Alicyclobacillus acidocaldarius Squalene-Hopene Cyclase: The Critical Role of Steric Bulk at Ala306 and the First Enzymatic Synthesis of Epoxydammarane from 2,3-Oxidosqualene

Article abstract of DOI:10.1002/cbic.201800281

The acyclic molecule squalene (1) is cyclized into 6,6,6,6,5-fused pentacyclic hopene (2) and hopanol (3; ca. 5:1) through the action of Alicyclobacillus acidocaldarius squalene-hopene cyclase (AaSHC). The polycyclization reaction proceeds with regio- and stereochemical specificity under precise enzymatic control. This pentacyclic hopane skeleton is generated by folding 1 into an all-chair conformation. The Ala306 residue in AaSHC is conserved in known squalene-hopene cyclases (SHCs); however, increasing the steric bulk (A306T and A306V) led to the accumulation of 6,6,6,5-fused tetracyclic scaffolds possessing 20R stereochemistry in high yield (94 % for A306V). The production of the 20R configuration indicated that 1 had been folded in a chair-chair-chair-boat conformation; in contrast, the normal chair-chair-chair-chair conformation affords the tetracycle with 20S stereochemistry, but the yield produced by the A306V mutant was very low (6 %). Consequently, bulk at position 306 significantly affects the stereochemical fate during the polycyclization reaction. The SHC also accepts (3R) and (3S)-2,3-oxidosqualenes (OXSQs) to generate 3α,β-hydroxyhopenes and 3α,β-hydroxyhopanols through polycyclization initiated at the epoxide ring. However, the Val and Thr mutants generated epoxydammarane scaffolds from (3R)-OXSQ; this indicated that the polycyclization cascade started in these instances at the terminal double bond position. This work is the first to report the polycyclization of oxidosqualene starting at the terminal double bond.

Full text of DOI:10.1002/cbic.201800281

Accepted Article
Title: Alicyclobacillus acidocaldarius squalene-hopene cyclase: the
critical role of the steric bulk at Ala306 and the first enzymatic
synthesis of epoxydammarane from 2,3-oxidosqualene
Authors: Natsumi Ideno, Shikou Umeyama, Takashi Watanabe, Mami
Nakajima, Tsutomu Sato, and Tsutomu Hoshino
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemBioChem 10.1002/cbic.201800281
Link to VoR: http://dx.doi.org/10.1002/cbic.201800281
A Journal of
www.chembiochem.org

10.1002/cbic.201800281
ChemBioChem
1
2
Alicyclobacillus acidocaldarius squalene-hopene cyclase: the critical role of steric
bulk at Ala306 and the first enzymatic synthesis of epoxydammarane from 2,3-
3
oxidosqualene
4
5
Natsumi Ideno, Shikou Umeyama, Takashi Watanabe, Mami Nakajima, Tsutomu Satoand
Tsutomu Hoshino*
6
7
8
Graduate School of Science and Technology and Department of Applied Biological
Chemistry, Faculty of Agriculture, Niigata University, Ikarashi 2-8050, Nishi-ku, Niigata
950-2181, Japan
9
10
11
12
13
14
15
16
17
Tsutomu Hoshino: Prof. Dr.
*Corresponding author
E-mail: hoshitsu@agr.niigata-u.ac.jp
1
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
Abstract
The acyclic molecule squalene (1) is cyclized into a 6,6,6,6,5-fused pentacyclic hopene
3
(2) and hopanol (3) (ca. 5:1) by Alicyclobacillus acidocaldarius squalene-hopene
cyclase (AaSHC). The polycyclization reaction proceeds with regio- and stereochemical
specificity under precise enzymatic control. This pentacyclic hopane skeleton is
generated by folding 1 into an all-chair conformation. The Ala306 in AaSHC is
conserved in known SHCs; however, increases in steric bulk (A306T and A306V) led to
the accumulation of 6,6,6,5-fused tetracyclic scaffolds possessing 20R stereochemistry
in high yield (94% for A306V). The production of the 20R configuration indicated that
1 had been folded in a chair-chair-chair-boat conformation; in contrast, the normal
chair-chair-chair-chair conformation affords the tetracycle with 20S stereochemistry, but
the yield produced by the A306V mutant was very low (6%). Consequently, bulk at
position 306 significantly affects the stereochemical fate during the polycyclization
reaction. SHC also accepts (3R) and (3S)-2,3-oxidosqualenes (OXSQs) to generate 3,
-hydroxyhopenes and 3, -hydroxyhopanols through a polycyclization reaction
initiated at the epoxide ring. However, the Val and Thr mutants generated
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
epoxydammarane scaffolds from (3R)-OXSQ, which indicated that the polycyclization
cascade started at the terminal double bond position. This work is the first to report the
polycyclization of oxidosqualene starting from the terminal double bond.
21 Introduction
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
The cyclization mechanisms by which acyclic squalene (1) and (3S)-2,3-
oxidosqualene form polycyclic triterpenes have attracted much attention from chemists
and biochemists for over a half a century.[1,2] Alicyclobacillus acidocaldarius squalene-
hopene cyclase (AaSHC) catalyzes the conversion of 1 into hopene (2) and hopanol (3)
(ca 5:1), pentacyclic triterpenes consisting of 6,6,6,6,5-fused A/B/C/D/E-ring systems,
as shown in Scheme 1A.[2a,b] The X-ray crystallographic structure of the SHC from A.
acidocaldarius (AaSHC) was first reported in 1997[3] and further refined later.[4]
Compound 1 is folded into an all-prechair conformation inside the reaction cavity and is
cyclized in a regio- and stereospecific fashion via a series of carbocationic
intermediates, leading to the formation of 5 new C-C bonds and 9 stereocenters. This
reaction consists of 8 reaction steps[2b] (Scheme 1B): (1) the acyclic molecule 1 is folded
into a chair conformation to give the intermediary monocyclic cation 4; (2) a second
ring formation reaction occurs to afford cation 5, consisting of a 6,6-fused bicyclic
cation; (3) a third ring formation reaction produces the 6,6,5-fused tricyclic cation 6; (4)
cation 6 undergoes a ring expansion to generate the secondary cation 7 with a 6,6,6-
2
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
fused A/B/C-ring system; (5) a further ring closure gives the tertiary cation 8 with a
6,6,6,5-fused tetracycle (17-epi-dammarenyl cation); (6) cation 8 undergoes a second
3
ring expansion reaction to provide the secondary cation 9 (prohopanyl cation); (7) the
final ring formation reaction furnishes the 6,6,6,6,5-fused pentacyclic cation 10
(hopanyl cation); and (8) proton elimination from Me-29 introduces a double bond to
produce 2, and a water molecule attacks the C-22 cation of 10 to give 3. Contrary to this
step-wise mechanism, there are reports that the polycyclization reactions proceed in a
concerted manner.[5a, b] (3S)-2,3-Oxidosqualene (11) is also accepted as a substrate by
SHC, yielding 3-hydroxyhopene (12) and 3-hydroxyhopanol (13) through an all-chair
conformation in a similar cyclization pathway to that of 1 (Scheme 1C). (3R)-2,3-
Oxidosqualene (14) is also converted into 3-hydroxyhopene (15) and 3-
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
hydroxyhopanol (16) through a boat-chair-chair-chair-chair conformation.[6a-d]
<Scheme 1>
To identify the active site residues of AaSHC and their functions, site-specific
mutagenesis experiments were conducted. The central Asp residue (Asp376) in the
D374XD376D377 motif functions as the proton donor to initiate the polycyclization
reaction.[7a,b] His451 hydrogen bonds with Asp376 to enhance the motif acidity.[2a,6a]
Asp377 likely stabilizes cation 4.[7a] Trp312 likely binds with the terminal methyl group
via a hydrophobic interaction (possibly CH- interaction) to place 1 in close proximity
to the DXDD motif, thus facilitating the initiation of the polycyclization reaction.[8]
Phe365 stabilizes cation 5.[9] Phe601 likely stabilizes cations 6 and 7,[10] and Phe605
stabilizes the intermediary cations 9 and 10.[11] All the Phe residues mentioned above
function to stabilize transient cations via cation/ interactions.[12] The main function of
Y420 is to stabilize the bicyclic cation 5 by pointing towards the active site cavity and
providing additional negative electrostatic potential.[2a,4,13] The steric bulk at the Leu607
position is critical for the correct chair-folding around the B-ring formation site.[13] The
steric bulk of another aliphatic residue, Ile261, is also crucial for enforcing the correct
folding around the D/E-ring formation sites.[14] The I261A variant generates cations 6
and 7 with 13-H and 17-H configurations, respectively, which are opposite to the
configurations of the genuine intermediary cations 6 (13-H) and 7 (17-H); thus, the
steric bulk at this position directs the stereochemical fate. There have been many recent
reports on the use of SHC mutants as biocatalysts, by which many known and unknown
compounds have been successfully created.[15]
Herein, we report the function of the Ala306 residue in AaSHC, which is widely
conserved in known SHCs (Fig. S1, Supporting Information). The X-ray crystal
3
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
structure of AaSHC complexed with a modeled hopene molecule indicates that Ala306
is located near the squalene molecule (Fig. S2).[3,4] To investigate the function of Ala306
3
of AaSHC in hopene biosynthesis, we constructed site-directed mutants in which the
steric bulk at this position is altered, incubated 1, 11, and 14 with these variants, and
determined the structures of the resulting enzymatic products. We provide clear
evidence that the Ala306 residue is located in close proximity to the D-ring formation
site and that the steric bulk at this position is a key determinant of the stereochemical
fate during the polyolefin cyclization reaction, similar to the Ile216 residue.
Furthermore, we report the first case in which the polycyclization reaction of
oxidosqualene was initiated from the terminal double bond instead of the terminal
epoxide ring, resulting in the production of a 17-epi-epoxydammarane scaffold.
4
5
6
7
8
9
10
11
12
13 Results
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Isolation of products generated from the incubation of 1 with A306X mutants
Fig. 1A shows the GC profile obtained by incubating 1 with the cell-free
homogenates of the A306V variant. Squalene 1 (200 mg) was incubated (pH 6.0, 45°C,
16 h) with the cell-free extracts (300 mL) obtained from a 6 L culture of the A306V
variant. The incubation mixture was freeze-dried, and the lipophilic materials, including
the enzymatic products, were extracted with hexane. These were then subjected to SiO2
column chromatography by eluting with 100% hexane to remove the Triton X-100
present in the incubation mixture and to partially separate the products into several
fractions. Only a small amount of 1 was recovered, indicating that the reaction by the
variant was almost complete. The purity of the partially separated fractions was
monitored by GC-MS. The fraction containing a mixture of 17, 19, and 20 was obtained
by partial fractionation with 100% hexane as the eluent. Complete separation of the
products was attained by column chromatography with activated SiO2 (heated in an
oven at 180°C for 2 h), yielding 2.2 mg of 17, 1.0 mg of 19, and 1.5 mg of 20 in a pure
state. However, separation of the other products was more difficult. The fraction
containing products 18, 21, 22, 23, and 24 was chromatographed on a 5% AgNO3-
impregnated SiO2 column, and pure 23 was obtained in a yield of 1.4 mg by eluting
with hexane/EtOAc (100:0.2), but the separation of 18 and 21 failed. This separation
was achieved by C18 reverse-phase HPLC (THF/H2O=60:40), which yielded 0.8 mg of
18 and 1.3 mg of 21 in a pure state. The fraction obtained by eluting with hexane/EtOAc
(100:1) contained a mixture of 21, 22, 23, and 24. Pure 24 (2.1 mg) was obtained by
reverse-phase HPLC (THF/H2O=70:30), but the separation of 21, 22, and 23 was not
successful. The fractions containing 18, 21, 22, and 23, which were not separable by 5%
4
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
argentation chromatography, were combined and subjected to 10%-AgNO3 SiO2 column
chromatography, and eluted with hexane/EtOAc (100:0.1) to yield a mixture of 18, 21,
3
and 22, and then C18-HPLC (THF/H2O=60:40) successfully gave pure 22 in a yield of
0.6 mg. The isolation yields of 18, 21, 22, and 23 were very low, despite the large peaks
detected by GC (Fig. 1A). The low yields could be ascribed to both the separation
difficulty and the use of an AgNO3-SiO2 column (as the recovery from AgNO3
chromatography is generally low). In addition, there were many remaining fractions that
have yet to be purified. Fig. S3.1 depicts the GC trace from the incubation of 1 with the
A306T variant. Products 25–27, which were not detected as products of the Val mutant,
were newly revealed in the GC traces, and 28 was identified as a new product of the
A306F mutant (Fig. S3.2). Products 25–28 were all identified by comparing their EIMS
fragment patterns and retention times to those of previously isolated authenticated
compounds.[2b]
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
<Figure 1>
Isolation of products generated from the incubation of 11 and 14 with A306X
mutants
Fig. 1B shows the GC chromatogram of the reaction mixture obtained by
incubating a mixture of 11 and 14 with the A306V mutant. To isolate the enzymatic
products 29–33, a racemic mixture of 11 and 14 (100 mg) was incubated (pH 6.0, 45°C,
20 h) with the cell-free extract (300 mL) from a 6 L culture of the A306V mutant. After
lyophilization of the reaction mixture, the reaction products were extracted with hexane,
which was passed through a short SiO2 column by elution with hexane/EtOAc (100:20)
to remove Triton X-100. The residues obtained after removing Triton X-100 were
subjected to SiO2 column chromatography in a stepwise gradient fashion
(hexane/EtOAc=100:1 to 100:2). Product 33 was isolated in a pure state (6.0 mg), but
the other products could not be separated. The inseparable fractions were divided into
the following 2 fractions: a mixture (22.4 mg) of 29, 31, and 33, and a mixture (14.7
mg) of 30 and 32. Products 29 and 31 were successfully separated by normal-phase
HPLC (hexane/2-propanol=100:0.1) to yield 29 (2.3 mg) and 31 (1.0 mg) in a pure
state. Products 30 and 32 were separated by normal-phase HPLC
(hexane/THF=100:1.0) with yields of 2.0 mg and 2.3 mg, respectively. The A306F
variant generated products 34, 35, and 36 from a mixture of 11 and 14, as shown in Fig.
S4. Fifty mg of a mixture of 11 and 14 was incubated with the A306F mutant. SiO2
column chromatography (hexane/EtOAC=100:1 to 100:5) afforded pure 36 (14.5 mg)
and a mixture (6.0 mg) of 34 and 35, which were then acetylated with Ac2O/Py (1:1).
5
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
Normal-phase HPLC (hexane:2-PrOH=100:0.02) successfully isolated the pure acetates
of 34 and 35 in yields of 0.6 mg and 0.8 mg, respectively. The isolation yields are not
3
consistent with the production yields because some fractions still contained inseparable
products.
4
5
6
Structural determination of products 17–36.
7
Fig. 2 lists the structures of enzymatic products that were identified. Products 17–
28, with the exception of 23, were previously isolated from the site-specific variants of
AaSHC.[2b]. Detailed NMR analyses of 23, a new product generated from AaSHC-
mutants, are described below. The EIMS and NMR spectra of 17–22 and 24 are shown
in Figs. S5–S10, and brief descriptions for proposing the structures are provided in the
Supporting Information. We isolated 17 from the L607F,[13] Y420A,[13], and Y609F
variants.[8] Compound 18 is identical to the product isolated from variants I261A[14] and
F605A.[11] Product 19 was isolated from mutants Y609A,[8] Y612A[8], and F365A.[9]
Product 20 is identical to the compound generated by the F601A mutant.[10] Compound
21 was isolated from the W489 variant.[2b, 16, 17] Product 22 was previously isolated from
the F605A mutant.[11] Product 24 was isolated from the W169F, W169H, and W489F
variants.[17] The EIMS spectra of minor products 25–28 are shown in Fig. S13. The
spectra and GC retention times of 25–28 were the same as those of samples we isolated
previously: 25 and 27 from F605A,[11] 26 from P263G, [19], and 28 from F365A.[9]
Accordingly, these structures are illustrated in Fig. 2.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
The detailed NMR analyses of product 23, including 2D NMR spectra (600 MHz,
C6D6, Fig. S11.9), revealed that 23 had the same tetracyclic skeleton as 22. Furthermore,
22 and 23 had identical EIMS spectra (Figs. S10.1 and S11.1), but their GCMS
retention times were different (Fig. 1A), indicating that the 2 compounds are
stereoisomers. The relative configurations of 23, which is present in the A/B/C/D-fused
tetracyclic scaffolds, were determined to be identical to those of 22 by detailed analyses
of the NOESY spectrum (Fig. S11.6). Thus, the C-20 stereochemistry of 23 is opposite
to that of 22 and can be assigned as the 20S configuration for 23. Consequently, 23 is
20S-tirucalla-7(8),24-diene, as shown in Fig. 2. The H of Me-21 of 20S-tirucalla-
7(8),24-diene in CDCl3 was reported to be H 0.895 (d, J=6.8 Hz), [18] which was close
to that of 23 (H 0.885, J=6.8 Hz) in CDCl3 solution. Thus, the 20S stereochemistry of
23 is credibly assigned.
Next, the structural analyses of the enzymatic products 29–36, which were
generated by incubating a mixture of 11 and 14, are described. We have not reported the
enzymatic reactions of 11 and 14 by the AaSHC variants; all the products are new SHC-
6
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
metabolites, and the NMR analyses are described below.
The 1H-NMR spectrum of 29 (400 MHz, C6D6, Fig. S14.2) showed the presence of
3
an isopropylidene residue: H 1.83 (s, 3H, Me-26) and 1.75 (s, 3H, Me-27). The 13C
NMR spectrum (100 MHz, C6D6, Fig. S14.4) revealed the presence of a tetrasubstituted
double bond (C 139.7, s; 134.9, s). Detailed HMBC analyses (Fig. S14.10) indicated
that 29 consists of a 6,6,6,5-fused tetracyclic system and that the double bonds of 29
exist at C-13–C-17 and C-24–C25. The multiplicity of H-3 (H 3.29) showed a broad
singlet, indicating that the hydroxyl group is -oriented. The doublet Me was located at
C-20 because Me-21 (H 1.14, d, J=7.2 Hz, 3H) showed a strong HMBC cross peak for
C-17. The chemical shift of Me-21 (H 0.918 d, J=6.8 Hz) in the CDCl3 solution (see
Fig. S14.3) was very close to the reported value (H 0.910 d, J=6.9 Hz) for 20R-
dammara-13(17),24-diene.[18] Thus, 29 was determined to be 3-hydroxy-20R-
dammara-13(17)-24-diene, as shown in Fig. 2.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
The EIMS spectrum of 30 (Fig. S15.1) is almost identical to that of 29 (Fig. S14.1),
indicating that 30 is a stereoisomer of 29. Detailed NMR analyses (Fig. S15.9) showed
that dammara-13(17),24-diene can be assigned to 30 as described in the structural
determination of 29. The C-20 stereochemistry of 30 was the same as that of 29 because
the Me-21 of 30 showed a H 0.919 (d, J=6.8 Hz, 3H) identical to that of 29 in CDCl3
solution (cf. in the C6D6 solution, H 1.14 for Me-21 of 29;H 1.15 for Me-21 for 30;
both compounds had the same chemical shifts in C6D6). The OH was determined to be
-oriented because H-3 (H 3.16) gave a dd splitting pattern (dd, J=10.2, 6.1 Hz). The
entire structure of 30 can be deemed 3-hydroxy-20R-dammara-13(17),24-diene, as
shown in Fig. 2.
The 1H-NMR (400 MHz, C6D6) and 13C-NMR (100 MHz, C6D6) spectra of 31 (Fig.
S16.3 and S16.4) revealed the presence of 2 double bonds: H 5.41 (2H: m, 1H, H-12;
m, 1H, H-24) and C 125.5 (d, C-24), 130.9 (s, C-25), 115.9 (d, C-12) and 149.2 (s, C-
13). Two Me groups (H 1.88, s, 3H, Me-26; 1.73, s, 3H, Me-27) had clear HMBC cross
peaks with C-25 and C-24, indicating that the isopropylidene moiety remained without
participation of the cyclization reaction, and thus, 31 is a tetracyclic product. Me-30 (H
1.13, s, 3H) showed a distinct HMBC correlation with C-13; Me-21 (H 1.05, d, J=6.8
Hz, 3H) had a clear HMBC contour with C-17 (C 48.94, d); and the multiplicity of C-
17 (d) indicated the presence of a proton at C-17. Thus, another double bond is
positioned at C-12–C-13. Indeed, the chemical shifts of H-11 [H 1.92 (m) and 2.16 (m)]
clearly indicated that H-11 is located at the allylic position. The OH group was -
oriented because the multiplicity of H-3 (H 3.29) showed a broad singlet. Detailed
NMR analyses (Fig. S16.10) indicated that 31 possesses 3-hydroxy-17-epi-dammara-
7
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
12(13),24-diene with R stereochemistry at C-20 because the chemical shift of Me-21 in
CDCl3 solution was H 0.789 (d, J=6.4 Hz, see Fig. S16.2.2), which was very close to
the reported value (H 0.784, d, J=6.8 Hz),[18] and to that of Me-21 (H 0.793, d, J=6.8
3
4
Hz, measured in CDCl3, see Fig. S9.9) in product 21.
5
Detailed NMR analyses (Fig. S17.9) indicated that product 32 also has a 17-epi-
dammara-12(13),24-diene scaffold, which was further confirmed by the fact that the
EIMS spectra of 31 (Fig. S16.1) and 32 (Fig. S17.1) were identical. The OH function
was oriented at the -position, as the multiplicity of H-3 (H 3.14) was a double doublet
(J=11.4, 4.7 Hz). Furthermore, the chemical shift of Me-21 (H 1.06, d, J=6.7 Hz, Fig.
S17.2.2) of 32 in C6D6 solution was the same as that (H 1.05, d, J=6.8 Hz, Fig. S16.3.2)
of 31 (cf. H 0.784 for 32 and H 0.789 for 31, in CDCl3 solution, Fig. S17.2.1). Thus,
32 was proposed to be 3-hydroxy-17-epi-20R-dammara-12(13),24-diene.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
No olefinic protons and no double bonds were found in the 1H-NMR (600 MHz,
acetone d6, Fig. S18.2) and 13C-NMR spectra (150 MHz, acetone d6, Fig. S18.3) of
product 33, indicating that the full cyclization reactions occurred. Three carbons
connected to the oxygen atom were found: C 86.06 (s, C-20), 88.23 (d, C-24) and 70.43
(s, C-25). Three Me groups had distinct HMBC correlations with these 3 carbons: Me-
21 (H 1.19, s, 3H)/C-20 and both Me-26 (H 1.13, s, 3H) and Me-27 (H 1.05, s, 3H)/C-
24/C-25, where the chemical shift assignments of Me-26 and Me-27 are exchangeable.
Me-21 exhibited further HMBC contours with C-22 (C 38.70, t) and C-17 (C 49.30,
d), and H-24 (H 3.71, dd, J=10.6, 5.2 Hz) was spin-coupled with H-23 (H 1.75, m, 1H;
1.85, m, 1H). Thus, the moiety 2-(tetrahydrofuran-2-yl)propan-2-ol is connected to C-
17. This THF moiety was further confirmed by EIMS, in which the fragment ion m/z
143 was observed as the base peak (Fig. 18.1). More detailed HMBC analyses revealed
the presence of a 6,6,6,5-fused tetracyclic ring, in which H-17 is positioned in a -
orientation because H-17 had a clear NOE with H-13 (H 2.09, m, 1H). Further
analyses, including 2D NMR (Fig. S18.9), indicated that product 33 is 20,24-
epoxydammarane-25-ol with 17-H (17-epi-epoxydammarane skeleton). No NOE was
found between Me-21 and H-24, indicating that either (20R, 24R) or (20S, 24S) can be
assigned for the THF ring. We have reported that epoxydammarane scaffolds are created
by the SHC-mediated reactions of non-natural (3R)- and (3S)-2,3-squalene diols and
2,3:22,23-dioxidosqualenes,[20] but this work is the first to report the production of the
epoxydammarane from oxidosqualene. The 1H- and 13C-NMR data of 33 (Fig. S18.9)
were indistinguishable from those of (20R, 24R)-17-epi-epoxydammarane (Fig.
S18.10), which was obtained from the reaction of (3R)-2,3-squalene diol with native
SHC.[20] (20S, 24S)-17-epi-Epoxydammarane and (20R, 24R)-17-epi-epoxydammarane
8
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
are diastereomers, thus the NMR spectra of the 2 diastereomers must be different, which
allowed us to propose (20R, 24R)-stereochemistry for 33.
3
Product 34-acetate showed peaks for 5 vinylic Me groups: H 1.69 (s, 3H, Me-29);
H 1.81 (s, 3H, Me-30); H 1.72 (s, 6H, Me-26 and Me-28); and H 1.78 (s, 3H, Me-27).
One tetrasubstituted double bond (C 126.2, s, C-8 and C 140.4, s, C-9) was found.
Furthermore, 3 aliphatic Me signals were observed: H 1.05 (s, 3H, Me-25), H 0.892 (s,
3H, Me-24), and H 1.05 (s, 3H, Me-23). These findings are suggestive of a bicyclic
ring system. Me-25 and Me-26 had distinct HMBC cross peaks with C-9 and C-8,
respectively, thus the tetrasubstituted double bond is arranged at C-8 and C-9. The H-3
(H 5.03) showed a broad singlet, indicating -oriented OH. Consequently, the structure
of 35 is polypoda-8(9),13,17,21-tetraen-3-ol, as shown in Fig. 2.
4
5
6
7
8
9
10
11
12
13
14
15
16
Detailed NMR analyses of 35-acetate (Fig. S20.9) and 36 (Fig. S21.9) revealed that
the structures of 35 and 36 are polypoda-7(8),13,17,21-tetraen-3-ol and polypoda-
7(8),13,17,21-tetraen-3-ol, respectively.
<Figure 2>
17 Discussion
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Mechanistic insights into the production of 17-36 from substrates 1, 11, and 14.
Scheme 2 shows the formation mechanisms of 6,6-fused bicyclic products.
Substrates 1 and 11 were folded into a chair-chair conformation to give 6,6-fused
bicyclic cations 5 and 37, respectively. Substrate 14 was folded into a boat-chair
conformation to afford the same bicyclic skeleton 38, but the OH-orientation (-
arrangement) from 14 was opposite to that (-orientation) from 11. This finding is
consistent with the cyclization mechanisms of 1, 11, and 14 (Scheme 1). Proton
elimination from Me-26 gave 17 (path a). The deprotonation of 7-H produced 19, 35,
and 36 (path b). The elimination of H-9 generated 34 (path d), which was obtained only
from substrate 14. Product 28 was produced by the following successive antiparallel
1,2-shifts: H-9C-8 cation, Me-25C-9, H-5C-10, and the deprotonation of 6-H to
give the double bond at C-5 and C-6 (path c).
Scheme 3A shows the formation mechanism of tricyclic 20. This compound was
produced only from 1. Substrate 1 was folded into a chair-chair-boat conformation (39)
to furnish the 6,6,5-fused A/B/C-ring system with 13-H (40). On the other hand, the
chair-chair-chair conformation (41) gives the 6,6,5-fused tricycle with 13-H (42),
which is identical to cation 7. Deprotonation from Me-27 of cation 40 afforded product
20 (path a).
Scheme 3B depicts the cyclization mechanisms leading to the pentacyclic products
9
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
25 and 26, the skeletons of which were produced only from 1 and not from 11 and 14.
Squalene 1 was folded into an all-chair conformation to afford the final hopanyl cation
3
10. The following successive rearrangement reactions of 10 with the C-22 cation
afforded 25 (path b): H-21 to C-22 cation, H-17 to C-21, Me-28 to C-17, followed by
deprotonation of H-13. H-21 was eliminated to give 26 (path c).
4
5
6
Scheme 3C illustrates the formation mechanism of product 33, which is produced
by the A306V and the A306T mutants (Fig. 1B and Fig. S4). The (20R, 24R)
stereochemistry of 33 indicates that this compound was produced from (3R)-
oxidosqualene 14, but not from 11. Substrate 14 was folded into an all-chair
conformation. Proton attack on the terminal double bond, but not on the epoxide ring,
triggered successive ring formation reactions to yield the 17-epi-6,6,6,5-fused
dammarenyl cation 43. The epoxide ring of 43 attacked the C-20 cation (Re-face
attack), and then the nucleophilic attack of a water molecule on the C-25 cation afforded
33. Intriguingly, the cyclization reaction started from the terminal double bond position
and not from the epoxide ring. This study is the first to report a polycyclization reaction
that initiated from the terminal double bond, despite the epoxide ring being present in
the substrate. As shown in Scheme 1B, the SHC-mediated cyclization reaction starts
from the epoxide ring side but not from the terminal double bond side. This
phenomenon has been clearly explained in terms of the higher nucleophilicity of
epoxide than that of double bond -electrons towards the DXDD motif, a proton donor
that can initiate the polycyclization reaction. [7] However, why epoxydammarane
compounds such as 33 are never produced from 11 and 14 by the native SHC, even in
small amounts, is unclear. An increase in the steric bulk at position 306 (A306V and
A306T variants) may have altered the geometry of the substrate inside the reaction
cavity, and the terminal double bond may have been positioned nearer to the DXDD
motif than the epoxide ring. In the case of the wild type SHC, the epoxide ring may be
in closer contact with the DXDD motif, and the more highly nucleophilic nature of the
epoxide ring may further facilitate the polycyclization reaction. We have reported that
native SHC can accept both (3R)- and (3S)-2,3-squalene diols to afford 24R-and 24S-
epoxydammarane, respectively. [20] Furthermore, the four different 2,3;22,23-
dioxidosqualenes (3R,S and 22R,S isomers) are also converted into four 3-hydroxy-
epoxydammarane scaffolds with four different configurations (3R, 24R; 3R, 24S; 3S,
24R; 3S, 24S); [20] that is, native SHC can accept both 22R- and 22S-isomers as
substrates, which correspond to (3R)- and (3S)-isomers in the case of the
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
polycyclization reactions of 2,3-oxidosqualene, respectively, resulting in the production
of both 24R- and 24S-epoxydammaranes. This fact that native SHC can accept both
10
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
(3R)-14 and (3S)-11 raises the question why (3S)-11 was not cyclized by the A306V and
A306T variants to produce 24S-epoxydammarane, while both (3R)-14 and (3S)-11
3
underwent cyclization reactions from the epoxide ring to generate 3- and 3-hydroxy-
tetracycle 29–32 and bicycle 34–36. It is also unclear why the mutated SHCs cannot
accept the monocyclic and bicyclic products of the premature cyclization step to yield
further cyclized products through reentry of the mon- and/or bicyclic product into the
reaction cavity after exiting. The substrate uptake and the product release occur through
the same channel,[3,4] which may be very compact; thus, the truncated cyclization
products cannot reenter the reaction cavity. In contrast, tetraprenyl--curcumene cyclase
from Bacillus megaterium (BmeTC) can accept 3-deoxyachilleol A (monocyclic
triterpene) to yield (+)-ambrein as a result of the re-entry of this monocyclic product
into the reaction cavity.[21] Further studies are necessary to provide detailed insight into
the cyclization mechanism underlying how 33 is produced from 14.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
<Scheme 2>
Scheme 4 shows the formation mechanisms of tetracyclic products from 1, 11, and
14. Two types of formation mechanisms are proposed: (A) the chair-chair-chair-boat
folding conformation 44 from 1, leading to the tetracycle 45 with 17-H; (B) the chair-
chair-chair-chair folding conformation 48 from 1, resulting in the tetracycle 49 with
17-H. Substrates 11 and 14 were also cyclized according to the folding conformation
of Type A, yielding the 3-hydroxytetracyclic cation 46 and 3-hydroxytetracyclic
cation 47, respectively. By contrast, Type B is limited to the cyclization reactions of 1.
The Newman projection of cations 45, 46, and 47 is shown in Scheme 4A. A minimal
motion (60° rotation) in the reaction cavity leads to 20R stereochemistry, but a large
rotation (120°) gives the 20S configuration. A minimal motion is preferable to a large
motion in the reaction cavity and is thus allowed to afford 20R stereochemistry. This
theory was proposed by Abe and Rohmer.[18] By contrast, the minimal motion (60°
rotation) of cation 49 with 17-H affords a 20S configuration (allowed, Scheme 4B),
but the large motion (120° rotation) results in the generation of 20R products (not
allowed, Scheme 4B). Thus, cation 49 led to products with 20S stereochemistry. Cation
45 underwent further structural modifications. A hydride shift of H-17 to the C-20
cation, followed by the deprotonation of H-13, furnished 18, 29, and 30 (path a). 1,2-
Shifts of H-17 to the C-20 cation and H-13 to C-17, followed by the removal of H-12,
gave 21, 31, and 32 (path b). The following consecutive 1,2-rearrangement reactions
occurred: H-17C-20 cation, H-13C-17, Me-30C-13, Me-18C14, and the
subsequent deprotonation of H-7 generated product 22 (path c). Proton elimination of
11
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
Me-21 of cation 49 provided product 24 (path d). Cation 49 underwent sequential 1,2-
shift reactions (path c): H-17C-20, H-13C-17, Me-30C-13, Me-18C14. The
3
subsequent deprotonation reaction of H-7 furnished 23 (path c). The nucleophilic
attack of a water molecule on the C-20 cation gave 27 (path e).
4
5
6
<Scheme 3>
7
Distribution ratio of products generated by altering the steric bulk at position 306
Table 1 shows the product distribution ratio obtained by the reactions of squalene 1.
To address the steric bulk, we employed the van der Waals volumes (nm3) reported by
Lin et al.: Gly, 0.00279; Ala, 0.05702; Thr, 0.19341; Val, 0.25674; Leu, 0.37876; and
Phe, 0.55298.[22] These values indicate that the steric bulk increases as follows:
Gly<Ala<Thr<Val<Leu<Phe. The Gly mutant, with decreased steric bulk, afforded the
normal cyclization products 2 and 3 in large quantities and the aberrant cyclization
products 25 and 26 (pentacycle) in small amounts, but did not produce any bi-, tri-, or
tetracyclic products. The Thr variant, with increased bulk, afforded decreased amounts
of 2 and 3 (25%), while the tetracyclic products were generated in increased quantities
(57%). The generation of the tetracyclic products suggests that the Ala306 residue is
located near the D-ring formation site. A further increase in steric bulk (A306V variant)
afforded small amounts of 2 (2%) and of bicyclic (4%) and tricyclic products (2%), but
large quantities of tetracyclic products were produced (90%). Furthermore, almost all
the tetracyclic products possessed a 20R configuration (94%) derived from cation 45,
while only 6% were 20S tetracycles derived from cation 49. In the case of the A306T
variant, 63% of the products consisted of 18, 21, and 22, derived from cation 45
(Scheme 4A), and 37% consisted of 23, 24, and 27, derived from cation 49 (Scheme
4B). Therefore, an increase in the steric bulk facilitates the formation of cation 45.
Cation 8, an intermediary tetracyclic cation (Scheme 1A), is the same as cation 49
(Scheme 4B), indicating that 20S tetracyclic products should be generated from the
intermediary cation 8 or 49 during the normal cyclization process. However, increased
steric bulk at the 306 position drastically increased the production of 20R tetracycles,
strongly indicating that the size of the residue at 306 is critical to the normal
polycyclization pathway. The Leu mutant furnished bicyclic products with a
significantly high yield (43%), and an almost equal amount of tetracyclic products
(39%), as shown in Table 1 and Fig. S3.3. The Leu and Val residues have the same
hydrophobic properties, but possess no polar OH and -electrons; the difference
between them is only the steric size (Leu>Val). Replacing Phe with bulk larger than Val
and Leu generated only bicyclic products and no tetracyclic products. This finding
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
12
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
indicates that the introduction of significantly larger substituents at the 306 position
altered the orientation of substrate 1 inside the reaction cavity. Consequently, the steric
3
bulk at position 306 significantly affects the stereochemical fate and the truncation steps
during the polycyclization cascade.
4
5
Table 2 shows the product distribution ratios obtained by the incubation of 11 and
14. The wild type produced 12 and 13 from 11, and 15 and 16 from 14, but no other
product was produced, as shown in Scheme 1B. The A306T variant produced tetracyclic
products in larger amounts (82%), in addition to 12 and 15 (8%). This result is
consistent with the outcomes from the incubation of 1 (Table 1), but a small amount
(8%) of epoxydammarane 33 was found by GC (Fig. S4). The A306V variant also
produced tetracyclic products with a 20R configuration (47%), and no products with
20S stereochemistry were present above the limit of GC detection. Interestingly, a larger
amount of 33 was produced (20%) than that by the A306T variant. The A306F mutant
produced only bicyclic products 34–36, which was consistent with the results of
squalene incubation with the A306F variant. Notably, products 29–32 were all generated
from cations 46 and 47 (equivalent to 45), which resulted from a chair-chair-chair-boat
folding conformation (Scheme 4A). In addition, 33 was produced only from 14 but not
from 11. The higher production of 33 by the Val mutant than by the Thr mutant indicates
that the production of 33 is also dependent on the steric bulk at position 306. In fact, the
larger Phe variant did not generate any 33 (Table 2). Thus, an appropriate steric size is
required for the formation of 33.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
This study clearly demonstrates that the appropriate steric bulk at position 306 of
AaSHC is critical for directing the normal polycyclization cascade. We describe the
enzymatic reactions of various substrate analogs with SHC, e.g., norsqualenes lacking
Me-group [6a, 23] and homosqualenes bearing ethyl groups on the squalene backbone, [24]
and we provide insight into how the polycyclization reaction is affected by the specific
modifications. The steric bulk at the Me-positions of squalene substrate significantly
altered the polycyclization pathway, e.g., alteration of the folding conformation leading
to different stereochemistry and a truncation of the ring-forming cascade, which in some
cases resulted in the generation of novel triterpene scaffolds.[23b ] The results of the site-
specific mutagenesis experiments provide further evidence that the appropriate steric
size of the active site residues is critical to correct folding and the full cyclization
pathway.
36 Conclusions
13
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
The following conclusions can be drawn from our study: (1) the appropriate steric
2
3
bulk of the Ala306 residue directs the normal polycyclization reaction without
premature termination; (2) the Ala side chain is in close proximity to the D-ring
formation site, as the A306V mutant produced large amounts of tetracyclic products
with the 20R configuration; and (3) this is the first report describing the initiation of the
polycyclization reaction from the terminal double bond rather than from the epoxide
ring. As discussed above, the production of 33 raises the question why the native SHC
does not produce epoxydammarane structure(s) from 11 or 14, specifically why the
proton attack occurs only at the epoxide ring, but not the terminal double bond, despite
both the terminal double bond and the epoxide ring entering the reaction cavity through
the substrate channel. The A306V and A306T variants are able to accept both heads of
oxidosqualene 14 because products 29–32 are generated by a proton attack on the
epoxide ring, while product 33 is produced by a proton attack on the terminal double
bond.
4
5
6
7
8
9
10
11
12
13
14
15
16
17 Experimental
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
General analytical methods: NMR spectra were recorded in C6D6 on a Bruker
DMX 600 or DPX 400 spectrometer, the chemical shifts () given in ppm relative to the
residual solvent peak H 7.280 and C128.0 as the internal reference for 1H- and 13C-
NMR spectra, respectively. In the case of CDCl3 solutions, the chemical shifts are given
in ppm relative to the solvent peak (H 7.26; C 77.0). The coupling constants J are given
in Hz. GC analyses were performed on a Shimadzu GC-8A chromatograph equipped
with a flame ionization detector (DB-1 capillary column, 0.53 mm  30 m). GC-MS
spectra were performed on a JEOL SX 100 spectrometer under electronic impact at 70
eV with a DB-1 capillary column (0.32 mm x 30 m). HR-EIMS was performed by
direct inlet system. Specific rotation values were measured at 25°C with a Horiba
SEPA-300 polarimeter.
Site-directed mutagenesis
The shc genes were mutated via oligonucleotide-directed in vitro mutagenesis using
the Pharmacia Unique Site Elimination Mutagenesis kit (Pfizer, New York, NY, USA).
Plasmid pKS, a pUC119 derivative containing a 1 kb KpnI-SacI fragment, was used as a
template for all mutagenesis reactions. The synthetic oligonucleotides for the mutations
(with changes to the wild type sequence shown in bold type) are as follows:
14
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
3
4
5
A306G: 5'-GGATGTTTCAGGGGTCGATATCGCCGGTGTG-3'
A306T: 5'-GGATGTTTCAGACTTCGATATCGCCGGTGTG-3'
A306V: 5'-GGATGTTTCAGGTTTCGATATCGCCGGTGTG-3'
A306L: 5'-GGATGTTTCAGTTGTCGATATCGCCGGTGTG-3'
A306F: 5'-GGATGTTTCAGTTTTCGATATCGCCGGTGTG-3'
6
7
Silent mutations were also used to screen the desired mutants by restriction fragment
analysis (underlined sequences). To ascertain that the desired mutation had been
achieved, the entire region of the inserted DNA was sequenced.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Incubation conditions and GC analysis
The SHC described here is from A. acidocaldarius.[7-14, 16-17, 19,20] Standard culture of
the cloned Escherichia coli and incubation conditions were performed according to our
published protocols.[7] The cell-free extract was prepared as follows. One liter of E. coli
culture encoding the native and mutated SHCs (pET3a) was harvested by centrifugation
50 mL of citrate buffer solution (pH 6.0, 50 mM) was added to the pellets, and
ultrasonication was then performed to disrupt the cells. The supernatant was used for
incubations after removal of all cell debris by centrifugation. One mL of the supernatant
contained ca. 200 g of the pure SHCs. One mg of the substrates, 1 and a mixture of 11
and 14, and 20 mg of Triton X-100 were emulsified with 1.0 mL dH2O and 3.0 mL Na-
citrate buffer solution (pH 6.0, 0.5 M). To this solution, 2.0 mL of the cell-free extract
was added, followed by incubation at 45°C for 12 h. To terminate the enzyme reaction,
6 mL of 15% KOH/MeOH was added and heated at 80°C for 30 min. The lipophilic
enzyme products and the substrate analogs, which remained unreacted, were extracted 4
times with hexane (5 mL) from the incubation mixtures, and the quantities of the
products and the starting materials were estimated by GC analyses with a DB-1
capillary column (30 m in length, J&W Scientific, Folsom, CA, USA). GC-EIMS
product profiles and substrate distributions, prepared after the saponification treatment
(quenching reaction), were identical to those obtained prior to the saponification
reaction, and hence there was no change in product distribution and structural
modification, or the substrates produced by the quenching conditions. The epoxide ring
was also unmodified by this quenching treatment. GC analyses shown in Fig. 1A were
performed under the following conditions: injection temperature, 290°C; column
temperature, 290°C; N2 carrier pressure, 0.75-1.5 kg/cm2. GC analyses shown in Fig.
1B were performed on the same GC instrument and GC column as (A), but the N2
carrier pressure was 0.5 kg/cm2.
Spectroscopic data of enzymatic products.
15
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
Almost the enzymatic products obtained from squalene incubation were reported in our
previous papers, as described in Text.[2b] To validate the identification, we isolated
3
products 17─24 and determined by the detailed NMR analyses: see the Figs. S5─S12 in
Supporting Information. Production of 23 by the mutated SHCs has not been reported
before. Production yields of 25─28 were significantly small, thus identifications were
made by GC-EIMS, as described in Fig. S13. NMR and EIMS data of products 29─36,
which have not been reported before. Therefore, the spectroscopic data of 23 and 29─36
are given here.
4
5
6
7
8
9
Product 23: []D25 = -0.565 (c=0.018, EtOH); 1H NMR (600 MHz, C6D6): =0.964 (s,
3H, Me-19), 0.967 (s, 3H, Me-28), 1.03 (s, 3H, Me-29), 1.04 (m, 1H, H-1), 1.05 (s, 3H,
Me-18), 1.12 (d, J=6.4 Hz, Me-21), 1.18 (s, 3H, Me-30), 1.27 (m, 2H, H-22 and H-3),
1.42 (m, 1H, H-16), 1.48 (dd, J=12.1, 5.4 Hz, 1H, H-5), 1.52 (m, 2H, H-2 and H-3),
1.58 (m, 1H, H-20), 1.60 (m, 1H, H-11), 1.62 (m, 1H, H-15), 1.63 (m, 1H, H-17), 1.66
(m, 1H, H-11), 1.68 (m, 1H, H-2), 1.72 (m, 1H, H-22), 1.74 (m, 1H, H-1), 1.75 (s, 3H,
Me-27), 1.77 (m, 2H, H-12 and H-15), 1.83 (s, 3H, Me-26), 1.92 (m, 1H, H-12), 2.00
(m, 1H, H-6), 2.09 (m, 1H, H-16), 2.14 (m, 1H, H-23), 2.25 (m, 1H, H-6), 2.33 (m, 1H,
H-23), 2.42 (m, 1H, H-9), 5.42 (bt, J=6.2 Hz, H-24), 5.47(bd, J=2.6 Hz, 1H, H-7) ; 13C
NMR (150 MHz, C6D6):=13.36 (q, C-19), 17.72 (q, C-27), 18.48 (t, C-11), 18.62 (q,
C-21), 19.44 (t, C-2), 21.50 (q, C-29), 22.25 (q, C-18), 24.70 (t, C-6), 25.53 (t, C-23),
25.88 (q, C-26), 27.56 (q, C-30), 28.60 (t, C-16), 33.16 (q, C-28), 33.29 (s, C-4), 34.29
(t, C-12), 34.46 (t, C-15), 35.39 (s, C-10), 36.37 (d, C-20), 36.64 (t, C-22), 39.26 (t, C-
1), 42.69 (t, C-3), 43.82 (s, C-13), 49.50 (2C, d, C-9 and s, C-14), 51.66 (d, C-5), 53.42
(d, C-17), 118.5 (d, C-7), 125.7 (d, C-24), 130.7 (s, C-25), 146.1 (s, C-8). Me-21
showed H 0.885 (d, J=6.8 Hz) in the CDCl3 solution. The assignments of C-13 and C-
14 may be exchangeable. MS(EI): m/z: 69 (30%), 109 (28%), 395 (100%), 410 (27%);
[M+]; HRMS (EI): m/z: calcd. for C30H50: 410.39125; found: 410.39187.
Product 29: []D25= -17.15 (c=0.206, EtOH); 1H NMR (400 MHz, C6D6): =0.872 (s,
3H, Me-29), 0.949 (s, 3H, Me-19), 1.045 (s, 3H, Me-28), 1.105 (s, 3H, Me-18), 1.141
(d, J=7.2 Hz, 3H, Me-21), 1.253 (s, 3H, Me-30), 1.39 (m, 1H, H-11), 1.40 (m, 1H, H-
15), 1.43 (m, 1H, H-6), 1.47 (m, 2H, H-1), 1.48 (m, 1H, H-7), 1.53 (2H, m, 1H, H-22;
m, 1H, H-5), 1.54 (m, 1H, H-6), 1.55 (m, 1H, H-2), 1.62 (m, 1H, H-22), 1.67 (m, 1H,
H-11), 1.70 (m, 1H, H-7), 1.71 (m, 1H, H-9), 1.752 (s, 3H, Me-27), 1.828 (s, 3H, Me-
26), 1.93 (m, 1H, H-2), 2.02 (m, 1H, H-12), 2.07 (m, 1H, H-15), 2.10 (m, 1H, H-23),
2.25 (m, 1H, H-23), 2.28 (m, 1H, H-16), 2.46 (m, 1H, H-16), 2.59 (dd, J=13.6, 3.6 Hz,
1H, H-12), 2.72 (m, 1H, H-20), 3.29 (br s, 1H, H-3), 5.42 (t, J=6.4 Hz, H-24); 13C NMR
(100 MHz, C6D6):=16.50 (q, C-19), 17.53 (q, C-18), 17.79 (q, C-27), 18.53 (t, C-6),
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
16
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
20.18 (q, C-21), 22.09 (t, C-11), 22.39 (q, C-29), 23.23 (q, C-30), 23.30 (t, C-12), 25.86
(q, C-26), 25.94 (t, C-2), 27.18 (t, C-23), 28.59 (q, C-28), 29.47 (t, C-16), 31.05 (t, C-
3
15), 32.24 (d, C-20), 33.90 (t, C-1), 35.72 (t, C-7), 36.02 (t, C-22), 37.74 (s, C-10),
37.79 (s, C-4), 41.60 (s, C-8), 49.71 (d, C-5), 51.82 (d, C-9), 56.94 (s, C-14), 75.80 (d,
C-3), 125.5 (d, C-24), 130.9 (s, C-25), 134.9 (s, C-17), 139.7 (s, C-13). The assignments
of C-4 and C-10 may be exchangeable. MS (EI) see Fig. S14.1. HRMS (EI): m/z: calcd.
for C30H50O: 426.38616; found: 410.38896.
4
5
6
7
8
Product 30: []D25= -10.9 (c=0.278, EtOH); 1H NMR (600 MHz, C6D6): =0.796 (dd,
J=11.8, 1.4 Hz, 1H, H-5), 0.909 (s, 3H, Me-19), 0.912 (s, 3H, Me-29), 0.92 (m, 1H, H-
1), 1.070 (s, 3H, Me-18), 1.149 (s, 3H, Me-28), 1.152 (d, J=6.7 Hz, 3H, Me-21), 1.275
(s, 3H, Me-30), 1.40 (2H, m, 1H, H-11; m, 1H, H-15), 1.44 (m, 1H, H-6), 1.46 (m, 1H,
H-7), 1.47 (m, 1H, H-9), 1.53 (m, 1H, H-22), 1.56 (m, 2H, H-2), 1.62 (2H: m, 1H, H-
22; m, 1H, H-7), 1.63 (m, 1H, H-11), 1.64 (m, 1H, H-6), 1.69 (m, 1H, H-1), 1.753 (s,
3H, Me-27), 1.831 (s, 3H, Me-26), 2.02 (m, 1H, H-12), 2.07 (m, 1H, H-15), 2.09 (m,
1H, H-23), 2.24 (m, 1H, H-23), 2.28 (m, 1H, H-16), 2.47 (m, 1H, H-16), 2.59 (ddd,
J=12.5, 4.7, 1.8 Hz, 1H, H-12), 2.72 (m, 1H, H-20), 3.16 (dd, J=10.2, 6.1 Hz, 1H, H-3),
5.41 (t, J=7.0 Hz, H-24); 13C NMR (150 MHz, C6D6):=15.83 (q, C-29), 16.61 (q, C-
19), 17.52 (q, C-18), 17.79 (q, C-27), 18.59 (t, C-6), 20.18 (q, C-21), 22.17 (t, C-11),
23.11 (q, C-30), 23.31 (t, C-12), 25.86 (q, C-26), 27.17 (t, C-23), 27.89 (t, C-2), 28.35
(q, C-28), 29.47 (t, C-16), 31.06 (t, C-15), 32.26 (d, C-20), 35.84 (t, C-7), 36.01 (t, C-
22), 37.62 (s, C-10), 39.15 (s, C-4), 39.26 (t, C-1), 41.34 (s, C-8), 52.02 (d, C-9), 56.29
(d, C-5), 56.83 (s, C-14), 78.46 (d, C-3), 125.4 (d, C-24), 130.9 (s, C-25), 135.1 (s, C-
17), 139.5 (s, C-13). MS (EI) see Fig. S15.1. HRMS (EI): m/z: calcd. for C30H50O:
426.38616; found: 410.38648.
Product 31: []D25= +6.3 (c=0.074, EtOH); 1H NMR (400 MHz, C6D6): =0.900 (s,
3H, Me-29), 1.011 (s, 3H, Me-18), 1.030 (s, 3H, Me-19 ), 1.047 (d, J=6.8 Hz, 3H, Me-
21), 1.065 (s, 3H, Me-28), 1.131 (s, 3H, Me-30), 1.32 (2H: m, 1H, H-1; m, 1H, H-15),
1.44 (m, 1H, H-22), 1.46 (m, 1H, H-7), 1.50 (m, 1H, H-6), 1.51 (m, 1H, H-2), 1.52 (m,
1H, H-1), 1.57 (2H; m, 1H, H-5; m, 1H, H-6), 1.62 (m, 1H, H-22), 1.63 (m, 2H, H-16),
1.730 (s, 3H, Me-27), 1.77 (m, 1H, H-15), 1.80 (m, 1H, H-7), 1.85 (m, 1H, H-2), 1.884
(s, 3H, Me-26), 1.92 (2H: m, 1H, H-11; m, 1H, H-9), 1.98 (m, 1H, H-20), 2.16 (m, 1H,
H-11), 2.23 (m, 2H, H-23), 2.59 (m, 1H, H-17), 3.29 (bs, 1H, H-3), 5.41 (2H: m, 1H, H-
12; m, 1H, H-24); 13C NMR (100 MHz, C6D6):=15.48 (q, C-21), 15.71 (q, C-19),
17.05 (q, C-18), 17.74 (q, C-27), 18.69 (t, C-6), 22.44 (q, C-29), 22.85 (q, C-30), 23.86
(t, C-11), 24.49 (t, C-16), 25.76 (t, C-2), 25.89 (q, C-26), 26.53 (t, C-23), 28.59 (q, C-
28), 32.07 (t, C-15), 33.46 (t, C-1), 34.71 (d, C-20), 35.06 (t, C-7), 36.35 (t, C-22),
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
17
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
37.60 (s, C-10), 37.61 (s, C-4), 38.52 (s, C-8), 48.36 (d, C-9), 48.94 (d, C-17), 49.75 (d,
C-5), 50.57 (s, C-14), 75.77 (d, C-3), 115.9 (d, C-12), 125.5 (d, C-24), 130.9 (s, C-25),
3
149.2 (s, C-13). The assignments of C-4 and C-10 may be exchangeable. MS (EI) see
4
Fig. S16.1. HRMS (EI): m/z: calcd. for C30H50O: 426.38616; found: 410.38432.
1
5
Product 32: []D25= +8.0 (c=0.213, EtOH); H NMR (600 MHz, C6D6): =0.865 (dd,
6
J=12.0, 10.3 Hz, 1H, H-5), 0.92 (m, 1H, H-1), 0.947 (s, 3H, Me-29), 0.973 (s, 3H, Me-
18), 0.991 (s, 3H, Me-19), 1.061 (d, J=6.7 Hz, 3H, Me-21), 1.148 (s, 3H, Me-30), 1.157
(s, 3H, Me-28), 1.30 (dd, J=6.3, 11.8 Hz, 1H, H-15), 1.42 (m, 1H, H-7), 1.47 (m, 1H, H-
22), 1.50 (m, 1H, H-6), 1.52 (m, 1H, H-2), 1.56 (2H: m, 1H, H-1; m, 1H, H-2), 1.62 (m,
2H, H-16), 1.63 (m, 1H, H-22), 1.64 (m, 1H, H-6), 1.68 (m, 1H, H-7), 1.72 (m, 1H, H-9),
1.732 (s, 3H, Me-27), 1.78 (m, 1H, H-15), 1.832 (s, 3H, Me-26), 1.89 (m, 1H, H-11), 1.99
(m, 1H, Me-20), 2.08 (m, 1H, H-11), 2.24 (m, 2H, H-23), 2.59 (m, 1H, H-17), 3.14 (dd,
J=11.4, 4.7 Hz, 1H, H-3), 5.39 (bs, 1H, H-12), 5.47 (t, J=7.4 Hz, H-24); 13C NMR (150
MHz, C6D6):= 15.48 (q, C-21), 15.88 (2C; q, C-19; q, C-29), 16.97 (q, C-18), 17.73 (q,
C-27), 18.72 (t, C-6), 22.83 (q, C-30), 23.91 (t, C-11), 24.50 (t, C-16), 25.88 (q, C-26),
26.53 (t, C-23), 27.71 (t, C-2), 28.32 (q, C-28), 32.08 (t, C-15), 34.72 (d, C-20), 35.08 (t,
C-7), 36.34 (t, C-22), 37.48 (s, C-10), 38.30 (s, C-8), 38.96 (t, C-1), 39.06 (s, C-4), 48.57
(d, C-9), 48.94 (d, C-17), 50.47 (s, C-14), 56.32 (d, C-5), 78.50 (d, C-3), 115.9 (d, C-12),
125.5 (d, C-24), 130.9 (s, C-25), 149.3 (s, C-13). MS (EI) see Fig. S17.1. HRMS (EI):
m/z: calcd. for C30H50O: 426.38616; found: 410.38648.
Product 33: []D25= -20.8 (c=0.239, EtOH); 1H NMR (600 MHz, acetone d6): =0.817
(s, 3H, Me-29), 0.82 (m, 1H, H-5), 0.849 (s, 3H, Me-28), 0.86 (m, 1H, H-1), 0.861 (s,
3H, Me-19), 0.969 (s, 3H, Me-18), 1.025 (s, 3H, Me-30), 1.050 (s, 3H, Me-27), 1.06 (m,
1H, H-15), 1.135 (s, 3H, Me-26), 1.14 (m, 1H, H-3), 1.188 (s, 3H, Me-21), 1.21 (m, 1H,
H-11), 1.26 (m, 1H, H-7), 1.32 (m, 1H, H-3), 1.38 (m, 1H, H-2), 1.39 (m, 1H, H-6),
1.43 (2H: m, 1H, H-15; m, 1H, H-9), 1.49 (m, 1H, H-12), 1.52 (m, 1H, H-6), 1.53 (m,
1H, H-16), 1.58 (m, 1H, H-11), 1.59 (m, 1H, H-22), 1.61 (2H: m, 1H, H-2; m, 1H, H-7),
1.65 (m, 1H, H-1), 1.75 (m, 1H, H-23), 1.76 (m, 1H, H-16), 1.85 (m, 1H, H-23), 1.92
(2H: m, 1H, H-12; m, 1H, H-22), 2.09 (m, 1H, H-13), 2.24 (m, 1H, H-17), 3.71 (dd,
J=10.6, 5.2 Hz, 1H, H-24); 13C NMR (150 MHz, acetone d6):= 16.33 (q, C-18), 16.64
(q, C-19), 16.94 (q, C-30), 19.37 (t, C-6), 19.47 (t, C-2), 21.86 (q, C-29), 23.21 (t, C-
11), 25.83 (q, C-27), 26.40 (t, C-23), 26.84 (t, C-12), 27.37 (q, C-26), 28.29 (t, C-16),
28.91 (q, C-21), 33.12 (t, C-15), 33.76 (q, C-28), 33.95 (s, C-4), 36.09 (t, C-7), 38.19 (s,
C-10), 38.70 (t, C-22), 41.35 (t, C-1), 41.68 (s, C-8), 42.89 (t, C-3), 44.60 (d, C-13),
49.36 (d, C-17), 49.85 (s, C-14), 51.71 (d, C-9), 57.79 (d, C-5), 70.43 (s, C-25), 86.06
(s, C-20), 88.23 (d, C-24). MS (EI) see Fig. S18.1. HRMS (EI): m/z 429 (M+-Me):
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
18
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
calcd. for C29H49O2: 426.37325; found: 429.37229.
Product 34-acetate: []D25= -9.5 (c=0.014, EtOH); 1H NMR (600 MHz,C6D6): =0.892
3
(s, 3H, Me-24), 1.051 (s, 3H, Me-25), 1.052 (s, 3H, Me-23), 1.50 (m, 1H, H-6), 1.62 (m,
1H, H-6), 1.68 (m, 1H, H-1), 1.691 (s, 3H, Me-29), 1.724 (s, 3H, Me-26), 1.724 (s, 3H,
Me-28), 1.779 (s, 3H, Me-27), 1.806 (s, 3H, Me-30), 1.82 (s, 3H, Me-32, acetyl Me),
1.85 (m, 1H, H-1, 1.88 (2H, m, H-2), 1.92 (m, 1H, H-5), 2.01(dd, J=6.0, 18.2 Hz, 1H,
H-7), 2.14 (m, 1H, H-12), 2.16 (m, 1H, H-7), 2.24 (4H: m, H-15 and H-19), 2.32 (m,
1H, H-12), 2.33 (m, 4H, H-16 and H-20), 2.38 (m, 2H, H-11), 5.03 (bs, 1H, H-3), 5.37
(t, J=6.8 Hz, 1H, H-21), 5.42 (t, J=6.8 Hz, 1H, H-13), 5.43 (t, J=6.8 Hz, 1H, H-17); 13C
NMR (150 MHz, C6D6):= 16.12 (q, C-28), 16.21 (q, C-27), 169.7 (s, C-31, acetyl
CO), 17.72 (q, C-29), 18.87 (t, C-6), 19.70 (q, C-26), 20.02 (q, C-25), 20.76 (q, C-32,
acetyl CH3), 21.84 (q, C-24), 23.71 (t, C-2), 25.83 (q, C-30), 27.20 (2C: t, C-16 and t,
C-20), 27.88 (q, C-23), 28.74 (t, C-12), 29.65 (t, C-11), 30.75 (t, C-1), 33.66 (t, C-7),
36.99 (s, C-4), 38.98 (s, C-10), 40.19 (t, C-15), 40.19 (t, C-19), 46.52 (d, C-5), 77.48 (d,
C-3), 124.7 (d, C-17), 124.9 (d, C-21), 125.3 (d, C-13), 126.2 (s, C-8), 131.1 (s, C-22),
134.8 (s, C-18), 135.1 (s, C-14), 140.4 (s, C-9). The assignments of C-14 and C-18 may
be exchangeable. MS (EI) see Fig. S19.1. HRMS (EI): m/z 408 (M+-CH3CO): calcd. for
C30H48: 408.37560; found: 408.37590.
Product 35-acetate: []D25= -20.0 (c=0.025, EtOH); 1H NMR (600 MHz,C6D6):
=0.867 (s, 3H, Me-25), 0.893 (s, 3H, Me-24), 0.957 (s, 3H, Me-23), 1.11 (m, 1H, H-1),
1.44 (m, 1H, H-11), 1.56 (m, 1H, H-1), 1.61 (m, 1H, H-11), 1.691 (s, 3H, Me-29), 1.740
(s, 6H, Me-27 and Me-28), 1.806 (s, 3H, Me-30), 1.83 (3H: m, 2H, H-2; m, 1H, H-5),
1.88 (m, 1H, H-9), 1.889 (s, 3H, Me-26), 1.890 (s, 3H, Me-32, acetyl Me), 1.96 (very
broad, 2H, H-6), 2.15 (m, 1H, H-12), 2.24 (m, 4H, H-15 and H-19), 2.33 (m, 4H, H-16
and H-20), 2.34 (m, 1H, H-12), 4.99 (t, J=2.4 Hz, H-3), 5.37 (t, J=6.8 Hz, H-21), 5.42
(t, J=6.8 Hz, H-13), 5.43 (t, J=6.8 Hz, H-17), 5.45 (very br s, 1H, H-7) ; 13C NMR (150
MHz, C6D6):= 13.57 (q, C-25), 16.10 (q, C-28), 16.28 (q, C-27), 169.6 (s, acetyl CO),
17.72 (q, C-29), 20.77 (q, acetyl CH3), 21.90 (q, C-24), 22.36 (q, C-26), 23.08 (t, C-2),
23.59 (t, C-6), 25.83 (q, C-30), 27.20 (2C: t, C-16 and C-20), 27.67 (2C: q, C-23; t, C-
11 ), 30.61 (t, C-12), 32.29 (t, C-1), 36.54 (s, C-4), 36.58 (s, C-10), 40.19 (t, C-19),
40.22 (t, C-15), 44.99 (d, C-5), 54.12 (d, C-9), 77.90 (d, C-3), 122.4 (d, C-7), 124.7 (d,
C-17), 124.9 (d, C-21),125.3 (d, C-13), 131.1 (s, C-22), 135.1 (s, C-18), 135.2 (s, C-14),
135.6 (s, C-8). The assignments may be exchangeable between C-4 and C-10, between
C-14 and C-18, between C-27 and C-28 and C-15 and C-19. MS (EI) see Fig. S20.1.
HRMS (EI): m/z 408 (M+-CH3CO): calcd. for C30H48: 408.37560; found: 408.37759.
Product 36: []D25= -34.4 (c=1.697, EtOH); 1H NMR (400 MHz,C6D6): =0.878 (s,
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
19
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
3H, Me-25), 0.975 (s, 3H, Me-24), 1.084 (s, 3H, Me-23), 1.11 (m, 1H, H-1), 1.22 (t,
J=8.4 Hz, H-5), 1.40 (m, 1H, H-11), 1.41 (m, 1H, H-2), 1.55 (m, 1H, H-2), 1.60 (m, 1H,
3
H-11), 1.69 (m, 1H, H-9), 1.694 (s, 3H, Me-29), 1.750 (s, 3H, Me-28), 1.768 (s, 3H,
Me-27), 1.807 (s, 3H, Me-30), 1.835 (s, 3H, Me-26), 1.87 (m, 1H, H-1), 2.02 (very b r,
2H, H-6), 2.16 (m, 1H, H-12), 2.27 (m, 4H, H-15 and H-19), 2.27-2.38 (m, 4H, H-16
and H-20), 2.38 (m, 1H, H-12), 3.15 (dd, J=10.8, 4.8 Hz, H-3), 5.37 (t, J=6.8 Hz, 1H,
H-21), 5.42 (t, J=6.8 Hz, 1H, H-13), 5.44 (t, J=6.8 Hz, 1H, H-17), 5.53 (very br s, 1H,
H-7); 13C NMR (100 MHz, C6D6):=13.71 (q, C-25), 131.1 (s, C-22), 16.12 (q, C-28),
16.26 (q, C-27), 17.73 (q, C-29), 22.27 (q, C-26), 23.81 (t, C-6), 25.30 (q, C-24), 25.84
(q, C-30), 27.13 (t, C-20), 27.23 (t, C-16), 27.70 (t, C-11), 27.93 (t, C-2), 28.12 (q, C-
23), 30.66 (t, C-12), 36.78 (s, C-10), 37.54 (t, C-1), 38.83 (s, C-4), 40.22 (2C, t, C-15
and C-19), 49.84 (d, C-5), 54.37 (d, C-9), 78.76 (d, C-3), 122.5 (d, C-7), 124.7 (d, C-
17), 124.9 (d, C-21), 125.3 (d, C-13), 135.1 (2C, s, C-14 and C-18), 135.2 (s, C-8). The
assignments may be exchangeable between C-14 and C-18, between C-16 and C-20 and
between C-27 and C-28. MS (EI) see Fig. S21.1. HRMS (EI): m/z: calcd. for C30H50O:
426.38616; found: 410.38354.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19 Acknowledgements
20
21
22
23
24
This work was supported in part by Grant-in-Aid for Scientific Research from Japan
Society for the Promotion of Science, Nos. 25450150 (C) and 18380001(B).
Keywords: squalene; cyclization, enzyme catalysis; hopene; epoxydammarane
25 References
26
27
28
29
30
31
32
33
34
35
36
[1] a) R. B. Woodward , K. Bloch, J. Am. Chem. Soc., 1953, 75, 2023-2024. b) A.
Eschenmoser, L. Ruzicka, O. Jeger , D. Arigoni, Helv. Chim. Acta 1955, 38, 1890–
1904. c) J. W. Cornforth, R. H. Cornforth , C. Donniger, G. Popjak, Y. Shimizu, S.
Ichii, E. Forchielli , E. Caspi, J. Am. Chem. Soc., 1965, 87, 3224-3228.
[2] For a review of squalene, oxidosqualene cyclases, see: a) K. U. Wendt, G. E. Schulz,
E. J. Corey, D. R. Liu, Angew. Chem., Int. Ed. 2000, 39, 2812–2833; b) T. Hoshino,
T. Sato, Chem. Commun. 2002, 291–301; c) R. A. Yoder, J. N. Johnston, Chem. Rev.
2005, 105, 4730–4756; d) M. J. R. Segura, B. E. Jackson, S. P. T. Matsuda, Nat.
Prod. Rep., 2003, 20, 304-317; e) I. Abe, Nat. Prod. Rep. 2007, 24, 1311–1331; f)
T. K. Wu, C. H. Chang, Y. T. Liu, T. T. Wang, Chem. Rec. 2008, 8, 302–325; g) W.
20
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
3
4
D. Nes, Chem. Rev. 2011, 111, 6423–6451; h) G. Siedenburg, D. Jendrossek, Appl.
Environ. Microbiol., 2011, 77, 3905-3915; i) P.-O. Syrén, S. Henche, A. Eichler, B.
M. Nestl, B. Hauer, Curr. Opin. Struct. Biol. 2016, 41, 73-82; j) T. Hoshino, Org.
Biomol. Chem., 2017, 15, 2869-2891.
5
6
[3] K. U. Wendt, K. Poralla, G. E. Schulz, Science, 1997, 277, 1811-1815.
[4] a) K. U. Wendt, A. Lenhart, G. E. Schulz, J. Mol. Biol., 1999, 286, 175-187; b) D. J.
Reinert, G. Balliano, G. E. Schulz, Chem. Biol., 2004, 6, 1717-1720.
[5] a) R. Rajamani, J. Gao, J. Am. Chem. Soc., 2003, 125, 12768-12781; b) L. Smentek,
B. A. Hess, Jr., J. Am. Chem. Soc., 2010, 132, 17111-17117.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
[6] a) T. Hoshino, S. Nakano, T. Kondo, T. Sato, A. Miyoshi, Org. Biomol. Chem., 2004,
2, 1456-1470; b) M. Rohmer, C. Anding, G. Ourisson, Eur. J. Biochem., 1980, 112,
541-547; c) P. Bouvier, Y. Berger, M. Rohmer, G. Ourisson, Eur. J. Biochem., 1980,
112, 549-556; d) R. Rohmer, P. Bouvier, G. Ourisson, Eur. J. Biochem., 1980, 112,
557-560.
[7] a) T. Sato, T. Hoshino, Biosci, Biotechnol. Biochem., 1999, 63, 2189-2198; (b) T.
Dang, G. D. Prestwich, Chem. Biol., 2000, 7, 643–649.
[8] T. Sato, T. Hoshino, Biosci, Biotechnol. Biochem., 2001, 65, 2233-2142.
[9] T. Hoshino, T. Sato, Chem. commun., 1999, 2005-2006.
[10] T. Hoshino, M. Kouda, T. Abe , S. Ohashi, Biosci, Biotechnol. Biochem., 1999, 63,
2038-2041.
[11] T. Hoshino, M. Kouda, T. Abe , T. Sato, Chem. Commun., 2000, 1485-1486.
[12] M. 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, 13148-13194.
[13] T. Sato, S. Sasahara, T. Yamakami, T. Hoshino, Biosci, Biotechnol. Biochem., 2002,
66, 1660-1670.
[14] T. Hoshino, T. Abe, M. Kouda, Chem. Commun., 2000, 441-442.
a) L. C. Kühnel, B. M. Nestl, B. Hauer, ChemBioChem, 2017, 18, 2222-2225; b) S.
[15]
A. Bastian, S. C. Hammer, N. Kreß, B. M. Nestle, B. Hauer, ChemCatChem., 2017,
9, 4364-4368; c) S. C. Hammer, A. Marjanovic, J. M.
Dominicus, B. M. Nestle, B.
Hauer, Nat. Chem. Biol., 2015, 11, 121-126; d) M. Seitz, P. O. Syrén, M. Seitz, B.
M. Nestle, B. Hauer, ChemBioChem, 2013, 14, 436-439.
Biosci, Biotechnol. Biochem., 1999, 63, 1171-1180.
[16] T. Sato and T. Hoshino,
[17] T. Sato, T. Abe, T. Hoshino, Chem. Commun., 1998, 2617-2618.
[18] I. Abe , M. Rohmer, J. Chem. Soc., Perkin Trans 1, 1994, 783-791.
[19] T. Sato, M. Kouda, T. Hoshino, Biosci., Biotechnol. Biochem., 2004, 68, 728-738.
[20] T. Hoshino, Y. Yonemura, T. Abe , Y. Sugino, Org. Biomol. Chem., 2007, 5, 792-801.
21
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
[21] D. Ueda, T. Hoshino, T. Sato, J. Am. Chem. Soc., 2013, 135,18335-18338.
[22] Z-h Lin, H-x Long, Z. Bo, Y-q Wang, , Y-z Wu, Peptides, 2008, 29, 1798-1805.
3
[23] a) T. Hoshino and T. Kondo, Chem. Commun. 1999, 731-732; b) T. Hoshino and S.
Ohashi , Org. Lett., 2002, 4, 2553-2556; c) S. Nakano, S. Ohashi and T. Hoshino,
Org. Biomol. Chem., 2004, 2, 2012-2022.
4
5
6
[24] a) K. Takahashi, Y. Sasaki, T. Hoshino, Eur. J. Org. Chem. 2018, 1477-1490; b) I.
Kaneko, Y. Terasawa, T. Hoshino, Chem. Eur. J., in press:
7
8
doi.org/10.1002/chem.201801668
9
10
11
12
13
22
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
Figure, Scheme, and Table Legends
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Scheme 1. (A) Polycyclization pathways of squalene (1) into hopene (2) and hopanol
(3) by squalene-hopene cyclase. Squalene 1 folds in an all-prechair conformation in the
23
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
3
4
5
6
7
8
9
reaction cavity. (B) The cyclization consists of 8 reaction steps. (C) Polycyclization
reactions of (3S)-2,3-oxidosqualene (11) and (3R)-2,3-oxidosqualene (14). Here, 11
folds into an all-prechair structure similar to that in the polycyclization reaction of 1,
leading to 3-hydroxyhopene (12) and 3-hydroxyhopanol (13). However, 14 folds into
a boat-chair-chair-chair-chair conformation, resulting in the production of 3-
hydroxyhopene (15) and 3-hydroxyhopanol (16).
24
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
3
4
21
5
22
(A)
6
18
7
8
23
17
9
19
1
24
2
20
10
11
12
13
14
15
16
17
18
19
20
21
22
16
14
22
18
20
12
8
10
Time (min)
(B)
11 and 14
30
33
29
31
50
32
20
40
70
30
60
Time (min)
23
24
25
26
27
28
29
30
31
32
Fig. 1. (A) Incubation mixture of squalene (1) with the A306V variant. (B) Incubation
mixture of 2,3-oxidosqualene (a mixture of 11 and 14) with the A306V variant. The
cell-free extracts (5 mL) were prepared from a 100 mL culture of the A306V variant.
Incubation of 1.0 mg of 1 and that of a mixture of 11 and 14 (1.0 mg) with 2.0 mL of
cell-free homogenates was carried out at 45°C for 24 h. The lipophilic materials were
extracted with hexane. Triton X-100 present in the hexane extract was removed by
passage through a short SiO2 column using hexane/EtOAc (100:20) as eluent. The
lipophilic materials thus collected were subjected to GC measurements. The GC
conditions are described in the Experimental section.
25
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Fig. 2. Structures of the enzymatic products produced by A306X variants. Products 17–
24 were produced from the incubation of 1 with the A306V mutant. Products 18, 21–27
26
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
were generated by the incubation of 1 with the A306T variant (Fig. S3.1). Products 17,
19, and 28 were produced by the reaction of 1 with the A306F mutant (Fig. S.1.2). Fig.
3
S4 shows the GC profiles of the incubation mixtures of a racemic mixture of 11 and 14
with the A360T and A306F variants. Products 29–33 were produced by the A306T
mutant. Products 34–36 were produced by the A306F mutant.
4
5
6
Compound names. 17: -polypodateraene, 18: dammara-13(17)-24-diene, 19: -
polypodateraene, 20: (17E)-(13H)-malabarica-14(27),17,21-triene, 21: 20R-dammara-
12(13), 24-diene, 22: 20R-eupha-7(8),24-diene, 23: 20S-tirucalla-7(8),24-diene (a new
SHC-mediated product), 24: 17-epi-dammara-20(21),24-diene, 25: neohop-13(18)-ene,
26: hop-21(22)-ene, 27: 20-hydroxy-17-epi-dammaraene, 28: neopolypoda-
5(6),13,17,21-teraene (rearranged skeleton of bicyclic polypodatetraene), 29: 20R-
dammara-13(17),24-dien-3-ol, 30: 20R-dammara-13(17),24-dien-3-ol, 31: 20R-
dammara-12(13),24-dien-3-ol, 32: 20R-dammara-12(13),24-dien-3-ol, 33: (20R,
24R)-17-epi-epoxydammarane, 34: polypoda-8(9),13,17,21-tetraen-3-ol, 35:
polypoda-7(8),13,17,21-tetraen-3-ol (3-hydroxy--polypodatetraene), 37: polypoda-
7(8),13,17,21-tetraen-3-ol (3-hydroxy--polypodatetraene).
7
8
9
10
11
12
13
14
15
16
17
27
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Scheme 2. Cyclization mechanisms of 1, 11, and 14 leading to bicyclic products 17, 19,
28, and 34–36. Substrates 1 and 11 fold in a chair-chair conformation, but substrate 14
folds in a boat-chair structure.
28
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800281
ChemBioChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Scheme 3. Cyclization mechanisms of 1 and 14 leading to tricyclic product 20 and
pentacyclic products 25, 26, and 33. (A) A chair-chair-boat conformation yielded the
tricyclic 20. (B) The all-chair conformation furnished the neohopane scaffold 25 and
hopane skeleton 26. (C) (3R)-oxidosqualene 14 was folded into an all-chair
conformation to yield 20R, 24R-17-epi-epoxydammarane 33. Notably, cyclization
started from the isopropylidene moiety rather than from the epoxide ring.
29
This article is protected by copyright. All rights reserved.