rabdosia japonica and by culturing Sac-
charomyces cerevisiae encoding Panax ginseng b-amyrin syn-
[
12]
thase; the Me-24 and Me-30 groups of compound 1 were
converted into Me-30 and Me-29 groups of compound 2, re-
[
7f,12]
spectively.
These results were obtained from in vivo experi-
ments. In the case of the Arabidopsis thaliana lupeol synthase,
[13]
[14]
synthase from hog liver
and SHC
have been reported;
the final proton elimination has been reported to occur at
[
12]
however, studies on the substrate analogues with b-amyrin
synthase have been limited. The polycyclization reactions of
a few substrate analogues by Pisum sativum (P. sativum) b-
both Me-24 and Me-30 groups of compound 1; that is, the
deprotonation of compound 5 does not occur in a regiospecific
fashion. This report has prompted us to review the stereo-
chemical destiny of the Me-24 and Me-30 groups of com-
[
15a–d]
amyrin synthase have been reported,
for example, 22,23-
[15b]
dihydro-2,3-oxidosqualene,
24,30-bisnoroxidosqualene (7)
and the truncated C25 analogue.
pound 1 to construct the b-amyrin skeleton 2 by using the
[
15a,c]
[15d]
2
(
Figure 1),
However,
EtAS. In vitro incubation of the [ H ]-labeled compound 10
3
how the branched methyl groups on the backbone of com-
with purified EtAS indicated that the deuterium atoms were re-
giospecifically incorporated into the Me-30 group of com-
pound 2. Furthermore, we examined the relative enzymatic ac-
tivities with substrates 7–9 compared to compound 1 to evalu-
ate the role of the methyl group(s) at the terminus on the
polycyclization cascade. The relative activities were compared
by estimating the yields of the enzymatic products, and the in-
cubations were performed at the same concentrations of com-
pounds 7–9 as that of compound 1 by using a fixed concen-
tration of the EtAS. Analogue 8 afforded the largest amounts
of products among the analogues 7–9. This finding further in-
dicates that the Me-30 group of compound 1 has a more criti-
cal role than the Me-24 group in the robust association of
compound 1 with the EtAS.
Figure 1. Chemical structure of 2,3-oxidosqaulene (1), its analogue 7–10,
and b-amyrin (2).
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15770
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Aldehyde 11 was subjected to
a Horner–Wadsworth–
Results and Discussion
Emmons reaction with (EtO) POCH CO Et to give the ester 14.
2
2
2
The olefinic protons at d =6.95 ppm (1H) and 5.81 (1H)
H
Syntheses of 24,30-bisnoroxidosqualene (7), 24-noroxido-
(
400 MHz, CDCl ) showed the following coupling patterns: dd,
3
2
squalene (8), 30-noroxidosqualene (9), and [24,24,24- H ]-
3
J=15.6, 7.0 Hz and d, J=15.6 Hz, respectively. The large cou-
pling constant J=15.6 Hz clearly indicated an E geometry for
2,3-oxidosqualene (10) and the incorporation pattern of the
deuterium atoms into b-amyrin (2)
1
the terminal double bond. The GC-MS profile and H NMR
Scheme 2 shows the methods used to synthesize compounds
spectrum showed no contamination of the Z geometry, sug-
gesting that the purity of the E isomer was greater than 98%.
7
(
–10. The starting material 1,1’,2-trisnorsqualene-3-aldehyde
11), which was prepared by treating 2,3-oxidosqualene with
H IO , was subjected to a Wittig condensation reaction with
Ester 14 was reduced with iBu AlH to afford alcohol 15 in 77%
2
yield. Alcohol 15 was then subjected to reaction with methane
sulfonyl chloride to afford compound 16, which was used for
the subsequent reaction without further purification. To com-
5
6
+
À
[
Ph PCH ] I to give compound 12. The bisnorsqualene 12 was
3 3
converted into the epoxide 7 by treating it with N-bromosucci-
nimide (NBS) and K CO /MeOH. For the synthesis of compound
pound 16 thus prepared, LiAlH was added, and the product
2
3
4
8
, which lacks the 23-E-Me group of squalene, a Wittig reaction
was purified on a SiO column to yield the 30-norsqualene 17
2
+
À
of compound 11 with [Ph PCH CH ] Br was conducted by em-
(87%). The epoxide 9 was prepared with NBS and K CO . The
3
2
3
2
3
ploying NaN(SiMe3)2 as the base, yielding compound 13
33.8%). The GC-MS profile showed that the peak correspond-
deuterium-labeled compound 10 was synthesized by a synthet-
ic method similar to that used for compound 9. A reaction of
(EtO) POCH(CH )CO Et with aldehyde 11 was conducted in the
(
ing to the 30-norsqaulene 17 was negligible (i.e., the Z isomer
of compound 13>98%). Compound 13 was converted into
2
3
2
presence of NaH, and purification of the major product by SiO2
column chromatography gave the pure compound 18 (80%),
the GC-MS profile of the isolated compound 18 indicated no
contamination of the Z isomer. The ester 18 was converted
into the epoxide by using NBS and K CO . Next, the epoxide
1
the desired epoxide 8 by using NBS and K CO . In the H NMR
2
3
spectrum of compound 8 (400 MHz, CDCl ), the olefinic pro-
3
tons H-22 and H-23 were observed at d =5.39 ppm (m, 2H),
H
and the other olefinic protons were observed at d=5.14 ppm
2
3
(
m, 4H). In addition, the epoxide ring proton was observed at
was subjected to reduction with LiAlD to yield compound 19,
4
dH =2.70 ppm (t, J=6.4 Hz).
which was followed by mesylation and LiAlD4 reduction to
afford the desired product 10. To confirm the deuterium-la-
1
13
beled position, the H and C NMR spectra of compound 10
were compared with those of the authentic 2,3-oxidosqualene
(
1), the signals at d =1.68 ppm (s, 3H) and d =25.68 ppm (t)
H C
corresponding to the 24-position in compound 1 were missing
in the spectra of compound 10. The deuterium content was
+
estimated to be >96% by comparing the intensity of the [M ]
peak (m/z=429) for compound 10 with that (m/z=426) for
compound 1 in the EI MS spectra.
Details of the syntheses and the NMR and MS data of the
synthetic intermediates are described in the Supporting Infor-
mation (see Figure S1 and pp. S5–S9).
Incubation of compound 10 with the purified EtAS was con-
ducted under the following conditions to obtain the deuteri-
um-labeled compound 2: the racemic compound 10 (4 mg),
partially purified EtAS (1.2 mg), dithiothreitol (DTT) (1 mm), Tri-
ton X-100 (0.1%), bovine serum albumin (BSA, 20 mg), and po-
tassium phosphate buffer (pH 7.4, 0.1m) in a total volume of
2
0 mL. From the incubation mixture, the deuterium-labeled
compound 2 was isolated in a yield of 1.3 mg. The EI MS spec-
+
tra showed m/z=429 ([M ]) for the labeled compound 2 and
+
m/z=426 ([M ]) for compound 1; the deuterium content was
estimated to be greater than 94% from the EI MS spectra.
1
Figure 2 shows a comparison of the H (Figure 2A) and
13
C NMR spectra (Figure 2B) between the unlabeled compound
2
(a) and the deuterium-labeled compound 2 (b). The protons
Scheme 2. Synthetic schemes of the analogues 7–10. Reagents and condi-
and carbon atoms were completely assigned through 1D and
+
À
tions: i) [Ph
MeOH, RT; iv) [Ph
v) (EtO) POCH CO
SO Cl, Et N, CH Cl
x) LiAlD
3
PCH
3
] I /BuLi in THF, À788C; ii) NBS in THF/H
2
O, RT; iii) K
O, À788C;
2
CO
3
/
1
1
2
D NMR experiments, including H– H COSY, HOHAHA, NOESY,
+
À
3
PCH
Et, NaH, À408C; vi) LiAlH
, 08C; viii) LiAlH
2
CH
3
] Br /NaN(SiMe
3
)
2
, Et
2
HSQC, and HMBC spectra, in conjunction with the literature
2
2
2
4
or iBu
POCH(CH
2
AlH/THF, À408C; vii) Me-
3 2
[
16]
NMR data. The proton signal of the Me-30 group was miss-
ing, and no decrease in the proton signal of the Me-29 group
2
3
2
2
4
; ix) (EtO)
2
)CO
Et, NaH, À408C;
4
.
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15771
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
13
6 6 6 6
Figure 2. A) H NMR (400 MHz, C D ) and B) C NMR spectra (100 MHz, C D )
of unlabeled b-amyrin (a, bottom spectra) and the deuterium-incorporated
b-amyrin (b, upper spectra), which was obtained by the incubation of the
1
deuterium-labeled oxidosqualene 10. In the H NMR spectrum A-b, the
13
signal of the Me-30 group was missing. In the C NMR spectrum B-b, the
signal of the C-30 atom was missing (the signal was significantly decreased)
and the signal of the C-20 atom was shifted to higher field (d from 31.25 to
0.03, Dd=1.22 ppm). This shift is due to the b-isotope effect, because the
C
3
signal of the carbon atom situated at the b position from the deuterium is
generally shifted up-field. These NMR data unambiguously demonstrated
that the deuterium atom is positioned on the C-30 atom. This result clearly
demonstrated that the cyclization of compound 1 into compound 2 by the
EtAS proceeded through the same pathway as that shown in Scheme 1.
Figure 3. GC traces of the hexane extracts obtained by incubating A) 2,3-oxi-
dosqualene (1), B) 24,30-bisnoroxidosqualene (7), C) 24-noroxidosqualene
1
was observed in the H NMR spectrum of the deuterium-la-
(
8), or D) 30-noroxidosqualene (9) with the native EtAS b-amyrin synthase.
1
3
beled compound 2 (Figure 2A, b). In the C NMR spectrum of
the labeled compound 2 (Figure 2B, b), the signal intensity of
the C-30 atom was significantly decreased due to the splitting
by the deuterium atom, but the intensity of the C-29 signal
Incubation conditions: the (3R,S)-substrates 7–9 (200 mg) were incubated
with the pure EtAS (1 mg) at 308C for 7 h in a total volume of 2.5 mL. After
the enzymatic reaction was terminated by the addition of 15% KOH/MeOH
(
3 mL) and heating of the mixture at 808C for 3 min, an aliquot of the squa-
À1
lene solution (50 mL, 1.0 mgmL dissolved in 0.1% Triton X-100) was added
to the reaction mixture as the internal standard. The lipophilic materials
were extracted with hexane (3”3 mL) and the resultant solution was evapo-
(
dC =33.56 ppm) of the labeled compound 2 remained un-
changed. Furthermore, the dC signal of the C-20 atom was
shifted to a higher field (Dd=1.22 ppm) because of the b-iso-
tope effect. These NMR data unambiguously demonstrated
that the deuterium atoms were positioned only at the C-30
atom of compound 2 and that no labeling of the C-29 atom
was observed, indicating that the E-Me (Me-24) and Z-Me (Me-
rated to a small amount of volume, which was subjected to a short SiO
2
column (hexane/EtOAc 100:20) to remove the Triton X-100 included in the
incubation mixture. The eluents were evaporated to dryness. To the residues
hexane (2.0 mL) was added and an aliquot of the solution (1.0 mL) was sub-
sequently injected to the GC apparatus to quantify the lipophilic materials.
GC conditions: DB-1 capillary column (30 m”0.25 mm, J&W Scientific Inc.),
À1
injection temperature: 3008C; column temperature: 245–2908C (28Cmin ),
3
0) groups of compound 1 can be regiospecifically converted
À1
flow rate (He gas): 1.50 mLmin . The several peaks of compounds 1, 7, 8,
into the Me-30 and Me-29 groups of compound 2, respectively,
as illustrated in Scheme 1 and Figure 2. Therefore, the EtAS
strictly controls the Me-24 and Me-30 groups of compound
and 9 indicate that the thermal degradation of these substrates occurred
during GC. Products 20–33 were numbered in order of their elution from
the GC analyses. The structures of these products and the distribution ratio
are shown in Figure 4. Isolation of the peaks A–E was unsuccessful because
of a tiny production.
1
during the biosynthesis of the pentacyclic scaffold 2, as does
[12]
the Panax ginseng b-amyrin synthase.
greater than 96.6%, with small amounts of the tetracyclic
Enzymatic products from 24,30-bisnor-2,3-oxidosqualene (7)
products (3.4%) butyrospermol and tirucalla-7(8)-24-dien-3b-ol
[
6]
Figure 3 depicts the GC chromatogram of hexane extracts ob-
tained by incubating four different substrates—2,3-oxidosqua-
lene (1) (Figure 3A), 24,30-bisnoroxidosqualene (7) (Figure 3B),
also produced. Abe et al. investigated the in vivo enzymatic
reaction of compound 7 with the P. sativum b-amyrin syntha-
[
15c]
se.
We re-examined the enzymatic reaction of compound 7
2
4-noroxidosqaulene (8) (Figure 3C), and 30-noroxidosqualene
9) (Figure 3D)—with the recombinant EtAS (i.e., the wild-type
EtAS). Figure 3A shows that this b-amyrin synthase is classified
by using a different b-amyrin synthase, that is, EtAS, in vitro
(
(Figure 3B). Five enzymatic products were observed in the GC
profile, whereas three products were reported as the enzymat-
[17]
[15b]
as a monofunctional OSC;
the production of b-amyrin is
ic products of the P. sativum b-amyrin synthase.
Hexane
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15772
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
was added to the incubation
mixture of compound 7 (23 mg)
with the EtAS, and the lipophilic
materials, including the enzy-
matic products, were extracted
with hexane. A mixture of the
enzymatic products (3 mg) was
isolated by using SiO2 column
chromatography (hexane/EtOAc
(
Ac=Acyl) 100:5), and these
products were then acetylated
with Ac O/Py (Py=pyridine). The
2
acetate of the main product 21
was isolated in a yield of 1.8 mg
through normal-phase HPLC (GL
Science Inertsil SIL-100A column,
2
2
1
50”7.6 mm ;
hexane/THF
00:0.2), and its structure was
determined to be 29,30-bisnor-
b-amyrin by detailed NMR analy-
ses, which included 2D NMR (see
the Supporting Information Fig-
ures S3-1 to S3-7 and Figure S3- Figure 4. Products of the enzymatic conversion of compounds 7, 8, and 9 by using the EtAS (i.e., compounds 20–
3). The product numbering is shown in Figure 3. The values given in parentheses refer to the amounts produced
[%]), which were estimated from three independent experiments. The estimated figures were collected by using
squalene as the internal standard, as described in the legend of Figure 3. Product names: 29,30-bisnor-d-amyrin
20), 29,30-bisnor-b-amyrin (21), 29,30-bisnor-germanicol (22), (19S)-30-norlup-12-en-3b-ol (23), 30-nor-b-amyrin
inferred from in vivo experi- (24), 27-nordammara-20(21),24-dien-3b-ol (25), 27-norbutyrospermol (27-noreuph-7(8)-24-diene-3b-ol) (26), (19R)-
0-norlup-12-en-3b-ol (27), 27-nortirucalla-7(8)-24-diene-3b-ol (28), 30-norbacchara-12,21-diene-3b-ol (29), 29-nor-
3
(
9
). The structure of compound
1 is illustrated in Figure 4. This
result was consistent with that
2
(
3
ments by using the P. sativum b-
b-amyrin (30), 29-norgermanicol (31), 29-nortaraxasterol (29-norurs-20(30)-en-3b-ol) (32), 27-nordammara-3b,20-
diol (33). The numbering of the carbon atoms of the chemical structures is used to assign the NMR data. Thus,
these numbering is somewhat different from those based on the nomenclatures of natural cyclic triterpene scaf-
[
15c]
amyrin synthase.
The mass
spectra of the trimethylsilyl
(
TMS) derivatives of compounds folds. It is likely that products A and B have pentacyclic scaffolds, based on the analyses of the MS data (see main
text and the comments in Figure S2 in the Supporting Information). The careful analyses of the GC-MS (fragmen-
2
0 and 22, which were prepared
tation and retention time) showed that products C and D were assigned to be compounds 30 and 31, respective-
ly. Isolation of product E and its structural determination failed.
with bis(trimethyl)amine, were
compared with those reported
[
15c]
in the literature,
which led to
the identification of compound 20 as 29,30-bisnor-d-amyrin
and compound 22 as 29,30-bisnorgermanicol. The identifica-
tion of other products A and B were unsuccessful because of
their small quantities; however, the GC-MS analyses indicated
that compounds A and B may have pentacyclic skeletons (see
the comments in Figure S2, p. S15 in the Supporting Informa-
tion). GC analysis (Figure 3B) showed that the products 21, 20,
the recovered compound 8 afforded quantities enabling struc-
tural analysis by NMR spectroscopy. The yields of the isolated
products 23 and 24 were 0.4 and 4.0 mg, respectively. The
NMR spectra of the acetate of compound 23 (600 MHz, C D )
6
6
are shown in Figures S5-1 to S5-7 in the Supporting Informa-
tion. One double bond [d =124.1 (d, C-12), 140.1 ppm (s, C-
C
13)] was observed for the C-12ÀC-13 bond; the latter carbon
22, and (A&B) were generated in a ratio of 6.4:1:2.7:1.1, which
atom was correlated with the Me-27 group [d =1.28 ppm (s,
H
was somewhat different from the ratio estimated from the P.
3H)], and the H-11 [d =1.72 (m), 1.92 ppm (m)] had cross
H
[15c]
sativum b-amyrin synthase (21/20/22=12.6:1:4.3).
peaks for the C-12 and C-13 atoms in the HMBC spectrum. The
1
ethyl group was definitively observed in the H NMR spectrum
[
d =1.09 (t, J=6.8 Hz, 3H), 1.88 ppm(m, 2H)] and was further
H
Enzymatic products 23 and 24 from 24-nor-2,3-oxidosqua-
lene (8)
1
1
confirmed by a H– H COSY cross-peak correlation between
them. The presence of the Et group suggests that product 23
has a norlupane skeleton, which was established by the de-
tailed analyses of other NMR data, including HMBC, HSQC,
COSY, HOHAHA, and DEPT spectra. The configurations of the C-
18 and C-19 atoms were determined from the NOESY spec-
Figure 3C shows the GC profile for the enzymatic reaction of
compound 8. The major product 24 and small amounts of
three other products (23 and C&D) were observed (24/23/
C&D=63.7:1:2). The large-scale incubation of compound 8
(
15.5 mg) with the purified EtAS (6.2 mg) was performed to iso-
trum. The absence of NOEs of H-18 [d =1.72 ppm (m, 1H)]/
H
late the enzymatic products. After the unreacted compound 8
Me-27 [d =1.28 ppm (s, 3H)] and of H-18/H-19 [d =2.06 ppm
H
H
was removed by SiO2 column chromatography, the product
(m, 1H)] indicated H-18b and H-19a configurations. This ste-
fraction was acetylated with Ac O/Py. Repeated incubation of
reochemistry was further supported by the definitive NOEs of
2
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15773
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
H-18/Me-28 [d =1.18 ppm (s, 3H)] and of H-20 [d =1.88 ppm
quently checked by GC-MS. Isolation of the acetate products
of compounds 25–32, which were produced in relatively larger
amounts, were successful; other products, including product E,
could not be isolated because of their very low yields.
H
H
(
m, 2H)]/Me-28. Thus, product 23 was determined to be (19S)-
0-norlup-12-en-3b-ol, as depicted in Figure 4. Complete NMR
assignments are shown in Figures S5-8 (C D ) and S5-10
3
6
6
1
13
(
CDCl ) in the Supporting Information.
The H, C NMR, and DEPT spectra of compound 25-Ac
3
1
The ethyl signal observed in the H NMR spectrum of prod-
(400 MHz, C D , Figures S8-1 and S8-2 in the Supporting Infor-
6
6
uct 23 was not observed in the spectrum of the main product,
mation) indicated the presence of one methylidene residue:
that is, the acetate of compound 24 (600 MHz, CDCl , Fig-
d =5.03 (s, 1H; s) and 5.10 ppm (s, 1H) for H-21, d =
3
H
C
ure S6-1 in the Supporting Information), instead, one doublet
108.2 ppm (t) for C-21, and d =152.3 ppm (s) for C-20. In addi-
C
Me signal [d =0.859 ppm (d, J=6.5 Hz, 3H; Me-29)] was ob-
tion, another double bond was observed: d =5.62 ppm (m,
H
H
served. Furthermore, no olefinic Me proton of compound 8
2H) for H-24 and H-25, and d =131.5 ppm (d) for C-24, and
C
(
observed at d =1.8–1.5 ppm) was observed; all of the Me sig-
dC =125.2 ppm (d) for C-25. The definitive HMBC cross peaks
H
nals of compound 24 appeared at higher fields of d =0.9–
were observed for H-21/C-17 [d =48.27 ppm (d)] and for H-
H
C
0
.5 ppm. Thus, product 24 consisted of a pentacyclic core re-
21/C-22 [d =34.70 ppm (t)]. From a biogenetic viewpoint, the
C
sulting from the complete cyclization. One double bond was
presence of the two double bonds suggested that compound
25 was a tetracyclic product. Detailed analyses of the corre-
sponding 2D NMR spectra further supported the tetracyclic
skeleton, as shown in Figure S8-8 in the Supporting Informa-
observed [d =122.0 (d, C-12), 145.3 ppm (s, C-13)]. The Me-27
C
group [d =1.14 ppm (s, 3H)] exhibited a HMBC cross-peak
H
correlation with the C-13 atom, verifying that the double bond
is positioned at the C-12ÀC-13 bond. On the basis of other
NMR spectra (Figures S6-1 to S6-7 in the Supporting Informa-
tion), a pentacyclic oleanane skeleton was assigned for com-
pound 24. The configurations of the C-17, C-18, and C-20
atoms were determined by using NOESY spectra. No NOE be-
tion. Strong NOEs were observed between Me-29 [d =
H
0.982 ppm (s, 3H)]/H-17 [d =2.42 ppm (m, 1H)]; however, no
H
NOE was observed between H-13 [d =1.87 ppm (m, 1H)]/H-
H
17, verifying the configurations to be H-13b and H-17a. Thus,
product 25 was assigned as 27-nordammara-20(21),24-dien-3b-
ol (Figure 4 and Figure S8-8 in the Supporting Information).
tween Me-27/Me-28 [d =0.805 ppm (s, 3H)] was observed;
H
however, definitive NOEs between Me-28/H-18 [d =1.80 ppm
The NMR spectra of product 26-Ac (400 MHz, CDCl ) are
H
3
(
m, 1H)] and between H-18/H-20 [d =1.25 ppm (m, 1H)] were
shown in Figures S9-1 to S9-7 in the Supporting Information.
H
13
observed. These NOEs strongly indicate the following configu-
rations: Me-28b, H-18b, and H-20b. Detailed NMR analyses (Fig-
ure S6-8 in the Supporting Information) verified that the struc-
ture of compound 24 was that of 30-nor-b-amyrin, as shown in
Figure 4. Isolation of the products C and D was unsuccessful
because of the significantly low yields. However, we observed
that the EI MS fragmentation patterns and the retention times
of products C and D were the same as those of compounds 30
and 31, respectively (see Figure 3; compare Figures S4-3 and
S4-4 with Figures S7-6 and S7-7 in the Supporting Information,
respectively). Thus, products C and D were indicated to be
compounds 30 and 31, respectively. Products 30 and 31 were
successfully isolated from the incubation mixtures of substrate
The C NMR spectrum showed that two double bonds were
involved in compound 26: dC =117.6 (d) for C-7 and
145.9 ppm (s) for C-8, and d =131.9 (d) for C-24 and d =
C
C
124.4 ppm (d) for C-25. The chemical shifts of the latter pair
were the same as those observed for substrate 9; thus, com-
pound 26 contained one unreacted double bond and pos-
sessed a tetracyclic core. The Me-29 group [d =0.971 ppm (s,
H
3H)] had an HMBC cross peak for the C-8 atom. Furthermore,
the chemical shift of H-9 [d =2.22 ppm (m, 1H)] indicated
H
that H-9 was situated at the allylic position. Thus, another
double bond was located at the C-7ÀC-8 bond. In the NOESY
spectrum, a definitive NOE correlation was observed between
the Me-29 group and H-17 [d =1.48 ppm (m, 1H)] and be-
H
9
, as described below.
tween the Me-18 group [d =0.801 ppm (s, 3H)] and H-9,
H
strongly supporting the configurations H-17b and Me-18a. To
determine the stereochemistry of the C-20 atom, we compared
Enzymatic products 25–32 from 30-nor-2,3-oxidosqualene
9)
the chemical shift of the Me-21 group [d =0.829 ppm (d, J=
H
(
6
.4 Hz, 3H)] of compound 26-Ac with that of authentic euph-
Numerous products were produced by the incubation of com-
pound 9 with the EtAS (Figure 3D). The large-scale incubation
of compound 9 (50 mg) was conducted to isolate these prod-
ucts. Two spots (high- and low-polarity materials) were ob-
7,24-diene with (20R) stereochemistry [d (CDCl )=0.835 ppm
H
3
(d, J=6.8 Hz)] and that of tirucall-7,24-diene with (20S) config-
[
18]
uration [dH (CDCl )=0.865 ppm (d, J=6.7 Hz)].
The nearly
3
identical chemical shifts of the Me-21 group of compound 26-
Ac to that of euph-7-ene verified that the stereochemistry at
the C-20 atom was R; thus, the structure of compound 26 was
served on SiO TLC plates as the enzymatic products. The un-
2
reacted compound 9 was recovered by SiO column chroma-
2
tography and then subjected to repeated incubations. The
low- and high-polarity products were isolated in yields of 18
and 1.7 mg, respectively. The higher-polarity product 33 was
further purified by normal-phase HPLC because some impuri-
ties were contaminated. The less-polar products were acetylat-
27-norbutyrospermol
(27-noreuph-7(8),24-diene-3b-ol),
as
shown in Figure 4 and Figure S9-8 in the Supporting Informa-
tion. Notably, all the stereochemistry of compound 26 coin-
cides with those of product 28 except for the C-20 stereo-
chemistry. The NMR spectra of compound 28-Ac (600 MHz,
C D ) are depicted in Figures S11-1 to S11-7 in the Supporting
ed with Ac O/Py and subjected to repeated separation by
2
6
6
normal-phase HPLC, and the purity of each acetate was subse-
Information, and the important 2D NMR correlations for the
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15774 ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
proposed structure are shown in Figure S11-8 in the Support-
ing Information. The NMR spectra of samples dissolved in
racyclic product. The Me-27 group [d =1.20 ppm (s, 3H)] had
H
clear HMBC correlations with the C-13 and C-15 atoms [d =
C
CDCl (600 MHz) are also shown in Figures S11-9 and S11-10 in
26.79 ppm (t)]. The COSY and HOHAHA spectra indicated the
3
1
1
the Supporting Information, and the detailed correlations ob-
tained by 2D NMR spectroscopy are illustrated in Figure S11-11
in the Supporting Information. These data allowed us to unam-
biguously propose the tetracyclic skeleton for compound 28,
which corresponds to that of compound 26. However, the
chemical shift of the Me-21 group [d (CDCl )=0.862 ppm, (d,
presence of a H– H spin-coupling network from H-12 to H-9.
The Me-28 group [d =0.971 ppm (s, 3H)] had HMBC cross
H
peaks for carbon atoms d =45.03 (t, C-18), 33.69 (t, C-16),
C
45.75 (t, C-19), 34.39 ppm (s, C-17), these cross peaks suggest-
ed that the D-ring of compound 29 consisted of a six-mem-
bered ring but not a five-membered ring. The HSQC spectrum
H
3
J=6.5 Hz)] was different from that of compound 26 and equiv-
showed that the proton signals of H-18 appeared at d =1.88
H
alent to that of authentic tirucalla-7(8),24-dien [d =0.865 ppm
(m) and 2.31 ppm (brd, J=16 Hz), which strongly indicated
that H-18 is situated at an allylic position. On the basis of the
detailed analyses of the 2D NMR data (Figure S12-8 in the Sup-
porting Information), we propose that the structure of com-
pound 29 possesses a 6/6/6/6-fused tetracyclic baccharane
scaffold. In the NOESY spectrum, a clear cross peak between
H
[
18]
(
d, J=6.7 Hz)], indicating that (20S) stereochemistry was as-
signable for compound 28. Thus, the structure of compound
8 was that of 27-nortirucalla-7(8), 24-dien-3b-ol, as shown in
2
Figure 4 and in Figures S11-8 and S11-11 in the Supporting In-
formation.
The NMR spectra of product 27-Ac (400 MHz, C D ) are de-
Me-27/H-19 [d =1.45 ppm(m, 2H)] was observed, indicating
6
6
H
picted in Figures S10-1 to S10-7 in the Supporting Information.
that the 17-side chain is of a orientation. Therefore, the struc-
ture of product 29 was proposed as 30-norbacchara-12-en-3b-
ol, as depicted in Figure 4.
One double bond [d =126.1 (d, C-12), 139.6 ppm (s, C-13)]
C
was found at the C-12ÀC-13 bond; the latter carbon atom was
correlated with the Me-27 group [d =1.16 ppm (s, 3H)], and
The NMR spectra of product 30-Ac (600 MHz) dissolved in
C D and CDCl are shown in Figures S13-1 to S13-7 in the Sup-
H
the H-11 [d =1.92 ppm (m, 2H)] had cross peaks with the C-
H
6
6
3
1
2 and C-13 atoms in the HMBC spectrum. The observation of
porting Information and Figures S13-9 to S13-10 in the Sup-
HMBC correlations of H-18 [d =2.57 ppm (d, J=11.2 Hz)] with
porting Information, respectively. In the C D6 solution, one
H
6
both C-12 and C-13 atoms further supported that the double
double bond was observed in the molecule: d =5.36 ppm
H
1
1
bond is located at the C-12ÀC-13 bond. The H– H COSY spec-
(brt, J=3.4 Hz, 1H; H-12), d =122.2 (d, C-12), 145.2 ppm (s, C-
C
trum revealed the presence of an ethyl group [d =1.08 ppm
13). The double-bond position was determined to be located
H
(
t, J=7.0 Hz, 3H), 1.34 (m, 1H), 1.93 ppm (m, 1H)]. The COSY
at the C-12ÀC-13 bond because, in the HMBC spectrum, the
1
1
and HOHAHA spectra revealed the H– H network from H-18
to H-22, which is indicative of the five-membered E-ring, defini-
tively demonstrating that product 23 consists of a norlupane
skeleton. The configurations of the C-18 and C-19 atoms were
determined from the NOESY spectrum. Unambiguous NOEs
Me-27 group [d =1.31 ppm (s, 3H)] had a clear cross-peak
H
correlation with the C-13 atom, and the H-12 also had defini-
tive correlations with the C-18 [d =45.49 ppm (d)] and C-9
C
atoms [d =47.85 ppm (d)]. The absence of an additional
C
double bond suggested that compound 30 had a fully cyclized
between H-18/H-19 [d =2.08 ppm (m, 1H)] and between H-
pentacyclic skeleton. The COSY and HOHAHA spectra revealed
H
1
8/Me-28 [d =1.22 ppm (s, 3H)] indicated H-18b and H-19b
the presence of the proton network from H-18 [d =2.23 ppm
H
H
configurations. Thus, product 27 was determined to be
18R,19R)-30-norlup-12-en-3b-ol, as depicted in Figure 4. Com-
plete assignments of the NMR data are shown in Figures S10-8
C D ) and S10-10 (CDCl ) in the Supporting Information. Prod-
(m, 1H)] to H-20 [d =2.08 ppm (m,1H)], the latter of which
H
(
was correlated with the Me-29 group [d =0.960 ppm (d, J=
H
7.2 Hz, 3H)], indicating that the Me-29 group was positioned
(
at the C-20 atom [d =39.5 ppm (d)]. The Me-29 group had
6
6
3
C
ucts 23 and 27 possessed the same norlupane scaffold, which
was further confirmed by the observation of the similarity of
the EI MS data (compare Figure S4-1 with Figure S7-3 in the
Supporting Information), however, the configurations at the C-
HMBC cross peaks with the C-19 [d =39.08 ppm (t)] and C-21
C
atoms [d =27.44 or 27.47 ppm (t)], and the Me-28 group
C
[d =1.09 ppm (s, 3H)] also exhibited clear HMBC correlations
H
with the C-16 [d =27.44 or 27.47 ppm (t)] and C-22 atoms
C
1
9 atom were opposite to each other, that is, S configuration
[d =35.44 ppm (t)], further supporting the presence of a six-
C
for compound 23 and R stereochemistry for compound 27.
Furthermore, the configuration at the C18 atom of compound
membered E-ring. The stereochemistry at the C-20 atom was
determined to be S because a clear NOE cross peak was ob-
served between H-18 and the Me-29 group. Thus, we inferred
that the structure of compound 30 was that of 29-nor-b-
amyrin (Figure 4 and Figure S13-8 in the Supporting Informa-
tion).
23 was the same as that of compound 27, that is, R. The for-
mation mechanisms of compounds 23 and 27 are shown in
Scheme 5.
The NMR spectra (400 MHz, C D ) of compound 29-Ac
6
6
showed that product 29 possessed two double bonds: d =
All the olefinic methyl protons (d (CDCl )=1.64–1.57 ppm)
H
H
3
5
.39 (brs, 1H; H-12), 5.63 ppm (m, 2H; H-21, H-22), d =121.2
of substrate 9 were shifted to higher fields (d =1.10–
C
H
(
d, C-12), 139.2 (s, C-13), 124.4 (d, C-22), 132.6 ppm (d, C-21).
0.73 ppm) for product 31-Ac, suggesting that the olefinic
methyl group was missing. This finding indicated that com-
pound 9 was fully cyclized by the enzymatic reaction to yield
The chemical shifts and the splitting patterns of the H-21 and
1
H-22 atoms in the H NMR spectrum of compound 29-Ac were
identical to those of substrate 9, suggesting that the terminal
double bond remained unreacted; thus, compound 29 is a tet-
1
13
a pentacyclic core. The H and C NMR spectra (400 MHz, C D )
6 6
indicated the presence of one double bond: d =5.27 ppm
H
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15775
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(
brs, 1H; H-19), d =125.9 (d, C-19), 144.9 ppm (s, C-18). The
0.977 ppm (s, 3H)]. The complete analysis of NMR data (Fig-
ure S16-8 in the Supporting Information) indicated that prod-
uct 33 was 27-nordammar-24-en-3b,20-diol; however, the ste-
reochemistry at the C-20 position has yet to be determined.
The structures of all the enzymatic products from substrates
8 and 9 are listed in Figure 4. The production ratio ([%]) of the
enzymatic products and the conformations of compounds 8
and 9 folded by the b-amyrin synthase are also shown in
Figure 4.
C
double-bond position was determined to be located at the C-
8ÀC-19 bond on the basis of the HMBC correlations between
Me-28 [d =1.24 ppm (s, 3H)]/C-18 and Me-29 [d =1.175 ppm
1
H
H
(
d, J=6.8 Hz, 3H)]/C-19. The chemical shifts of the protons of
the Me-28 group were very similar to those of the Me-29
group; thus, the orientation of the Me-29 group was difficult
to infer, but the following NOEs were clearly observed: Me-27/
H-16a/H-21a/H-20a (see Figure S14-9 in the Supporting Infor-
mation). Thus, we inferred the b arrangement of the Me-29
group. The proton signals of the Me-29 [d =0.926 ppm (d, J=
H
The folding conformations of substrates 7–9 leading to the
enzymatic products 20–33
7
.0 Hz)] and Me-28 groups [d =1.03 ppm (s, 3H)] measured in
H
CDCl (600 MHz) were more separated. A clear NOE between
3
Me-28/Me-29 was observed (Figure S14-12 in the Supporting
Information), further ensuring the b orientation of the Me-29
group. The detailed analyses of the 2D NMR spectra measured
in C D and in CDCl are described in Figures S14-8 and S14-13
13
2
By using feeding experiments of the [23- C]- and [23,23- H ]-
2
[15c]
labeled compound 7, Abe et al.
reported the intriguing find-
ing that the polycyclization reaction of compound 7 proceed-
ed through a pathway that differed from that depicted in
6
6
3
in the Supporting Information, respectively. The structure of
13
Scheme 1. Scheme 3 shows the incorporation pattern of C in
compound 31 was determined to be 29-norgermanicol
13
13
[
23- C]-7 (marked with a closed square) to yield the C-la-
(
Figure 4 and Figures S14-8 and S14-13 in the Supporting Infor-
mation).
Similar to compounds 30 and 31, product 32-Ac did not
13
beled compounds 20–22. C-Labeled compound 7 was folded
in a chair–chair–chair–boat conformation, which, upon a ring
expansion, yielded the bis-norbaccharenyl cation 34 through
a normal cyclization pathway. However, further cyclization re-
sulted in the formation of the secondary cation 35 with a six-
membered chair ring (Scheme 3, path a). This pathway is un-
usual because the bis-norlupanyl cation 36 was not produced
as the intermediate. If the cyclization pathway proceeded via
have the vinylic Me group observed in substrate 9, indicating
that product 32 also had a pentacyclic skeleton. The presence
of the methylidene moiety was demonstrated by NMR spec-
troscopy (600 MHz, C D ): d =4.93 (brs, 1H; H-29), 4.94 ppm
6
6
H
(
brs, 1H; H-29), these two protons correlated with the C-29
atom [d =107.8 ppm (t)] in the HSQC spectrum. The H-18
C
cation 36 (Scheme 3, path b), the intermediate 37 (like cation
5) would have been produced; however, the C-labeled posi-
[
(
d =1.34 ppm (m, 1H)] was spin-coupled with H-19 [d =2.27
13
H
H
3
1
1
m), 2.39 ppm (m)], which was confirmed by the H- H COSY
spectrum. Both H-19 and H-21 [d =1.84 (m), 2.41 ppm (m)]
H
had HMBC cross peaks with the C-29 and C-20 atoms [d =
C
1
49.9 ppm (s)]. This finding indicated that the methylidene
moiety was positioned at the C-20ÀC-29 bond. In the NOESY
spectrum, strong NOEs between Me-27 [d =0.973 ppm (s,
H
3
H)/H-18/H-9 [d =1.28 ppm (dd, J=12.7, 3.3 Hz)] were ob-
H
served, indicating that the H-18 and H-9 exhibited a disposi-
tion. The detailed analyses of the other NMR data (Figure S15-
8
in the Supporting Information) verified that the structure of
compound 32 was 29-nortaraxasterol (29-norurs-20(30)-en-3b-
ol), as shown in Figure 4.
The highly polar product 33 showed the following two alco-
holic carbon atoms (C D , 150 MHz): d =78.50 (d, C-3),
6
6
C
7
4.67 ppm (s, C-20). One olefinic methyl group [d =1.74 ppm
H
(
d, J=4.9 Hz)] and one double bond [d =5.62 ppm (m, 2H; H-
H
2
4, H-25), d =132.3 (d), 124.8 ppm (d)] were observed in the
C
1
13
H (C D , 600 MHz) and C NMR spectra of compound 33;
6
6
these chemical shifts were almost the same as those of the ter-
minal double bond of compound 9. No other vinylic methyl
group was observed in the NMR spectra of compound 33.
Thus, compound 33 was a tetracyclic compound. The Me-21
13
2
2
Scheme 3. The polycyclization mechanism of [23- C]- and [23- H ]-labeled
compound 7 to form products 20–22.
[15c]
13
2
The positions of the C and H iso-
topes are marked with a closed squares and D, respectively. The labeling ex-
[15c]
periments conducted by Abe et al.
have demonstrated that the bisnor-
substrate 7 undergoes polycyclization through path a (without enzymatic
control), which significantly differs from the normal pathway (i.e., path b,
also shown in Scheme 1). Product 21’, which has a structure similar to that
of product 21, was postulated to be produced through path b, but com-
pound 21’ was not found as the enzymatic product. Compounds 21’ and 21
differ only in their isotope-labeling positions. This scheme shows the reason
of why compound 7 undergoes the cyclization through path a, but not
through path b.
group [d =1.19 ppm (s, 3H)] had HMBC cross peaks with the
H
C-20, C-17 [d =50.46 ppm (d)], and C-22 atoms [d =
C
C
4
0.69 ppm (t)], which suggested that the OH group was at-
tached to the C-20 group. The H-17 [d =1.78 ppm (m, 1H)]
H
exhibited an a arrangement because a strong NOE was ob-
served between the H-17 and the Me-29 group [d =
H
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15776
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tion differs between cations 35 and 37. The labeling positions
of the deuterium isotopes also differ between paths a and
b (Scheme 3), that is, the C-19 atom for path a and the C-20
atom for path b. Intermediate 35 was subjected to the follow-
ing backbone rearrangements: D-19b!C-20, H-18a!C19, and
H-13b!C18, which was followed by deprotonation of H-12a
to yield compound 21. Final proton elimination reactions from
H-13b and H-18a of cation 35 yielded compounds 20 and 22,
respectively. Path a (Scheme 3) provided products with the
deuterium atoms positioned at both C-19 and C-20 atoms.
However, path b (Scheme 3) leads to product 21’, which has
1
3
the deuterium and C only at the C-20 position; however,
[
15c]
compound 21’ was not observed as the enzymatic product.
These results raise the question why the cyclization proceed-
ed through path a and not through path b (Scheme 3). The
secondary cation 35 is more stable than the primary cation 36.
Furthermore, a six-membered ring is less hindered than a five-
membered one. Therefore, E-ring formation for the enzymatic
reaction of compound 7 would have proceeded under thermo-
[15c]
dynamic control but not through enzymatic control.
Howev-
er, in the case of substrate 1, the Me-24 and Me-30 groups
were transformed in a stereospecific fashion into the Me-30
and Me-29 groups of compound 2, respectively, as shown in
Figure 2, indicating that the terminal methyl groups were
strongly captured by this enzyme. To address the question of
which terminal methyl group (Me-24 or Me-30 of compound 1,
or both) is critical to the normal cyclization cascade, 24-noroxi-
dosqualene (8) and 30-noroxidosqualene (9) were subjected to
enzymatic reactions.
Scheme 4 shows the cyclization pathway of compound 8
with the Z-Me moiety at the terminus to afford the main prod-
uct 24 and small amounts of compounds 23, 30, and 31. The
substrate analogue 8 was folded in a chair–chair–chair–boat–
boat conformation to form a norbaccharenyl cation (4’ with Z-
Me, like compound 4). Further cyclization could afford the 30-
norlupanyl cation 5’ (like cation 5), which could undergo a ring
expansion to form the 30-noroleanyl cation 6’ (like cation 6),
as shown in path a (Scheme 4). Wagner–Meerwein rearrange-
ment reactions of H-18a to C-19 and H-13b to C-18 in antipar-
allel fashion, followed by elimination of H-12a, could give com-
pound 24 in a high yield (95.5%). This cyclization mechanism
is identical to that shown in Scheme 1 (normal pathway).
To confirm this pathway from compound 8 to compound
Scheme 4. Cyclization pathways of compound 8 to yield the main product
24 and the minor products 23, 30, and 31. Product 24 is generated through
path a, which is identical to the pathway shown in Scheme 1. The minor
products 30 and 31 are produced either through path b or path c. Both
paths b and c involve bond rotations through the C-19ÀC-20 axis of inter-
mediate 5’ (like cation 5) and through the C-20ÀC-21 axis of intermediate 4’
(like intermediate 4), respectively. The bond rotations indicate that the affini-
ty was decreased between the recognition site of the cyclase and the termi-
nal methyl group; however, the production of compounds 30 and 31 was
negligible. Compound 23 was produced as a result of the difficulty of the
ring expansion of the lupanyl cation 5’ (like cation 5) to the oleanyl cation 6’
(like cation 6), thus, the polycyclization cascade terminated at the stage of
the lupanyl cation with a 6/6/6/6/5-fused ring system but with a negligible
quantity. The netted graphic represents the binding site of the Me-30 group
of substrate 1. The analogue 8 possesses a Me-30 group. The tight associa-
tion of the Me-30 group of compound 8 with the recognition site could fur-
nish the normal folding conformation of a chair–chair–chair–boat–boat
structure, resulting in the complete cyclization without any production of
the premature cyclization products.
13
24, [23- C]-labeled compound 8 was incubated with this b-
amyrin synthase. The labeled compound 8 was synthesized by
the method shown in Figure S1 in the Supporting Information.
1
3
A mixture of [1- C]-bromoethane and triphenylphosphine was
through the mechanism depicted in Scheme 1 but not through
the mechanism shown in Scheme 3. Cation 5’ underwent 1,2-
shifts of the three hydrogen atoms of H-19b to C-20, H-18a to
C-19, and H-13b to C-18 in an antiparallel manner, followed by
elimination of H-12a, which resulted in the production of com-
pound 23 (1.5%). Notably, the stereochemistry at the C-20
atom of compounds 30 and 31 is opposite (i.e., S) that of com-
pound 24. The bond rotation in cation 5’ through the C-19ÀC-
20 bond afforded the 29-norlupanyl cation 5’’, which was then
subjected to a ring-enlargement process to produce the 29-
noroleanyl cation 6’’; then, the sequential 1,2-hydride shifts
1
3
heated to reflux in toluene to give C-ethyltriphenylphospho-
nium bromide, which was then subjected to a Wittig reaction
13
with the C -aldehyde 11 and NaN(SiMe ) , yielding the C-la-
beled compound 8. Figure 5 depicts the C NMR spectra of
27
3 2
1
3
1
3
the unlabeled compound 24 (Figure 5A) and the [ C]-labeled
compound 24 (Figure 5B).
The C signal was selectively enriched at the C-20 position
of the labeled compound 24, but no enrichment of the signal
of the C-19 position was observed, verifying that compound
13
2
4 was generated according to path a of Scheme 4, that is,
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15777
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
a chair–chair–chair–boat conformation to give the 27-nordam-
marenyl cation 3’’ (like cation 3) through path a (Scheme 5).
The tetracyclic products 25, 26, and 33 were produced in
a yield of 31.1% from intermediate 3’’ (Scheme 5, path b).
Attack of a water molecule on the C-20 position of the cation
3
’’ gave compound 33. Deprotonation from the Me-21 group
furnished compound 25. Four steps of sequential 1,2-shift re-
actions (H-17a!C-20, H-13b!C-17, Me-29a!C-13, and Me-
1
8b!C-14), followed by removal of H-7, yielded compound
2
6, where the least motion of the side chain (608 rotation)
prior to the shift of H-17a to C-20 resulted in a (20R) configura-
tion. In contrast to compound 26, product 28 had a (20S) ste-
reochemistry. The production of compound 28 also resulted
from 1,2-shifts of hydrides and Me groups and the subsequent
deprotonation of H-7, similar to the production of compound
1
3
Figure 5. C NMR spectrum of product 24 obtained by incubating the unla-
beled compound 8 with the EtAS (A) and that obtained by incubating the
1
3
[
23- C]-labeled compound 8 with this enzyme (B). A comparison of A) and
13
B) indicated that the regiospecific incorporation of C at the C-20 position
occurred. This result demonstrates that substrate 8 was converted into com-
pound 24 through the pathway shown in Scheme 1, that is, path a of
Scheme 4 and that the labeling position was different from that of the bis-
noroxidosqualene 7 (compare Scheme 3 with Scheme 4). In the case of sub-
26. A large motion angle (1208 rotation) of the side chain of in-
termediate 3’’ (Scheme 5, path c’) may have provided a (20S)
stereochemistry. However, the alternative pathway may also be
viable: analogue 9 could be folded in a chair–chair–chair–chair
conformation to afford the 17-epi-27-nordammarenyl cation
3’’’ with H-17b (Scheme 5, path c). A small motion (608 rota-
tion) of the side chain of intermediate 3’’’ could afford a (20S)
[
15c]
strate 7, the C-19 position has been reported to be labeled.
and deprotonation of H-12a gave compound 30 (1.5%),
whereas proton elimination of H-18a gave compound 31
1.5%). With respect to the formation of compounds 30 and
1, the alternative pathway (path c in Scheme 4) would also be
possible. Cation 4’ could undergo bond rotation through the
C-20ÀC-21 bond to afford the cation 34’ (like cation 34,
Scheme 3), then, the terminal double bond could attack to the
C-18 cation to give the 6/6/6/6/6-fused cation 35’ (like cation
[
18]
[13e]
(
configuration. The groups of Rohmer and Corey
have in-
3
troduced the concept of a limited motion of the side chain to
determine the C-20 stereochemistry. We surmise that this con-
cept may be applicable to the enzymatic reaction by using
both the original substrate (i.e., 2,3-oxidosqualene) and intact
cyclases (wild-type enzymes).
3
5), which could then undergo 1,2-hydride and methyl shifts
A constraint of substrate 1 imposed by the enzyme may
have been released, as indicated by the absence of the Me-30
group in compound 1, which, in turn, may have allowed
a large or free motion of the side chain of compound 3“
(Scheme 5, path c’), possibly leading to a (20S) stereochemistry.
Construction of a chair structure, but not a boat structure, for
the D-ring formation of compound 3’’’ would further suggest
that somewhat free motion had occurred around the D-ring
formation site because a chair structure is thermodynamically
more stable than a boat structure. In any case, the looser bind-
ing of the terminal Me group would have afforded product 28.
A ring expansion to the six-membered ring from the five-mem-
bered ring in structure 3’’ gave the 30-norbaccharenyl cation
4’’ (like cation 4), which underwent further cyclization to pro-
duce the 29-norlupanyl cation 5’’ (like cation 5, 35.4%) with
the terminal isoprene-like unit having been correctly folded.
However, the side chain of compound 4’’ was improperly ar-
ranged as a result of the bond rotation through the C-20ÀC-21
axis in compound 4’’ (curved arrow), to give the cation 4’’’, as
shown in path d (Scheme 5). This cation could undergo further
cyclization, yielding compound 5’’’, which bears H-18b and H-
19a (7.1%) conformations. Notably, the configurations at the
C-18 and C-19 atoms in compound 5’’’ are opposite those in
compound 5’’. Sequential hydride shifts (H-19a!C-20, H-
18b!C-19, and H-13b!C-18) and the subsequent elimination
of H-12 furnished compound 27 as the enzymatic product. A
normal polycyclization cascade gave the 29-noroleanyl cation
6’’. 1,2-Hydride shifts in an antiparallel fashion (H-18a!C-19,
H-13b!C-18) in cation 6’’, followed by the elimination of H-
and deprotonation of H-12a, yielding compound 30. The
proton elimination of H-18a of compound 35’ gave product
31, where no ring expansion process from the lupanyl cation
to the oleanyl cation is involved. This alternative pathway is
analogous to the cyclization reaction of compound 7, which is
shown in Scheme 3. The occurrence of bond rotation observed
in paths b and c (Scheme 4) led to the opposite configurations
of the C-20 atom between compounds 24 and 30 (31) and fur-
ther indicates that the binding of the Me-30 group of com-
pound 1 with the recognition site of the cyclase was weak-
ened. However, the production of the aberrant cyclization
products 30 and 31 was negligible (3%). A lack of the Me-24
group also resulted in the production of compound 23, indi-
cating that the ring-expansion process from cation 5’ to inter-
mediate 6’ was suppressed slightly (1.5%) and that the tight
binding facilitates the ring expansion process from cation 5’
(
5) to cation 6’ (6). Therefore, the Me-24 group (E-Me) of com-
pound 1 is concluded to have little effect on the polycycli-
zation pathway and the Me-30 group (Z-Me) of compound 1 is
concluded to play a key role in adopting the normal folding
conformation of compound 1 to generate compound 2. The
recognition site for the Me-30 group tightly holds the Me-30
group to construct the final oleanyl cations 6 or 6’, whereas
the binding site for the Me-24 group is significantly loose, so
that the small hydrogen atom at the Me-24 position is also ac-
ceptable to form the final cations 6 or 6’.
Scheme 5 illustrates the polycyclization mechanisms of com-
pound 9 to the products 25–33. This analogue was folded in
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15778
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 5. Cyclization pathways of compound 9 to yield the products 25–33. Paths a and b show the normal cyclization pathway. Products 25, 26, 29, 30,
3
1, 32, and 33 were generated through the normal folding conformation, as does compound 1 undergo. Path c, c’, and d show the unusual conformation
and the aberrant cyclization pathways, leading to the production of compounds 27 and 28. Products 25, 26, 28, 29, and 33 with tetracyclic ring systems
were produced due to the decreased binding of the nor-substrate to the cyclase, leading to the failure to hold a normal folding conformation. Thus, the poly-
cyclization cascade terminated at the premature cyclization stage (i.e., the tetracyclic stage) without directing the expansion from a five- to a six-membered
ring (3!4) and further cyclization (4!5). The netted image represents the binding site of the Me-30 group of substrate 1.
1
2, resulted in the formation of compound 30. Elimination of
sulted in the production of compounds 27 and 28 (16.2%). A
high yield of the abortive cyclization products with a tetracyclic
core can be attributed to the looser binding of compound 9
around the E-ring formation site. This decreased binding
halted the polycyclization cascade at the premature tetracyclic
stage. Thus, we conclude that the Z-Me moiety (30-Me) at the
terminus of compound 1 is more strongly captured by its rec-
ognition site of the EtAS and that the free motion of the Z-Me
group is prohibited in the reaction cavity, leading to the cor-
rect folding conformation shown in Scheme 1 and to the selec-
tive transformation of the 30-Me group of compound 1 into
the Me-29 group of compound 2. The loss of the E-Me group
(substrate 8) had little to no effect on the polycyclization path-
way, indicating that the affinity of the E-Me group of com-
pound 1 toward its binding site is very loose. Notably, the final
deprotonation reaction of compound 8 occurred only at the C-
12 atom, which suggests that the formed intermediate 6’ is
not distorted; that is, it is identical to the ordered shape of
H-18 in intermediate 6’’ gave compound 31. A 1,2-shift of the
H-20a to the C-19 atom and deprotonation from the Me-30
group provided compound 32. The formation of compound
2
9 can be explained in terms of the shift of H-13b to the C-18
atom in cation 4’’, followed by elimination of H-12a, to create
the C-12ÀC-13 double bond.
As shown in Figure 4, only a few products were produced
from compound 8, but a variety of enzymatic products were
obtained from compound 9 (Scheme 5). The normal cyclization
product 24 (30-nor-b-amyrin) was generated from compound
8
in a very high yield of 95.5% (Scheme 4), whereas the yield
of compound 30 (29-nor-b-amyrin) from compound 9 was
only 18.6% (Scheme 5). Furthermore, analogue 8 produced no
tetracyclic product. However, compound 9 afforded a large
amount of the tetracyclic products (41.4%) that were furnished
from the 27-nordammarenyl cation 3’’ and the 30-norbacchar-
enyl cation 4’’. Furthermore, the misfolding conformation re-
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15779
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
structure 6. Thus, the H-12 remains in proximity to the basic
active site residue(s) responsible for abstracting the H-12 of in-
termediate 6’, resulting in the almost exclusive production of
compound 24 as oleanane skeletons. In contrast, compound 9
afforded the three oleanane skeletons 30–32, which differ in
the deprotonation positions. The formed intermediate 6’’
would have been somewhat distorted in shape, such that the
H-12 of structure 6’’ could be arranged distal to the normal po-
sition of the basic residue(s). This arrangement may have de-
creased the production of compound 30, resulting in the pro-
duction of compounds 31 and 32 with the different double-
bond positions. This pathway also supports the idea that the
Me-30 group plays critical roles in constructing the ordered ar-
chitecture of a chair–chair–chair–boat–boat structure, facilitat-
ing the ring-expansion reactions, and performing the final re-
action of the proton elimination at the correct position.
enzyme quantities used. At the high concentration (i.e., 2 mg
of the enzyme), the whole amount of the substrate (3S)-1 used
was completely converted into b-amyrin (ꢁ100%). Notably, all
of the nor-analogues 7–9 had decreased conversion yields
compared with that of compound 1. The conversion ratio of
(3S)-8 (66%) was significantly higher than that of (3S)-9 (29%),
indicating that compound 8 is a better substrate. This result
further supports the aforementioned hypothesis that the Z-Me
group (i.e., the Me-30 group) at the terminus more tightly
binds to the enzyme than the E-Me group (i.e., the Me-24
group), that is, the lack of the Z-Me group (analogue 9) signifi-
cantly decreased the yields. This fact was further demonstrated
by the finding that the conversion yields at the lower concen-
tration of the EtAS (i.e., 1.0 or 0.5 mg) decreased in the order
1>8>7>9 (Figure 6), the conversion yields of compounds 1,
8, and 9 were estimated to be approximately 80–90, 25–35,
and 11–14%, respectively (Figure 6). This result also indicated
that compound 8 is a more favorable substrate than com-
pound 9. In contrast, compound 7 had a relatively higher yield
(46%) than compound 9 (29%). The binding of compound 7
to this enzyme is assumed to be decreased compared to that
of compound 9 because of the lack of two Me groups; howev-
er, compound 7 had a higher conversion yield than compound
Relative conversion yields of compounds 7, 8, and 9 com-
pared to that of compound 1
We evaluated the enzymatic activity of each of the substrates
by estimating the conversion yields of compound 1 and com-
pounds 7–9. Figure 6 shows the relative activities of com-
pounds 7–9 against that of compound 1. The quantities of the
9
(Figure 6), which raises the questions of why does com-
pound 7 have a higher conversion yield than compound 9 and
why does compound 7 afford only the fully cyclized 6/6/6/6/6-
fused pentacyclic products 20–22 (90.1%) without any produc-
tion of the premature cyclization products.
(
3R,S) substrates employed were 200 mg. Enzyme quantities of
0
7
.5, 1.0, and 2.0 mg were used. Incubations were conducted for
[6]
.0 h under optimized catalytic conditions, where each of the
(
3S) substrates can only be converted into products, but their
Careful inspection of the EI MS fragments of peaks A and B
in Figure 3 showed that the carbon frameworks of compounds
A and B did not correspond to any type of (6/6/6/5-fused) tet-
racycle or (6/6/6/6/5-fused) pentacycle such as those observed
in compounds 23, 25, 26, 27, 28, and 33 (compare the MS
spectra of structures A, B, 23, 25, 26, 27, 28, and 33, which are
shown in Figure S2 in the Supporting Information), possibly
suggesting that compounds A and B would be completely cy-
clized products (6/6/6/6/6-fused skeleton, see also the com-
ments on p. S15 in the Supporting Information). The produc-
tion of the unidentified compounds A and B accounted for
only 9.9% among all the products from compound 7. If we hy-
pothesize that compounds A and B are tetracyclic products,
then the production of the tetracyclic products is much lower
enantiomer (3R) substrates remain unreacted. The conversion
yields were plotted by estimating the GC peak areas of all the
products and the unreacted substrates remaining after incuba-
tions (Figure 3). The conversion yields were dependent on the
(
only 9.9%) compared to that of compound 9 (41.4%), despite
the loss of another methyl group; that is, a higher conversion
yield of compound 7 than of compound 9 cannot be fully ex-
plained by the binding force between the methyl group and
the recognition site. This result further demonstrates that the
cyclization mechanism of compound 7 is quite different from
Figure 6. Relative enzymatic activities of the EtAS toward 24-noroxidosqua-
lene (8), 30-noroxidosqualene (9), and 24,30-bisnoroxidosqualene (7) com-
pared to that of oxidosqualene (1). The activities were estimated by measur-
ing the total amounts of the products on the basis of the GC peak areas.
The activities measured at three different EtAS concentrations are shown.
This figure indicates that, on the basis of its highest conversion ratio, ana-
logue 8 is the superior substrate for the EtAS among the analogues 7, 8,
and 9. OXSQ represents oxidosqualene. Incubations were conducted at cata-
that of compounds 8 and 9 (compare Scheme 3 with
[15e]
Schemes 4 and 5). In a previous paper,
we reported that 29-
noroxidosqualene exhibited a strong time-dependent inhibi-
tion, resulting in very poor conversion (ꢀ3%) of the 29-nor-
analogue by the EtAS. However, compounds 7, 8, and 9 had
a relatively high conversion yield, as shown in Figure 6. Kitz–
Wilson plots revealed that the time-dependent inhibition was
minimal (Figure S17 in the Supporting Information) for the en-
zymatic reactions of the analogues 7, 8, and 9.
[
6]
lytically optimal conditions for 7.0 h. The incubation scales were as follows:
00 mg for the (3R,S) substrates; 2, 1 and 0.5 mg for the EtAS enzyme used in
2
the experiments. After incubation was completed, 20 mg of squalene were
added as an internal standard for the GC measurements. The use of 2 mg of
the EtAS afforded the approximately complete conversion yield (100%). The
error bars show the difference for three independent experiments.
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15780
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tion (4!5). The hydrogen atom substituted at the C-23 posi-
tion of compound 9 is too small to be tightly captured by the
Me-30-recognition site, resulting in the failure to maintain the
normal folding conformation. This conformational failure
would have led to a significantly high production of the tetra-
cyclic products. This fact further suggests that the polycycli-
zation of compound 1 toward the tetracyclic stage proceeds
under concerted catalysis, but the fifth ring formation requires
time for the complicated movement of the last isoprene unit.
This suggestion has already been proposed by the group of
Schulz for the fifth ring formation of hopene by SHC based on
Conclusion
To inspect the role of the two methyl groups at the terminus
of 2,3-oxidosqualene (1) on the polycyclization cascade by b-
amyrin synthase, we synthesized 24-noroxidosqualene (8) and
3
0-noroxidosqualene (9), which lack the E-Me and Z-Me group,
respectively, at the C-23 position of compound 1. These sub-
strate analogues were subjected to the enzymatic reactions
under catalytically optimal conditions. The substrate analogue
8
afforded 30-nor-b-amyrin (24) (95.5%) almost exclusively. The
minor products (19S)-30-norlup-12-en-3b-ol (23) (1.5%), 29-
nor-b-amyrin (30) (1.5%), and 29-norgermanicol (31) (1.5%)
were identified. Substrate 9 was converted into the following
enzymatic products consisting of a 6/6/6/5-fused tetracycle,
a 6/6/6/6-fused tetracycle, and a 6/6/6/6/6-fused pentacycle:
[
19a]
the co-crystal structure complexed with 2-azasqualene.
On
the other hand, the Me-24 binding site can accept both the
hydrogen atom and the Me-24 group without any disturbance
to the normal folding conformation, indicating that the affinity
of the E-Me group of compound 1 toward its binding site is
very loose. Thus, the Me-30 group is concluded to play critical
roles in constructing the ordered architecture of a chair–chair–
chair–boat–boat structure, facilitating the ring expansion reac-
tions, and performing the final proton elimination at the cor-
rect position. We have revealed that the critical role of the Me-
30 group is also true of the polycyclization cascade of squalene
2
7-nordammara-20(21),24-dien-3b-ol (25) (3.9%), 27-norbutyro-
spermol (27-noreuph-7(8),24-diene-3b-ol) (26) (7.7%), (19R)-30-
norlup-12-en-3b-ol (27) (7.1%), 27-nortirucalla-7(8),24-diene-3b-
ol (28) (9.1%), 30-norbacchara-12, 21-diene-3b-ol (29) (10.3%),
2
2
2
9-nor-b-amyrin (30) (18.6%), 29-norgermanicol (31) (12.9%),
9-nortaraxasterol (29-norurs-20(30)-en-3b-ol) (32) (3.9%), and
7-nordammara-3b, 20-diol (33) (19.5%). The cyclization mech-
[
14d,h]
anisms leading to these products are proposed as shown in
Schemes 4 and 5. Analogue 9 produced large amounts of nu-
merous abortive cyclization products (64.6%) and the fully cy-
clized product (35.4%) having the 6/6/6/6/6-fused pentacycle;
however, the b-amyrin skeleton 30 was present in only 18.6%
of these products. In contrast, analogue 8 preferentially pro-
duced the b-amyrin skeleton 24. Furthermore, the Me-30
group (Z-Me) of compound 8 was selectively transformed into
the Me-29 group of the b-amyrin scaffold, similar to the cycli-
zation reaction of compound 1. These findings strongly indi-
cate that the Me-30 group (Z-Me) of compound 8 is essential
to the correct folding of compound 1 for the b-amyrin biosyn-
thesis, whereas the role of the Me-24 group (E-Me) is minor.
Abe et al. conducted a cyclization reaction of 24,30-bis-noroxi-
by SHC to form 6/6/6/6/5-fused pentacyclic hopene.
The
fifth ring construction of hopene is facilitated by stabilizing the
hopanyl cation by the Phe601 and Phe605 residues through
[
2a,b,9a,e,19]
cation–p interactions.
The active-site residues for the
formation of the fifth ring in the b-amyrin skeleton and the
effect of the larger steric size at the C-23 position (e.g., substi-
tution with an ethyl group) on the polycyclization reactions
will be reported in due course.
Experimental Section
Materials and methods
NMR spectra of the enzymatic products were recorded in CDCl on
3
[
15c]
Bruker DMX 600 and DPX 400 spectrometers, the chemical shifts
dosqualene (7),
they reported that the C-19 position of the
are given in [ppm] relative to the solvent peak d =7.26 and d =
H
13
C
bisnor-b-amyrin product was labeled by the feeding experi-
1
7
7.0 ppm as the internal reference for the H and C NMR spectra,
13
2
ments by using [23- C]- and [23- H ]-labeled compound 7. In
2
respectively. In the C D solution, the chemical shifts are given in
6
6
contrast, the C-20 position of compound 24 was selectively la-
[
ppm] relative to the solvent peak (d =7.28 and d =128.0 ppm).
H C
1
3
beled by the enzymatic reaction of the [23- C]-labeled com-
pound 8. This fact further demonstrates that the cyclization of
compound 8 to compound 24 proceeded in the same manner
as that of compound 1 to b-amyrin, as depicted in Scheme 4,
but the cyclization pathway of compound 8 is significantly dif-
ferent from that of compound 7, although the same scaffold
with a 6/6/6/6/6-fuse pentacyclic ring system was constructed
both from compounds 8 and 7. No time-dependent inhibition
was observed for the enzymatic reactions of compounds 7, 8,
and 9; thus, these analogues do not act as suicide substrates.
The production of compounds 25, 26, 28, 29, and 33 with
tetracyclic ring systems from compound 9 occurred because of
the decreased binding of the substrate to the cyclase, which
led to the failure to hold a normal folding conformation. Thus,
the polycyclization cascade of compound 9 terminated at the
premature cyclization stage without directing the expansion
from a five- to a six-membered ring (3!4) and further cycliza-
The coupling constants J are given in [Hz]. GC analyses were done
on a Shimadzu GC-2014 chromatograph equipped with a flame
ionization detector (a DB-1 capillary column, 30 m”0.25 mm”
0
.25 mm, J&W Scientific). GC-MS spectra were recorded on
a JEOLSX100 or a JEOL JMS-Q1000GCK9 instrument equipped with
a ZB-5 ms capillary column (30 m”0.25 mm”0.25 mm; Zebron) by
using the EI mode operated at 70 eV. High-resolution mass spec-
trometry (HRMS) was performed on a JMS-T100GCV by using the
electron ionization (EI) mode. HPLC was carried out with a Hitachi
L-1700 (pump) and L-7405 (UV detector), the HPLC peaks were
monitored at l=210 or 214 nm. Specific rotation values were mea-
sured with a Horiba SEPA-300 polarimeter.
Incubation experiments of the analogues 7–9
The purification of the EtAS and the incubation conditions were
carried out according to the published protocol. The detailed in-
cubation conditions for Figures 3 and 6 are described in the text
and the captions of the figures.
[6]
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15781
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Synthetic experiments for the substrate analogues 7–10
1H; H-18), 1.87 (m, 2H; H-11), 1.97 (m, 1H; H-16), 2.046 (s, 3H;
CH CO), 4.50 (dd, J=11.9, 4.5 Hz, 1H; H-3), 5.20 ppm (brt, J=3.4,
3
The detailed synthetic procedures and the spectral data of the syn-
thetic compounds are described in the Supporting Information.
13
1
H; H-12); C NMR (150 MHz, CDCl ): d=15.56 (q, 1C; C-25), 16.69
3
(
q, 1C; C-24), 16.81 (q, 1C; C-26), 18.27 (t, 1C; C-6), 21.28 (q, 1C;
CH CO), 22.56 (q, 1C; C-29), 23.54 (t, 1C; C-2), 23.59 (t, 1C; C-11),
3
Spectroscopic data of the enzymatic products
25.85 (q, 1C; C-27), 26.31 (t, 1C; C-15), 27.22 (t, 1C; C-16), 28.04 (q,
1
1
C; C-23), 28.42 (q, 1C; C-28), 30.98 (t, 1C; C-21), 32.52 (s, 1C; C-
7), 32.65 (t, 1C; C-7), 33.67 (d, 1C; C-20), 36.86 (s, 1C; C-10), 37.73
2
D
0
1
Product 21 acetate: Solid; [a] = +58.1 (c=0.08, CHCl ); H NMR
3
(
(
600 MHz, CDCl ): d=0.767 (s, 3H; Me-28), 0.85- (m, 1H; H-5), 0.86
m, 1H; H-16), 0.863 (s, 3H; Me-24), 0.871 (s, 3H; Me-23), 0.961 (s,
H; Me-25), 0.97 (m, 1H; H-15), 0.971 (s, 3H; Me-26), 1.06 (m, 1H;
H-1), 1.10 (m, 1H; H-20), 1.135 (s, 3H; Me-27), 1.16 (m, 1H; H-22),
3
(
s, 1C; C-4), 38.31 (t, 1C; C-1), 39.80 (s, 1C; C-8), 41.54 (t, 1C; C-22),
4
1
1
1.72 (s, 1C; C-14), 43.19 (t, 1C; C-19), 47.59 (d, 1C; C-9), 51.72 (d,
C; C-18), 55.28 (d, 1C; C-5), 80.41 (d, 1C; C-3), 122.0 (d, 1C; C-12),
3
45.3 (s, 1C; C-13), 169.7 ppm (s, 1C; CH CO) (the assignments of
3
1
1
1
2
1
4
5
1
6
2
1
2
.34 (m, 1H; H-7), 1.36 (m, 1H; H-21), 1.36 (m, 1H; H-19), 1.39 (m,
H; H-6), 1.41 (m, 1H; H-22), 1.43 (m, 1H; H-21), 1.52 (m, 1H; H-6),
.53 (m, 1H; H-7), 1.58 (m, 1H; H-9), 1.60 (m, 1H; H-20), 1.62 (m,
H; H-2), 1.63 (m, 1H; H-1), 1.74 (m, 1H; H-18), 1.76 (m, 1H; H-19),
.78 (m, 1H; H-15), 1.87 (m, 2H; H-11), 2.02 (ddd, J=13.0, 13.0,
the C-2 and C-11 atoms may be exchangeable); HRMS (EI): m/z
calcd for C H O : 454.38108; found: 454.38040.
31
50
2
2
D
0
1
Product 25 acetate: Oil; [a] = +37.2 (c=0.047, CHCl ); H NMR
3
(
400 MHz, C D ): d=0.82 (dd, J=11.6, 4.8 Hz, 1H; H-5), 0.84 (m,
6 6
1
H; H-1), 0.878 (s, 3H; Me-19), 0.982 (s, 3H; Me-29), 1.025 (s, 3H;
.6 Hz, 1H; H-16), 2.046 (s, 3H; CH CO), 4.50 (t, J=8.8 Hz, 1H; H-3),
3
1
3
Me-27), 1.037 (s, 3H; Me-28), 1.042 (s, 3H; Me-18), 1.21 (m, 1H; H-
1), 1.22 (m, 1H; H-15), 1.30 (m, 1H; H-9), 1.33 (m, 1H; H-7), 1.40
m, 1H; H-12), 1.43 (m, 1H; H-6), 1.50 (m, 1H; H-11), 1.53 (m, 1H;
H-6), 1.56 (m, 1H; H-1), 1.62 (m, 1H; H-7), 1.65 (m, 1H; H-16), 1.72
m, 2H; H-2, H-15), 1.74 (d, J=5.2 Hz, 3H; Me-26), 1.85 (m, 1H; H-
.19 ppm (brs, 1H; H-12); C NMR (150 MHz, CDCl ): d=15.57 (q,
3
1
(
C; C-25), 16.69 (q, 1C; C-24), 16.83 (q, 1C; C-26), 18.29 (t, 1C; C-
), 21.28 (q, 1C; CH CO), 22.31 (t, 1C; C-21), 23.54 (t, 1C; C-11),
3
3.60 (t, 1C; C-2), 25.85 (q, 1C; C-27), 26.30 (t, 1C; C-15), 27.26 (t,
C; C-16), 27.40 (t, 1C; C-20), 28.06 (q, 1C; C-23), 28.73 (q, 1C; C-
8), 32.66 (t, 1C; C-7), 33.07 (s, 1C; C-17), 34.19 (t, 1C; C-19), 36.88
(
2
2
1
5
), 1.861 (s, 3H; CH CO), 1.87 (m, 1H; H-13), 2.08 (m, 1H; H-16),
.26 (m, 2H; H-22), 2.39 (m, 2H; H-23), 2.41 (m, 1H; H-12), 2.42 (m,
3
(
s, 1C; C-10), 37.74 (s, 1C; C-4), 38.33 (t, 1C; C-1), 39.83 (s, 1C; C-8),
H; H-17), 4.81 (dd, J=11.6, 4.4 Hz, 1H; H-3), 5.03 (s, 1H; H-21),
4
1
1
1.57 (t, 1C; C-22), 41.74 (s, 1C; C-14), 47.60 (d, 1C; C-9), 51.84 (d,
C; C-18), 55.31 (d, 1C; C-5), 80.98 (d, 1C; C-3), 121.5 (t, 1C; C-12),
1
3
.10 (s, 1H; H-21), 5.62 ppm (m, 2H; H-24, H-25); C NMR
(
(
100 MHz, C D ): d=15.75 (q, 1C; C-18), 16.17 (q, 1C; C-29), 16.41
6 6
q, 1C; C-19), 16.81 (q, 1C; C-28), 18.08 (q, 1C; C-26), 18.45 (t, 1C;
45.6 (s, 1C; C-13), 170.9 ppm (s, 1C; CH CO); HRMS (EI): m/z calcd
3
for C H O : 440.36543; found: 440.36481. The NMR data in C D
6
solution are given in Figures S3-9 and S3-10 in the Supporting In-
3
0
48
2
6
C-6), 20.85 (q, 1C; CH CO), 21.62 (t, 1C; C-11), 24.09 (t, 1C; C-2),
3
2
1
3
1
1
8.12 (q, 1C; C-27), 29.34 (t, 1C; C-16), 31.73 (t, 1C; C-15), 31.99 (t,
C; C-23), 34.10 (t, 1C; C-12), 34.70 (t, 1C; C-22), 35.64 (t, 1C; C-7),
7.23 (s, 1C; C-10), 38.07 (s, 1C; C-4), 38.77 (t, 1C; C-1), 40.69 (s,
C; C-8), 45.70 (d, 1C; C-13), 48.27 (d, 1C; C-17), 49.65 (s, 1C; C-
4), 51.04 (d, 1C; C-9), 56.14 (d, 1C; C-5), 80.47 (d, 1C; C-3), 108.2
formation. In C D solution, the signal of H-3 appeared at d =
6
6
H
4
.83 ppm, and the splitting pattern was dd (J=12.0, 4.4 Hz).
2
D
0
Product 23 acetate: Semi-solid; [a] = +68.5 (c=0.027, CHCl );
3
1
H NMR (600 MHz, C D ): d=0.874 (brd, J=11.3 Hz, 1H; H-5), 0.93
6
6
(
m, H-1), 0.974 (s, 3H; Me-25), 1.036 (s, 3H; Me-23), 1.047 (s, 3H;
Me-24), 1.088 (t, J=6.8 Hz, 3H; Me-29), 1.105 (s, 3H; Me-26), 1.12
m, 2H; H-15), 1.181 (s, 3H; Me-28), 1.276 (s, 3H; Me-27), 1.32 (m,
H; H-21, H-16), 1.42 (m, 1H; H-6), 1.44 (m, 1H; H-7), 1.50 (m, 1H;
H-1), 1.54 (m, 1H; H-6), 1.57 (m, 1H; H-7), 1.63 (m, 1H; H-9), 1.64
m, 2H; H-22), 1.71 (m, 1H; H-2), 1.72 (m, 2H; H-11, H-18), 1.80 (m,
(
t, 1C; C-21), 125.2 (d, 1C; C-25), 131.5 (d, 1C; C-24), 152.3 (s, 1C;
C-20), 169.9 ppm (s, CH CO); HRMS (EI): m/z calcd for C H O :
4
3
31 50
2
(
2
54.38108; found: 454.37986.
2
D
0
1
Product 26 acetate: Oil; [a] = +38.3 (c=0.06, CHCl ); H NMR
3
(
400 MHz, CDCl ): d=0.764 (s, 3H; Me-19), 0.801 (s, 3H; Me-18),
3
(
0
.829 (d, J=6.4 Hz, 1H; H-21), 0.849 (s, 3H; Me-27), 0.932 (s, 3H;
1
2
H; H-16), 1.83 (m, 1H; H-2), 1.879 (s, 3H; CH CO), 1.88 (m, 2H; H-
0), 1.92 (m, 1H; H-11), 2.05 (m, 1H; H-21), 2.06 (m, 1H; H-19), 4.83
3
Me-28), 0.971 (s, 3H; Me-29), 1.00 (m, 1H; H-22), 1.22 (m, 1H; H-1),
1
2
2
1
1
2
1
.25 (m, 1H; H-16), 1.40 (m, 1H; H-20), 1.42 (m, 1H; H-5), 1.45 (m,
H; H-15), 1.48 (m, 1H; H-17), 1.51 (m, 2H; H-11), 1.61 (m, 1H; H-
2), 1.64 (m, 2H; H-2), 1.64 (m, 3H; Me-26), 1.66 (m, 1H; H-12),
.68 (m, 1H; H-1), 1.79 (m, 1H; H-12), 1.84 (m, 1H; H-23), 1.91 (m,
(
dd, J=11.8, 4.5 Hz, H-3), 5.38 ppm (brt, J=3.4 Hz, 1H; H-12);
13
C NMR (150 MHz, C D ): d=12.91 (q, 1C; C-29), 15.71 (q, 1C; C-
6
6
2
(
5), 17.05 (q, 1C; C-24), 17.47 (q, 1C; C-26), 18.59 (t, 1C; C-6), 20.82
q, 1C; CH CO), 23.83 (t, 1C; C-11), 23.95 (t, 1C; C-2), 26.48 (q, 2C;
3
H; H-16), 1.96 (m, 1H; H-6), 2.054 (s, 3H; CH CO), 2.06 (m, 1H; H-
C-27, C-28), 27.06 (t, 1C; C-15), 27.83 (t, 1C; C-20), 28.23 (q, 1C; C-
3), 29.15 (t, 1C; C-21), 31.99 (t, 1C; C-16), 33.42 (t, 1C; C-7), 37.08
s, 1C; C-10), 37.89 (s, 1C; C-4), 38.39 (t, 1C; C-1), 39.76 (s, 1C; C-8),
3
3), 2.12 (m, 1H; H-6), 2.22 (m, 1H; H-9), 4.51 (dd, J=11.4, 3.8 Hz,
2
(
4
1
8
H; H-3), 5.25 (brs, 1H; H-7), 5.41 ppm (m, 2H; H-24, H-25);
1
3
C NMR (100 MHz, CDCl ): d=13.13 (q, 1C; C-19), 15.87 (q, 1C; C-
0.40 (t, 1C; C-22), 41.53 (s, 1C; C-17), 42.46 (s, 1C; C-14), 47.52 (t,
C; C-19), 47.94 (d, 1C; C-9), 55.69 (d, 1C; C-5), 61.24 (t, 1C; C-18),
0.58 (t, 1C; C-3), 124.1 (t, 1C; C-12), 140.1 (s, 1C; C-13), 169.9 ppm
3
2
2
8), 17.96 (q, 1C; C-26), 18.14 (t, 1C; C-11), 18.49 (q, 1C; C-21),
1.34 (q, 1C; CH CO), 22.05 (q, 1C; C-18), 23.75 (t, 1C; C-6), 24.19
3
(
t, 1C; C-2), 27.31 (q, 1C; C-29), 27.56 (q, 1C; C-27), 28.44 (t, 1C; C-
(
s, 1C; CH CO) (the assignment of the C-2 and C-11 atoms may be
3
1
3
6), 29.84 (t, 1C; C-23), 33.67 (t, 1C; C-12), 33.93 (t, 1C; C-15),
4.80 (s, 1C; C-10), 34.92 (t, 1C; C-22), 35.61 (d, 1C; C-20), 36.81 (t,
exchangeable); HRMS (EI): m/z calcd for C H O : 454.38108;
found: 454.38015.
3
1
50
2
2
D
0
1
1C; C-1), 37.83 (s, 1C; C-4), 43.51 (s, 1C; C-13), 48.78 (d, 1C; C-9),
Product 24 acetate: Solid; [a] = +80.6 (c=0.132, CHCl ); H NMR
(
0
3
5
1
1
0.73 (d, 1C; C-5), 51.27 (s, 1C; C-14), 53.18 (d, 1C; C-17), 81.14 (d,
C; C-3), 117.6 (d, 1C; C-7), 124.4 (d, 1C; C-25), 131.9 (d, 1C; C-24),
600 MHz, CDCl ): d=0.805 (s, 3H; Me-28), 0.85 (m, 1H; H-16),
3
.859 (d, J=6.5 Hz, 3H; Me-29), 0.86 (m, 1H; H-5), 0.864 (s, 3H;
45.9 (s, 1C; C-8), 171.0 ppm (s, 1C; CH CO); HRMS (EI): m/z calcd
3
Me-24), 0.874 (s, 3H; Me-23), 0.963 (s, 3H; Me-25), 0.972 (s, 3H;
Me-26), 1.06 (m, 2H; H-1, H-21), 1.135 (s, 3H; Me-27), 1.14 (m, 1H;
H-15), 1.23 (m, 1H; H-22), 1.25 (m, 1H; H-20), 1.34 (m, 2H; H-7, H-
9), 1.38 (m, 1H; H-6), 1.39 (m, 1H; H-21), 1.42 (m, 1H; H-22), 1.43
m, 1H; H-19), 1.52 (m, 1H; H-6), 1.53 (m, 1H; H-7), 1.58 (m, 1H; H-
), 1.62 (m, 2H; H-2), 1.63 (m, 1H; H-1), 1.77 (m, 1H; H-15), 1.80 (m,
for C H O : 454.38108; found: 454.38108.
31
50
2
2
D
0
1
Product 27 acetate: Oil; [a] =À21.5 (c=0.061, CHCl ); H NMR
3
1
(
9
(400 MHz, C D ): d=0.85 (m, 1H; H-1), 0.86 (m, 1H; H-5), 0.957 (s,
6 6
3H; Me-25), 1.031 (s, 3H; Me-26), 1.042 (s, 6H; Me-23, Me-24), 1.08
(m, 2H; H-15), 1.082 (t, J=7.0 Hz, 3H; Me-29), 1.159 (s, 3H; Me-27),
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15782
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
.215 (s, 3H; Me-28), 1.34 (m, 1H; H-20), 1.36 (m, 1H; H-16), 1.42
132.6 (d, 1C; C-22) 139.2 (s, 1C; C-13), 169.9 ppm (s, 1C; CH CO)
3
(
m, 1H; H-6), 1.44 (m, 1H; H-1), 1.46 (m, 1H; H-7), 1.52 (m, 1H; H-
(the assignments of the C-21 and C-22 atoms may be exchangea-
6
2
), 1.53 (m, 2H; H-7, H-22), 1.62 (m, 2H; H-9, H-16), 1.65 (m, 1H; H-
2), 1.70 (m, 1H; H-2), 1.78 (m, 2H; H-21), 1.82 (m, 1H; H-2), 1.872
ble); HRMS (EI): m/z calcd for C H O : 454.38119; found:
31
50
2
454.38108.
Product 30 acetate: White solid; [a] = +40.8 (c=0.083, CHCl );
(
s, 3H; CH CO), 1.92 (m, 2H; H-11), 1.93 (m, 1H; H-20), 2.08 (m, 1H;
20
D
3
3
H-19), 2.57 (d, J=11.2 Hz, 1H; H-18), 4.81 (dd, J=11.6, 4.4 Hz, 1H;
H-3), 5.34 ppm (brs, 1H; H-12); C NMR (100 MHz, C D ): d=14.16
1
H NMR (600 MHz, C D ): d=0.82 (m, 1H; H-16), 0.869 (brd, J=
6
6
1
3
6
6
1
2
2
1
1.6 Hz, 1H; H-5), 0.92 (m, 1H; H-1), 0.960 (d, J=7.2 Hz, 3H; Me-
(
q, 1C; C-29), 15.67 (q, 1C; C-25), 16.97 (q, 1C; C-24), 17.30 (q, 1C;
9), 0.962 (s, 3H; Me-25), 1.043 (s, 3H; Me-24), 1.047 (s, 3H; Me-
3), 1.088 (s, 3H; Me-28), 1.10 (m, 1H; H-15), 1.12 (m, 1H; H-19),
.129 (s, 3H; Me-26), 1.313 (s, 3H; Me-27), 1.34 (m, 1H; H-22), 1.37
C-26), 18.59 (t, 1C; C-6), 20.80 (q, 1C; CH CO), 23.26 (q, 1C; C-27),
3
2
1
2
3.77 (t, 1C; C-2), 23.84 (t, 1C; C-11), 25.49 (t, 1C; C-21), 26.18 (t,
C; C-15), 28.13 (q, 1C; C-23), 29.28 (q, 1C; C-28), 30.51 (t, 1C; C-
0), 32.96 (t, 1C; C-16), 33.59 (t, 1C; C-7), 37.09 (s, 1C; C-10), 37.79
(
m, 1H; H-7), 1.39 (m, 1H; H-21), 1.40 (m, 2H; H-6, H-7), 1.48 (m,
1
1
1
1
2
H; H-1), 1.55 (m, 1H; H-6), 1.64 (m, 1H; H-9), 1.65 (m, 1H; H-22),
.70 (m, 1H; H-2), 1.82 (m, 1H; H-2), 1.84 (m, 1H; H-11), 1.85 (m,
H; H-21), 1.92 (m, 1H; H-15), 1.93 (m, 1H; H-11), 2.00 (m, 1H; H-
(
s, 1C; C-4), 38.05 (t, 1C; C-1), 39.43 (s, 1C; C-8), 40.64 (s, 1C; C-17),
4
1
1
1.31 (s, 1C; C-14), 42.12 (t, 1C; C-22), 47.97 (d, 1C; C-9), 48.07 (d,
C; C-19), 54.17 (d, 1C; C-18), 55.63 (d, 1C; C-5), 80.94 (d, 1C; C-3),
9), 2.03 (m, 1H; H-16), 2.049 (s, 3H; CH CO), 2.08 (brs, , 1H; H-20),
3
26.1 (s, 1C; C-12), 139.6 (s, 1C; C-13), 169.9 ppm (s, 1C; CH CO)
3
.23 (m, 1H; H-18), 4.83 (dd, J=11.9, 4.5 Hz, 1H; H-3), 5.36 ppm
(
the assignments of the carbon signals between the C-2 and C-11
(
brt, J=3.4, 1H; H-12) (the assignments of the Me-23 and Me-24
atoms and between the Me-24 and Me-26 groups may be ex-
changeable); HRMS (EI): m/z calcd for C H O : 454.38108; found:
4
13
groups may be exchangeable); C NMR (150 MHz, C D ): d=15.64
6
6
3
1
50
2
(
q, 1C; C-25), 17.02 (q, 1C; C-24), 17.06 (q, 1C; C-26), 17.35 (q, 1C;
54.38108.
C-29), 18.56 (t, 1C; C-6), 20.82 (q, 1C; CH CO), 23.82 (t, 1C; C-11),
3
2
D
0
1
2
1
2
3.95 (t, 1C; C-2), 26.23 (q, 1C; C-27), 26.57 (t, 1C; C-15), 27.44 (t,
C; C-21), 27.47 (t, 1C; C-16), 28.23 (q, 1C; C-23), 28.82 (q, 1C; C-
8), 32.94 (t, 1C; C-7), 33.31 (s, 1C; C-17), 35.44 (t, 1C; C-22), 37.02
Product 28 acetate: Oil; [a] =À86.6 (c=0.072, CHCl ); H NMR
3
(
(
600 MHz, C D ): d=0.871 (s, 3H; Me-19), 0.982 (s, 3H; Me-27), 1.02
6 6
m, 1H; H-1), 1.029 (s, 3H; Me-29), 1.090 (d, J=6.2 Hz, 3H; Me-21),
.106 (s, 3H; Me-28), 1.138 (s, 3H; Me-18), 1.27 (m, 1H; H-22), 1.41
m, 1H; H-16), 1.43 (m, 1H; H-5), 1.45 (m, 1H; H-11), 1.54 (m, 1H;
H-11), 1.56 (m, 1H; H-1), 1.58 (m, 1H; H-20), 1.62 (m, 2H; H-15, H-
7), 1.72 (m, 1H; H-22), 1.75 (m, 3H; H-2, H-12, H-15), 1.777 (d, J=
.1 Hz, 3H; Me-26), 1.868 (s, 3H; CH CO), 1.88 (m, 1H; H-12), 1.89
(
1
(
3
s, 1C; C-10), 37.91 (s, 1C; C-4), 38.36 (t, 1C; C-1), 39.08 (t, 1C; C-
9), 39.50 (d, 1C; C-20), 40.12 (s, 1C; C-8), 42.00 (s, 1C; C-14), 45.49
d, 1C; C-18), 47.85 (d, 1C; C-9), 55.59 (d, 1C; C-5), 80.57 (d, 1C; C-
), 122.2 (d, 1C; C-12), 145.2 (s, 1C; C-13), 170.0 ppm (s, 1C;
1
(
1
5
CH CO) (the assignments of the C-2 and C-11 atoms may be ex-
3
3
changeable and the assignments of the C-16 and C-21 atoms may
(
2
(
m, 1H; H-2), 1.99 (m, 1H; H-1), 2.07 (m, 1H; H-16), 2.13 (m, 1H; H-
3), 2.16 (m, 1H; H-1), 2.25 (m, 1H; H-9), 2.30 (m, 1H; H-23), 4.83
dd, J=11.9, 4.5 Hz, 1H; H-3), 5.41 (brs, 1H; H-7), 5.64 ppm (m, 2H;
also be exchangeable); HRMS (EI): m/z calcd for C H O :
31 50
2
4
54.38119; found: 454.38148.
1
3
2
D
0
H-24, H-25); C NMR (150 MHz, C D ): d=13.33 (q, 1C; C-19), 16.09
(
6
6
Product 31 acetate: White solid; [a] = +29.4 (c=0.031, CHCl );
3
1
q, 1C; C-28), 18.08 (q, 1C; C-26), 18.41 (t, 1C; C-11), 18.60 (q, 1C;
H NMR (400 MHz, C D ): d=0.79 (m, 1H; H-5), 0.85 (ddd, J=13.2,
6
6
C-21), 20.82 (q, 1C; CH CO), 22.23 (q, 1C; C-29), 24.08 (t, 1C; C-6),
3
13.2, 3.6 Hz, 1H; H-1), 0.918 (s, 3H; Me-25), 1.004 (s, 3H; Me-27),
2
1
1
3
1
8
1
4.61 (t, 1C; C-2), 27.46 (q, 1C; C-18), 27.71 (q, 1C; C-27), 28.54 (t,
C; C-16), 30.09 (t, 1C; C-23), 34.25 (t, 1C; C-12), 34.44 (t, 1C; C-
5), 34.98 (s, 1C; C-10), 36.18 (d, 1C; C-20), 36.48 (t, 1C; C-22),
6.84 (t, 1C; C-1), 38.04 (s, 1C; C-4), 43.89 (s, 1C; C-13), 49.10 (d,
C; C-9), 51.08 (d, 1C; C-5), 51.48 (s, 1C; C-14), 53.45 (t, 1C; C-17),
0.57 (d, 1C; C-3), 118.4 (d, 1C; C-7), 124.6 (d, 1C; C-25), 132.3 (d,
1
1
.026 (s, 3H; Me-23), 1.039 (s, 3H; Me-24), 1.145 (s, 3H; Me-26),
.175 (d, J=6.8 Hz, 3H; Me-29), 1.22 (m, 1H; H-15), 1.243 (s, 3H;
Me-28), 1.26 (m, 1H; H-11), 1.28 (m, 1H; H-9), 1.33 (m, 1H; H-12),
.39 (m, 3H; H-6, H-7, H-21), 1.51 (m, 1H; H-11), 1.52 (m, 2H; H-6,
H-16), 1.53 (m, 1H; H-7), 1.59 (m, 1H; H-22), 1.61 (m, 1H; H-1), 1.62
m, 1H; H-16), 1.65 (m, 1H; H-12), 1.68 (m, 1H; H-22), 1.73 (m, 1H;
H-2), 1.76 (m, 1H; H-21), 1.87 (m, 1H; H-2), 1.883 (s, 3H; CH CO),
1
(
C; C-24), 146.0 (s, 1C; C-8), 169.9 ppm (s, 1C; CH CO) (the assign-
3
3
ments of the C-25 and C-25 atoms may be exchangeable); HRMS
EI): m/z calcd for C H O : 454.38123; found: 454.38108.
1
.94 (m, 1H; H-15), 2.35 (m, 1H; H-20), 2.43 (brd, 11.6 Hz, 1H; H-
(
3
1
50
2
13), 4.83 (dd, J=11.6, 4.8 Hz, 1H; H-3), 5.27 ppm (brs, 1H; H-19);
1
3
2
D
0
C NMR (100 MHz, C
26), 16.85 (q, 1C; C-25), 16.90 (q, 1C; C-24), 18.40 (t, 1C; C-6), 20.87
q, 1C; CH CO), 21.34 (t, 1C; C-11), 22.72 (q, 1C; C-29), 24.05 (t, 1C;
D ): d=15.12 (q, 1C; Me-27), 16.21 (q, 1C; C-
6 6
Product 29 acetate: White solid; [a] =À15.9 (c=0.011, CHCl );
3
1
H NMR (400 MHz, C D ): d=0.89 (brd, J=10.8 Hz, 1H; H-5), 0.94
6
6
(
3
(
m, 1H; H-1), 0.971 (s, 3H; Me-28), 0.991 (s, 3H; Me-25), 1.026 (d,
J=4.8 Hz, 3H; Me-29), 1.043 (s, 3H; Me-23), 1.049 (s, 3H; Me-24),
.143 (s, 3H; Me-26), 1.18 (m, 1H; H-15), 1.201 (s, 3H; Me-27), 1.42
m, 2H; H-6, H-16), 1.45 (m, 2H; H-19), 1.50 (m, 1H; H-20), 1.52 (m,
C-2), 25.69 (q, 1C; Me-28), 26.49 (t, 1C; C-12), 27.84 (t, 1C; C-15),
2
1
3
1
5
1
2
8.04 (q, 1C; C-23), 28.47 (t, 1C; C-21), 32.32 (d, 1C; C-20), 34.58 (s,
C; C-17), 34.78 (t, 1C; C-7), 37.20 (s, 1C; C-10), 37.97 (s, 1C; C-4),
8.44 (t, 1C; C-16), 38.54 (t, 1C; C-1), 39.00 (d, 1C; C-13), 40.91 (s,
C; C-8), 41.03 (t, 1C; C-22), 43.78 (s, 1C; C-14), 51.26 (d, 1C; C-9),
5.71 (d, 1C; C-5), 80.94 (d, 1C; C-3), 125.9 (d, 1C; C-19), 144.9 (s,
1
(
1
1
1
H; H-9), 1.53 (m, 2H; H-7, H-16), 1.55 (m, 2H; H-1, H-6), 1.63 (m,
H; H-7), 1.72 (m, 1H; H-2), 1.80 (m, 1H; H-11), 1.83 (m, 1H; H-2),
.85 (m, 1H; H-11), 1.88 (m, 1H; H-18), 1.881 (s, 3H; CH CO), 1.92
3
C; C-18), 169.9 ppm (s, 1C; CH CO) (the carbon signals of the C-
3
(
4
(
m, 1H; H-15), 2.18 (m, 1H; H-20), 2.31 (brd, J=16 Hz, 1H; H-18),
.84 (dd, J=11.6, 4.8 Hz, 1H; H-3), 5.39 (brs, 1H; H-12), 5.63 ppm
m, 2H; H-21, H-22); C NMR (100 MHz, C D ): d=16.11 (q, 1C; C-
6 6
4 and C-25 atoms may be exchangeable); HRMS (EI): m/z calcd for
1
3
C H O : 454.38119; found: 454.38077.
31 50 2
20
2
1
5), 17.16 (q, 1C; C-24), 17.66 (q, 1C; C-26), 18.17 (q, 1C; C-29),
Product 32 acetate: White solid; [a] = +86.7 (c=0.015, CHCl );
D
3
1
8.47 (t, 1C; C-6), 20.85 (q, 1C; CH CO), 21.80 (q, 1C; C-28), 21.97
H NMR (600 MHz, C D ): d=0.77 (brd, J=11.0 Hz, 1H; H-5), 0.83
6 6
3
(
q, 1C; C-27), 23.31 (t, 1C; C-11), 24.01 (t, 1C; C-2), 26.79 (t, 1C; C-
(brt, J=13.2, 1H; H-1), 0.888; (s, 3H; Me-25), 0.973 (s, 3H; Me-27),
0.981 (s, 3H; Me-28), 1.01 (m, 1H; H-12) 1.023 (s, 3H; Me-23), 1.036
(s, 6H; Me-24, Me-26), 1.08 (m, 1H; H-15), 1.28 (dd, J=12.7, 3.3 Hz,
1H; H-9), 1.32 (m, 1H; H-16), 1.34 (m, 1H; H-18), 1.37 (m, 1H; H-
22), 1.38 (m, 1H; H-6), 1.39 (m, 1H; H-7), 1.43 (m, 1H; H-16), 1.51
(m, 2H; H-6, H-22), 1.54 (m, 1H; H-13), 1.57 (m, 1H; H-1), 1.65 (m,
1
3
1
4
1
5), 27.21 (t, 1C; C-20), 28.27 (q, 1C; C-23), 33.69 (t, 1C; C-16),
4.38 (t, 1C; C-7), 34.39 (s, 1C; C-17), 37.02 (s. 1C; C-10), 37.92 (s,
C; C-4), 38.68 (t, 1C; C-1), 39.34 (s, 1C; C-8), 43.28 (s, 1C; C-14),
5.03 (t, 1C; C-18), 45.75 (t, 1C; C-19), 48.03 (d, 1C; C-9), 55.78 (d,
C; C-5), 80.52 (d, 1C; C-3), 121.2 (d, 1C; C-12), 124.4 (d, 1C; C-21),
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15783
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
H; H-12), 1.72 (m, 1H; H-2), 1.77 (m, 1H; H-15), 1.84 (m, 2H; H-7,
[5] R. Thoma, T. Schulz-Gasch, B. D’Arcy, J. Benz, J. Aebi, H. Dehmlow, M.
Hennig, M. Stihle, A. Ruf, Nature 2004, 432, 118–122.
H-21), 1.86 (m, 1H; H-2), 1.887 (s, 3H; CH CO), 2.27 (m, 1H; H-19),
2
1
3
[6] R. Ito, Y. Masukawa, T. Hoshino, FEBS J. 2013, 280, 1267–1280.
.39 (m, 1H; H-19), 2.41 (m, 1H; H-21), 4.81 (dd, J=11.9, 4.5 Hz,
13
[7] a) H. H. Rees, E. I. Mercer, T. W. Goodwin, Biochem. J. 1966, 99, 726–734;
b) H. H. Rees, G. Britton, T. W. Goodwin, Biochem. J. 1968, 106, 659–665;
c) D. H. R. Barton, G. Mellows, D. A. Widdowson, J. J. Wright, J. Chem.
Soc. C 1971, 1142–1148; d) T. Suga, T. Shishibori, Phytochemistry 1975,
14, 2411–2417; e) S. Seo, Y. Tomita, K. Tori, J. Am. Chem. Soc. 1981, 103,
2075–2780; f) S. Seo, Y. Yoshimura, A. Uomori, K. Takeda, H. Seto, Y. Ebi-
zuka, U. Sankawa, J. Am. Chem. Soc. 1988, 110, 1740–1745.
H; H-3), 4.93 (brs, 1H; H-29), 4.94 ppm (brs, 1H; H-29); C NMR
(
(
150 MHz, C D ): d=15.91 (q, 1C; C-26), 16.33 (q, 1C; C-25), 16.33
6 6
q, 1C; C-28), 16.80 (q, 1C; C-24), 18.46 (t, 1C; C-6), 20.82 (q, 1C;
CH CO), 21.09 (t, 1C; C-11), 24.07 (t, 1C; C-2), 24.98 (t, 1C; C-12),
2
1
3
6.81 (t, 1C; C-15), 28.09 (q, 1C; C-23), 30.92 (t, 1C; C-19), 33.98 (s,
C; C-17), 34.11 (t, 2C; C-7, C-21), 36.96 (d, 1C; C-13), 37.16 (s, 1C;
C-10), 37.55; (t, 1C; C-16), 37.98 (s, 1C; C-4), 38.42 (t, 1C; C-1),
[8] a) C. H. Chang, H. Y. Wen, W. S. Shie, C. T. Liu, M. E. Li, Y. T. Liu, W. H. Li,
T. K. Wu, Org. Biomol. Chem. 2013, 11, 4214–4219; b) C. H. Chang, Y. C.
Chen, S. W. Tseng, Y. T. Liu, H. Y. Wen, W. H. Li, C. Y. Huang, C. Y. Ko, T. T.
Wang, T. K. Wu, Biochimie 2012, 94, 2376–2381; c) T. K. Wu, C. H. Chang,
H. Y. Wen, Y. T. Liu, W. H. Li, T. T. Wang, W. S. Shie, Org. Lett. 2010, 12,
4
1
1
1
4
1.04 (s, 1C; C-8), 42.13 (s, 1C; C-14), 43.33 (t, 1C; C-22), 45.35 (d,
C; C-18), 50.21 (d, 1C; C-9), 55.64 (d, 1C; C-5), 80.94 (d, 1C; C-3),
07.8 (t, 1C; C-29), 14.63 (q, 1C; C-27), 149.9 (s, 1C; C-20),
69.9 ppm (s, 1C; CH CO); HRMS (EI): m/z calcd for C H O :
3
31 50
2
500–503.
54.38119; found: 454.38123.
[
9] a) N. Morikubo, Y. Fukuda, K. Ohtake, N. Shinya, D. Kiga, K. Sakamoto, M.
Asanuma, H. Hirota, S. Yokoyama, T. Hoshino, J. Am. Chem. Soc. 2006,
128, 13184–13194; b) T. Hoshino, K. Shimizu, T. Sato, Angew. Chem. Int.
Ed. 2004, 43, 6700–6703; Angew. Chem. 2004, 116, 6868–6871; c) T.
Sato, T. Hoshino, Biosci. Biotechnol. Biochem. 2001, 65, 2233–2242; d) T.
Hoshino, T. Abe, M. Kouda, Chem. Commun. 2000, 441–442; e) T. Hoshi-
no, M. Kouda, T. Abe, S. Ohashi, Biosci. Biotechnol. Biochem. 1999, 63,
2
D
0
1
Product 33 acetate: Oil; [a] = +19.4 (c=0.098, CHCl ); H NMR
3
(
600 MHz, C D ): d=0.78 (m, 1H; H-5), 0.89 (m, 1H; H-1), 0.897 (s,
6 6
3
3
H; Me-19), 0.934 (s, 3H; Me-28), 0.977 (s, 3H; Me-29), 1.053 (s,
H; Me-18), 1.150 (s, 3H; Me-27), 1.17(m, 1H; H-15), 1.186 (s, 3H;
Me-21), 1.25 (m, 1H; H-11), 1.36 (m, 2H; H-7, H-11), 1.39 (m, 1H; H-
9
1
1
2
), 1.47 (m, 1H; H-6), 1.55 (m, 1H; H-11), 1.57 (m, 2H; H-2), 1.60 (m,
H; H-15), 1.61 (m, 1H; H-6), 1.62 (m, 1H; H-7), 1.63 (m, 1H; H-16),
.64 (m, 1H; H-1), 1.66 (m, 2H; H-22), 1.742 (d, J=4.9 Hz, 3H; Me-
6), 1.78 (m, 1H; H-17), 1.79 (m, 1H; H-13), 1.80 (m, 1H; H-16), 2.01
2
038–2041.
10] R. Ito, I. Hashimoto, Y. Masukawa, T. Hoshino, Chem. Eur. J. 2013, 19,
7150–17158.
[
1
[11] R. Ito, Y. Masukawa, C. Nakada, K. Amari, C. Nakano, T. Hoshino, Org.
Biomol. Chem. 2014, 12, 3836–3846.
(
brd, J=11.2 Hz, 1H; H-12), 2.22 (m, 1H; H-23), 2.31 (m, 1H; H-23),
3
.16 (dd, 10.9, 5.4 Hz, 1H; H-3), 5.62 ppm (m, 2H; H-24, H-25);
[12] T. Kushiro, M. Shibuya, K. Masuda, Y. Ebizuka, J. Am. Chem. Soc. 2000,
13
1
22, 6816–6824.
C NMR (150 MHz, C D ): d=15.67 (q, 2C; C-18, C-28), 16.40 (q,
6
6
[
13] a) T. Hoshino, A. Chiba, N. Abe, Chem. Eur. J. 2012, 18, 13108–13116;
b) T. Hoshino, Y. Sakai, Tetrahedron Lett. 2001, 42, 7319–7323; c) T.
Hoshino, Y. Sakai, Chem. Commun. 1998, 1591–1592; d) T. Hoshino, E.
Ishibashi, K. Kaneko, J. Chem. Soc. Chem. Commun. 1995, 2401–2402;
e) E. J. Corey, S. C. Virgil, J. Am. Chem. Soc. 1991, 113, 4025–4026; f) E. J.
Corey, S. C. Virgil, S. Sarshar, J. Am. Chem. Soc. 1991, 113, 8171–8172;
g) E. E. van Tamelen, R. P. Hanzlik, K. B. Sharpless, R. B. Clayton, W. J.
Richter, A. L. Burlingame, J. Am. Chem. Soc. 1968, 90, 3284–3286.
1
6
C; C-19), 16.69 (q, 1C; C-29), 18.08 (q, 1C; C-26), 18.64 (t. 1C; C-
), 21.83 (t, 1C; C-11), 25.15 (t, 1C; C-16), 25.79 (q, 1C; C-21), 27.52
(
t, 1C; C-23), 27.93 (t, 2C; C-2, C-12), 28.27 (q, 1C; C-27), 31.56 (t,
1
3
1
9
C; C-15), 35.66 (t, 1C; C-7), 36.31 (t, 1C; C-1), 37.33 (s, 1C; C-10),
9.20 (s, 1C; C-4), 40.64 (s, 1C; C-8), 40.69 (t, 1C; C-22), 42.51 (t,
C; C-13), 50.46 (d, 1C; C-17), 50.50 (s, 1C; C-14), 51.08 (d, 1C; C-
), 56.32 (d, 1C; C-5), 74.67 (s, 1C; C-20), 78.50 (d, 1C; C-3), 124.8
(
d, 1C; C-25), 132.3 ppm (d, 1C; C-24) (the assignments of the C-24
[14] a) Y. Yonemura, T. Ohyama, T. Hoshino, Org. Biomol. Chem. 2012, 10,
4
2
40–446; b) T. Hoshino, Y. Kumai, T. Sato, Chem. Eur. J. 2009, 15, 2091–
100; c) T. Abe, T. Hoshino, Org. Biomol. Chem. 2005, 3, 3127–3139;
and C-25 atoms may be exchangeable); HRMS (EI): m/z calcd for
C H O ÀH O: 412.37.51; found: 412.36957.
29
50
2
2
d) T. Hoshino, S. Nakano, T. Kondo, T. Sato, A. Miyoshi, Org. Biomol.
Chem. 2004, 2, 1456–1470; e) T. Hoshino, Y. Kumai, I. Kudo, S. Nakano,
S. Ohashi, Org. Biomol. Chem. 2004, 2, 2650–2657; f) S. Nakano, S.
Ohashi, T. Hoshino, Org. Biomol. Chem. 2004, 2, 2012–2022; g) T. Hoshi-
no, S. Ohashi, Org. Lett. 2002, 4, 2553–2556; h) T. Hoshino, T. Kondo,
Chem. Commun. 1999, 731–732.
Acknowledgements
This work was supported in part by Grant-in-Aid for Scientific
Researchfrom the Japan Society for the Promotion of Science(-
Nos. 25450150and 18380001).
[
15] a) E. J. Corey, S. K. Gross, J. Am. Chem. Soc. 1968, 90, 5045–5046; b) I.
Abe, Y. Sakano, H. Tanaka, W. Lou, H. Noguchi, M. Shibuya, Y. Ebizuka, J.
Am. Chem. Soc. 2004, 126, 3426–3427; c) I. Abe, Y. Sakano, M. Sodeya-
ma, H. Tanaka, H. Noguchi, M. Shibuya, Y. Ebizuka, J. Am. Chem. Soc.
2
004, 126, 6880–6881; d) H. Noma, H. Tanaka, H. Noguchi, M. Shibuya,
Keywords: alkenes
polycycles · terpenoids
·
cyclization
·
enzyme catalysis
·
Y. Ebizuka, I. Abe, Tetrahedron Lett. 2004, 45, 8299–8301; e) T. Hoshino,
Y. Yamaguchi, K. Takahashi, R. Ito, Org. Lett. 2014, 16, 3548–3551.
16] a) S. A. Knight, Org. Magn. Reson 1974, 6, 603–611; b) S. Seo, Y. Tomita,
K. Tori, Tetrahedron Lett. 1975, 16, 7–10; c) J. W. Blunt, M. H. G. Munro,
Org. Magn. Reson. 1980, 13, 26–27.
17] a) D. R. Phillips, J. M. Rasbery, B. Bartel, S. P. T. Matsuda, Curr. Opin. Plant
Biol. 2006, 9, 305–314; b) S. Lodeiro, Q. Xiong, W. K. Wilson, M. D. Koles-
nikova, C. S. Onak, S. P. T. Matsuda, J. Am. Chem. Soc. 2007, 129, 11213–
[
[
[
[
1] A. Eschenmoser, L. Ruzicka, O. Jeger, D. Arigoni, Helv. Chim. Acta 1955,
8, 1890–1940.
2] Reviews: a) T. Hoshino, T. Sato, Chem. Commun. 2002, 291–301; b) K. U.
Wendt, G. E. Schulz, E. J. Corey, D. R. Liu, Angew. Chem. Int. Ed. 2000, 39,
3
2
812–2833; Angew. Chem. 2000, 112, 2930–2952; c) R. A. Yoder, J. N.
11222; c) R. Ito, K. Mori, I. Hashimoto, C. Nakano, T. Hoshino, Org. Lett.
Johnston, Chem. Rev. 2005, 105, 4730–4756; d) T. K. Wu, C. H. Chang,
Y. T. Liu, T. T. Wang, Chem. Rec. 2008, 8, 302–305; e) I. Abe, Nat. Prod.
Rep. 2007, 24, 1311–1331; f) W. D. Nes, Chem. Rev. 2011, 111, 6423–
2011, 13, 2678–2681.
[
[
18] I. Abe, M. Rohmer, J. Chem. Soc. Perkin Trans. 1 1994, 783–791.
19] a) D. J. Reinert, G. Balliano, G. E. Schulz, Chem. Biol. 2004, 11, 121–126;
b) K. U. Wendt, A. Lenhart, G. E. Schulz, J. Mol. Biol. 1999, 286, 175–187.
6
451.
[
[
3] R. Xu, G. C. Fazio, S. P. T. Matsuda, Phytochemistry 2004, 65, 261–291.
4] A. Ruf, F. Müller, B. D’Arcy, M. Stihle, E. Kusznir, C. Handschin, O. H.
Morand, R. Thoma, Biochem. Biophys. Res. Commun. 2004, 315, 247–
Received: June 20, 2015
2
54.
Published online on September 9, 2015
Chem. Eur. J. 2015, 21, 15769 – 15784
www.chemeurj.org
15784
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim