Squalene-Hopene Cyclase: On the Polycyclization Reactions of Squalene Analogues Bearing Ethyl Groups at Positions C-6, C-10, C-15, and C-19

Article abstract of DOI:10.1002/ejoc.201800010

Squalene-hopene cyclase (SHC) has been found to convert acyclic squalene into 6,6,6,6,5-fused pentacyclic triterpenes hopene and hopanol. The enzymatic reactions of squalene analogues bearing ethyl groups in lieu of methyl groups at positions C-6, C-10, C-15, and C-19 have been examined to investigate whether the larger ethyl substituents (a C₁ unit increment) are accepted as substrates and to investigate how these substitutions affect polycyclization cascades. Analogue 6-ethylsqualene 19a did not cyclize, which indicates that substitution with the bulky group at C-6 completely inhibited the polycyclization reaction. In contrast, 19-ethylsqualene 19b afforded a wide spectrum of cyclization products, including mono-, bi-, tetra-, and pentacyclic products in a ratio of 6:6:1:2. The production of tetra- and pentacyclic scaffolds suggests that the reaction cavity for D-ring formation site is somewhat loosely packed and can accept the 19-ethyl group, and that a robust hydrophobic interaction exists between the 19-ethyl group and the binding site. In contrast to 19b, 10-ethylsqualene 20a and 15-ethylsqualene 20b afforded mainly mono- and bicyclic products, that is, the polycyclization cascade terminated prematurely at the bicyclic reaction stage. Therefore, the catalytic domains for the 10- and 15-methyl binding sites are tightly packed and cannot fully accommodate the Et substituents. The cyclization pathways followed by the ethyl-substituted substrates in the presence of SHC and lanosterol and β-amyrin synthases are compared.

Full text of DOI:10.1002/ejoc.201800010

A Journal of
Accepted Article
Title: Squalene-hopene cyclase: on the polycyclization reactions of
squalene analogs bearing ethyl groups at positions C-6, C-10,
C-15 and C-19
Authors: kazunari Takahashi, Yusuke Sasaki, 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: Eur. J. Org. Chem. 10.1002/ejoc.201800010
Link to VoR: http://dx.doi.org/10.1002/ejoc.201800010
Supported by

10.1002/ejoc.201800010
European Journal of Organic Chemistry
Squalene-hopene cyclase: on the polycyclization reactions of squalene
analogs bearing ethyl groups at positions C-6, C-10, C-15 and C-19
Kazunari Takahashi,＋Yusuke Sasaki＋and Tsutomu Hoshino*
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
*Corresponding author. Prof. Dr.
E-mail address: hoshitsu@agr.niigata-u.ac.jp
http://www.agr.niigata-u.ac.jp/~satot/index.html
[+] These authors contributed equally to this work.
Supporting Information for this article is available under WWW. https······
1
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
Abstract
Squalene-hopene cyclase (SHC) converts acyclic squalene into the 6,6,6,6,5-fused
pentacyclic triterpene hopene and hopanol. Enzymatic reactions of squalene analogs
bearing ethyl groups in lieu of methyl groups at positions C-6, C-10, C-15 and C-19
were examined in order to investigate whether the larger ethyl-substituents (C1-unit
increment) are accepted as substrates and to investigate how these substitutions affect
polycyclization cascades. Analog 6-ethylsqualene 19a was not cyclized, indicating that
substitution with the bulky group at C-6 completely inhibited the polycyclization
reaction. In contrast, 19-ethylsqualene 19b afforded a wide spectrum of cyclization
products, including mono-, bi-, tetra-, and pentacyclic products in a ratio of 6:6.3:1:2.
Production of tetra- and pentacyclic scaffolds suggests that the reaction cavity for the D-
ring formation site is somewhat loosely packed and can accept the 19-ethyl group, and a
robust hydrophobic interaction exists between the 19-ethyl group and the binding site.
In contrast to 19b, 10-ethylsqualene 20a and 15-ethylsqualene 20b afforded mainly
mono- and bicyclic products, that is, the polycyclization cascade terminated prematurely
at the bicyclic reaction stage. Therefore, the catalytic domain for the 10- and 15-methyl
binding sites are tightly packed and cannot fully accommodate the Et substituents. The
cyclization pathways of ethyl-substituted substrates are compared between SHC and
lanosterol and -amyrin synthases.
Introduction
The enzymatic cyclization of acyclic squalene 1 and 2,3-oxidosqualene to form
polycyclic triterpenes have fascinated chemists and biochemists for over half a
century.[1, 2 ] The X-ray crystallographic structure of the squalene-hopene cyclase from
Alicyclobacillus acidocaldarius (AaSHC) was reported in 1997.[3] AaSHC catalyzes the
conversion of 1 into the pentacyclic triterpenes hopene 2 and hopanol 3 at a ratio of ca.
5:1 (Scheme 1).[2a, 2b] This polycyclization cascade is attained by a single enzyme.
Compound 1 is folded in all pre-chair conformations inside the reaction cavity and
cyclized in a regio- and stereospecific fashion via a series of carbocationic
intermediates, leading to the formation of five new C-C bonds and nine chiral centers.
Studies on AaSHC have revealed that the polycyclization reaction consists of 8 reaction
steps (Scheme 1): [2b] (1) first, cyclization to form A-ring 4 by proton attack on the
terminal double bond, donated by the DXDD motif;[4] (2) second, ring closure to yield
the B-ring (6/6-fused A/B ring system 5);[5,6] (3) third, cyclization to yield the 5-
membered C-ring (6/6/5-fused A/B/C-tricyclic ring system 6) by Markovnikov
closure,[6, 7] which then undergoes (4) ring expansion to form the 6-membered C-ring
2
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
(6/6/6-fused tricyclic ring system 7);[7] (5) fifth, cyclization to yield the
thermodynamically favored 5-membered D-ring (6/6/6/5-fused A/B/C/D ring system 8,
17-epi-dammarenyl cation) [8-10] followed by (6) the second ring enlargement process to
form the 6-membered D-ring (6/6/6/6-fused A/B/C/D-ring system, prohopanyl cation 9)
[8-12] and (7) the last ring closure process to construct the 6/6/6/6/5-fused A/B/C/D/E-
ring system (10, hopanyl cation),[12] and (8) the final deprotonation reaction of 10 to
introduce the double bond.[13] Recently, the Hauer group has succeeded in creating the
AaSHCs with novel catalytic activities by site-directed mutagenesis, which have
enabled the cyclization reactions of truncated squalene and terpene-like analogs to yield
unnatural products.[2i]
<Scheme 1>
We have reported the polycyclization reactions of various norsqualene analogs by
the native AaSHC; in these analogs, the methyl positions of 1 are substituted with
hydrogen atoms. Squalene 1 is a symmetrical molecule and possesses two
isopropylidene moieties at both terminal positions. We have revealed that the
isopropylidene moiety is essential for initiating polycyclization by AaSHC. [13] In
previous paper, we reported the cyclization products of 26-norsqualene 11. [14] This
compound possesses isopropylidene moieties at both left and right termini, and thus,
two cyclization pathways are possible, i.e., from the left (path a) and right side (path b).
Therefore, two chemical names were proposed for this compound based on the
cyclization pathway: 26-norsqualene 11a for path a and 29-norsqualene 11b for path b,
as shown in Scheme 2A. Substrate 11a afforded 12, and 11b afforded 13 and 14. [14] The
structures of all the products indicate that both 11a and 11b are folded in a normal
folding conformation, as shown in Scheme 1. Thus, the substitutions of the less-bulky
hydrogen atom at C-6 and C-19 had little influence on the polycyclization cascade. On
the other hand, 27-norsqualene 12a generated the unprecedented triterpene skeletons 15
and 16, which consist of 6/5+5/5 and 6/5+5/5+6 ring systems, respectively. [15] The
structures of 15 and 16 indicate that 12a underwent the unusual folding conformation,
as depicted in Scheme 2. In contrast, 28-norsqualene 12b afforded the pentacyclic
hopane skeletons 17 and 18 as the enzymatic products. Product 17 was produced from
the all-chair conformation, as shown in Scheme 1, but product 18 was generated from a
chair/chair/chair/chair/boat folding conformation.[15] The results from 12a and 12b
indicated that the decrease in steric size at C-10 of 1 had a greater effect on the folding
conformation than that at C-15.
3
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
<Scheme 2>
However, there is no report regarding the cyclization reactions of squalene analogs
19 and 20, which bear larger substituents (ethyl moieties, C1-increment) at the C-6 or C-
19 and the C-10 or C-15 positions. Analog 19 has an Et group at C-6 or C-19 (19a or
19b, respectively) based on the cyclization pathways a and b, respectively. Homolog 20
possesses Et groups at C-10 or C-15 (20a or 20b, respectively). Analogs 19 and 20 were
synthesized and incubated with AaSHC. Introduction of the Et moiety at C-6 (19a) gave
no enzymatic product, while that at C-19 (19b) afforded mono-, bi-, tetra- and
pentacyclic scaffolds. Replacement with an Et group at C-10 (20a) and C-15 (20b) led
to the production of only mono- and bicyclic compounds with the exception of the
generation of a tricyclic product from 20a. Herein, we report the detailed experimental
results and discuss the mechanism of the cyclization of 19 and 20, which exhibit larger
steric volumes (C1-increment). This investigation provides important information on the
molecular recognition mechanism between AaSHC and squalene substrate, and the
differences in the polycyclization results between lanosterol and -amyrin synthases are
discussed.
Results
Syntheses of ethyl-substituted analogs 19 and 20
Squalene 1 was treated with mCPBA (m-chloroperbenzoic acid) in CH2Cl2 to
produce an epoxide mixture, which was then treated with H5IO6, yielding a mixture of
C17 21, C22 22, and C27 aldehydes.[13, 16] Scheme 3 shows the synthetic designs of 19 and
20. Synthetic intermediates of 23 were prepared from aldehyde 21 according to the
published protocol.[17] Aldehyde 21 was subjected to a Wittig-Horner reaction using
triethyl 2-phosphonobutyrate to yield an E/Z-mixture of alkenes, which were separated
by SiO2 column chromatography to isolate the E-isomer. The ethyl ester thus obtained
was reduced with DIBAL-H to yield the corresponding alcohol, which was then
converted to bromide 23 using PBr3. Geraniol 24 was treated with PBr3 to yield geranyl
bromide. The solution of bromide in DMF was slowly added to sodium
benzenesulfinate in DMF, yielding geranyl phenylsulfone 25. Compounds 23 and 25
were subjected to a coupling reaction using n-BuLi, yielding the phenylsulfone
derivative of the ethyl-substituted analog 26. To remove the phenylsulfonyl group, 26
was treated with Super-Hydride reagent to synthesize the required C31 analog 20. To
synthesize analog 19, aldehyde 22 was subjected to the same reactions as the
preparation of 23 from 21, affording bromide 26. Commercially available 3-methylbut-
4
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
2-en-1-ol was transformed to the bromide and then to the corresponding phenylsulfone
derivative 27 by using the same reactions as the preparation of 25 from 24. The
coupling reaction of 26 with 27 and the subsequent removal of the phenylsulfone group
were carried out, yielding the C31 analog 19. The synthetic experiments and the NMR
data of the synthetic intermediates and Et-substituted analogs 19 and 20 are described in
the Supporting Information (Fig. S1).
<Scheme 3>
Enzymatic reactions of 19 and 20 with native AaSHC
A buffered reaction mixture (2.0 mL, pH 6.0, 60 mM citrate buffer) consisting of
Triton X-100 (0.2%, w/v), 1 [200 g, 19 or 20], and purified histidine-tagged pET26b
AaSHC (5 g)[18] was incubated at 55 °C for 6 h.[18,19] Then, 15% KOH/MeOH (7.5 mL)
was added, and the resulting mixture was heated to 70–80 °C for 30 min. The lipophilic
materials were extracted with hexane, and geranylgeraniol (5 g) was added to the
hexane extract as the internal standard. Triton X-100 present in the extract was removed
by elution through an SiO2 column (hexane/EtOAc, 100:10), and the eluent was
evaporated to dryness. Then, hexane (150 L) was added to the residue, and the
resultant solution (1.0 L) was subjected to GC analysis (GC: gas chromatography) to
quantify the lipophilic materials.
Fig. 1 shows the gas chromatograms of the hexane extracts obtained by incubating
1 (A), 19 (B) and 20 (C) with the purified His-tagged AaSHC. In Fig. 1A, no peak
corresponding to substrate 1 was observed (A), indicating that 1 was completely
converted to 2 and 3 under the incubation conditions described above. Incubation of 19
afforded a broad product spectrum, as shown by the products numbered 28─33;
however, a significantly large peak corresponding to substrate 19 was observed (B),
indicating that the conversion of 19 was markedly decreased relative to that of 1. Fig.
1C shows that substrate 20 also exhibited significantly low conversion (a large peak
corresponding to substrate 20) despite many products 34─42 having been generated.
Consequently, substitution of the bulkier moiety (increment of C1-unit) at positions C-6,
C-10, C-15 and C-19 resulted in significant perturbation of the polycyclization reaction.
<Fig. 1>
Isolation and structures of enzymatic products 28─33 from substrate 19
To isolate products and determine their structures, we cultured Escherichia coli
BL21(DE3) harboring pET3a-AaSHC and used the cell-free extracts for large scale
incubation experiments.[19] Substrate analog 19 (27 mg) was incubated with the cell-free
extract (60 mL), which was prepared from a 1.2-L culture, at optimal conditions (60 C,
5
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
pH 6.0) for 16 h. [19] After saponification with a solution of 15% KOH/MeOH, the
lipophilic materials were extracted with hexane, and then, the Triton X-100 was
removed from the hexane extract by a short SiO2 column chromatographic purification
(hexane:EtOAc=100:20). Careful SiO2 column chromatography (activated by heating
SiO2 at 170 C for 3 h), eluting with hexane, afforded 28, 30 and 32 in pure states. A
fraction containing 29 and impurities was subjected to normal phase HPLC with 100%
hexane to afford pure 29. Fractions enriched with 31 and 33 were subjected to
argentation SiO2 column chromatography (5% AgNO3, 100% hexane), which afforded
pure 31 and 33. Final purifications of all products were performed by normal phase
HPLC (100% hexane).
Substrates 19 and 20 possess seven methyl groups and one ethyl group on the
squalene backbone. In the 1H-NMR spectrum of product 28 (400 MHz, CDCl3), four
vinylic Me groups were found at H (ppm) 1.59 (3H, s), 1.61 (6H, s, Me x 2) and 1.68
(3H, s), and one olefinic Et group was observed at H 0.959 (3H, t, J=7.6 Hz, Me-31)
and 2.03 (2H, m, H-28), suggesting that 28 is a monocyclic product and that the Et
group is present on the double bond of 19. Two aliphatic Me groups (H 0.911, 3H, s for
Me-23; 0.833, 3H, s for Me-24) and one methylidene group (H 4.54, 1H, s and H 4.75,
1H, s for H-25) were also detected. Distinct HMBC cross peaks were observed for Me-
23/Me-24/C-5 (C 53.6, d)/C-3 (C 36.4, t)/C-4 (34.9, s) and for H-25/C-5/ C-6 (C
149.4, s). Detailed analyses of 2D NMRs, including 1H-1H COSY, HOHAHA, NOESY,
HSQC and HMBC spectra (Figs. S2.2─S2.8, Supporting Information), in addition to
DEPTs data demonstrated that the structure of 28 is as depicted in Fig. 2. Complete
assignments of 1H- and 13C-NMR spectra are shown in Fig. S2.9.
In the 1H-NMR spectrum of product 29 (600 MHz, C6D6), five olefinic Me groups
and one olefinic Et group were detected in addition to five olefinic protons (H 5.38, t,
J=6.8 Hz, 2H, H-17 & H-21; H 5.47, 2H, s, H-10 & H-13; 5.59, 1H, s, H-3). Since
seven olefinic Me groups and one Et group, in addition to six olefinic protons, are
involved in substrate 19, the cyclization product of 19 is likely 29. Indeed, two aliphatic
Me groups were found at H 0.979 (3H, d, J=6.8 Hz, Me-25) and 1.01 (3H, s, Me-24),
suggesting that 29 is a monocyclic product with one double bond in the ring. Olefinic
Me-23 (H 1.79, 3H, s) had clear HMBC cross peaks for C-3 (C 124.5, d), C-4 (C
139.7, s) and C-5 (C 40.7, s), and Me-24 showed distinct HMBC cross peaks for C-4,
C-5 and C-6 (C 33.5, d). Detailed HMBC analyses (Fig. S3.9) showed that 29
possesses a 1,5,6-trimethylcyclohexene moiety (IUPAC numbering, Fig. S3.10B). This
ring system was further confirmed by electron impact mass spectrometry (EIMS), with
fragment ion m/z 123 as a base peak (Fig. S3.1). The relative stereochemistry of the
6
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
cyclohexane ring was determined by the NOESY spectrum (Fig. S3.6). Distinct NOEs
of Me-24/H-6 and Me-25/H-7 were observed, but no NOE was observed for H-6/H-7 or
for Me-25/Me-24 (see Figs. S3.9 and S3.10A). These findings strongly indicate that the
relative configurations of positions 5 and 6 of the cyclohexene ring all possess R
stereochemistry, indicating that substrate 19 was folded in a boat conformation but not
in a chair conformation (see Fig. S3.10A and Scheme 4E). Thus, the structure of product
29 was determined as shown in Fig. 2 and in Figs. S3.9 and S3.10. We have reported
that this 1,5R,6R-trimethylcyclohexene ring (IUPAC numbering, Fig. S3.10B) was
produced by incubating 1 with the L607F SHC variant. We have proposed the name
neoachillapentaene for the structure of the fundamental skeleton of 29.[19c] This unusual
product with the same relative stereochemistry as 29 was also obtained from the
incubation experiment of 10-ethyl-2,3-oxidosqualene with hog-liver lanosterol cyclase
(see 44 shown in Fig. S14.1).[20] The absolute stereochemistry at positions 5 and 6
should be verified by other methods different from NMR.
The 1H-NMR spectrum of product 30 (400 MHz, CDCl3) showed that three olefinic
Me groups were identified: H 1.57 (3H, s, Me-27), 1.60 (3H, s, Me-30) and1.68 (3H, s,
Me-29). In addition, one olefinic Et group was observed: H 0.963 (3H, t, J=7.6 Hz, Me-
31) and 2.03 (2H, m, H-28). Three aliphatic methyl groups were also identified: H
0.666 (3H, s, Me-25), H 0.800 (3H, s, Me-24) and H 0.868 (3H, s, Me-23). In addition,
one methylidene group was observed: H 4.54 (1H, s, H-26) and 4.82 (1H, s, H-26).
These findings suggest that 30 is a bicyclic compound, which was further confirmed by
the following HMBC correlations (Fig. S4.9): H-26/ C-8 (C 148.8, s) and H-26/C-9 (C
56.2, d), Me-25/C-9, Me-23/C-5 (C 55.6, d), Me-24/C-5, Me-25/C-5. A strong NOE
was observed between H-5 (H 1.08, 1H, dd, J=12.8, 2.8 Hz) and H-9 (H 1.61, 1H, m).
Complete NMR assignments are shown in Fig. S4.9. The structure of 30 thus
determined is shown in Fig. 2 and Fig. S4.9.
Product 31 showed two olefinic Me groups (600 MHz, C6D6): H 1.84 (3H, s, Me-
26) and 1.74 (3H, s, Me-27). An Et group was connected to C-20 (C 39.8, d), which
was determined by the HMBC correlation between Me-31 (H 0.959, 3H, t, J=7.1 Hz)
and C-20. The C-20 signal appeared at significantly higher field than that (C 140.8, s)
of the corresponding position of 19. Thus, the Et group is not present on the double
bond of the squalene backbone, indicating that olefinic Et changed to aliphatic Et. The
presence of the Et group on C-20 was further confirmed by the following HOHAHA
connectivity: Me-31/H-21 (H 1.46, 1H, m; 1.51, 1H, m)/H-20 (H 2.42, 1H, m). Thus,
product 31 is a tetracyclic compound. Strong HMBC cross peaks were observed
between H-21 and C-17 (C 132.7, s), and between Me-30 (H 1.31, 3H, s) and C-13 (C
7
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
142.0, s), indicating that the double bond is situated at C-13 and C-17. Further analysis
of the HMBC spectrum (Figs. S5.8 and S.5.9) allowed us to propose the fundamental
structure of 31 as dammara-13(17), 24-diene. Unambiguous NOEs were observed for
Me-28 (H 1.04, 3H, s)/H-5 (H 0.940, 1H, m), H-5/H-9 (H 1.60, 1H, m) and H-9/Me-30
(H 1.31, 3H, s), and thus, the complete structure, including stereochemistry, can be
proposed as shown in Fig. 2 and Fig. S5.9, except for the C-20 stereochemistry.
The 1H-NMR spectrum (400 MHz, CDCl3) of product 32 showed four olefinic Me
groups:H 1.60 (3H, s, Me-30), 1.61 (3H, s, Me-27), 1.68 (3H, s, Me-29), 1.71 (3H, s,
Me-26). In addition, the spectrum showed one olefinic Et group: H 0.980 (3H, t, J=6.8
Hz, Me-31) and 2.03 (2H, m, H-28). Furthermore, three aliphatic Me groups were found
at H 0.852 (3H, s, Me-23), 0.874 (3H, s, Me-24) and 0.741(3H, s, Me-25). These data
suggest that 32 may be monocyclic. However, the HMBC cross peaks were clearly
observed for Me-25/C-9 (C 54.2, d), Me-26/C-9, Me-26/C-7 (C 122.0, d) and Me-
26/C-8 (C 135.6, s), indicating that 32 was indeed a bicyclic compound and that double
bond is situated at positions C-7 and C-8. Detailed analysis of 2D NMR spectra
revealed that the complete structure of 32 is as shown in Fig. 2 and Fig. S6.9. Structural
difference between 30 and 32 is found only at the double bond position.
Product 33 possessed only one double bond, which was observed at H 5.11 (1H, s,
H-29) and H 5.07 (1H, s, H-29) in the 1H-NMR (600 MHz, C6D6, Fig. S7.2) spectrum
and C 109.9 (t, C-29) and C 148.9 (s, C-22) in the 13C-NMR (150 MHz, C6D6, Fig.
S7.3) spectrum. The two olefinic protons (H-29) were correlated with C 109.9 (t) in the
HSQC spectrum (Fig. S7.7) and had HMBC cross peaks for C-22 (Fig. S7.8). Thus, the
double bond is assigned to a methylidene group. No other double bond was observed.
This finding definitively demonstrated that product 33 was produced by the complete
cyclization reaction, suggesting that 33 has a hopene scaffold. The detailed 2D NMR
analyses are shown in Fig. S7.9. In the HMBC spectrum, H-21 (H 2.72, 1H, m) had
clear cross peaks with the following carbons: C-22, C-29, C-30 (C 25.7, q), C-17 (C
56.3, d), and C-20 (C 28.1, t). H-21 exhibited unambiguous HOHAHA correlations for
the following protons: H-17/H-16/H-15/H-20/H-19 (Fig. S7.5). The HMBC and
HOHAHA data strongly indicated that 33 possesses a hopene skeleton with a 5-
membered E-ring. An Et group is attached to the C-18 position, as the Me protons (H
1.02, 3H, t, J=7.4 Hz, Me-31) displayed a strong HMBC cross peak for C-18 (C 47.4,
s). Taking NOE data into consideration, the whole structure of product 33, including the
stereochemistry, can be proposed as shown in Fig. 2 and Fig. S7.9.
<Fig. 2>
8
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
Isolation and structures of enzymatic products 34─42 from substrate 20
Substrate analog 20 (40 mg) was incubated with the cell-free extract (100 mL) at
optimal conditions (60C, pH 6.0) for 16 h. The extraction of the lipophilic materials
from the incubation mixture was carried out with the same method that was used for the
incubation of 19. By careful SiO2 column chromatography with a step-wise elution
(hexane/EtOAc, 100:0–100:5), four partially purified fractions were obtained. Each
fraction was further purified by 5% or 10% AgNO3-SiO2 column chromatography
(hexane/EtOAc, 100: 0─100:3). Final purification was conducted by normal phase
HPLC (100% hexane). Complete separation of 34 and 35, 36 and 38 and 39 and 40 was
unsuccessful. Therefore, each mixture of two compounds was directly subjected to
NMR analyses without further purification.
Fig. S8.1 shows the GC-EIMS results of a mixture of 34 and 35, which exhibited
identical fragment patterns, suggesting that the two products have similar structures and
that the ratio of the yields of 34 to 35 was ca. 1 to 1.5. The 1H-NMR spectrum (600
MHz, C6D6) is shown in Fig. S8.2. Two aliphatic Me groups were found with high
intensity peaks: H 0.974 (3H, s, Me-24) and 1.08 (3H, s, Me-23). In addition, an Et
group was identified (H 1.11, 3H, t, J=5.2 Hz, Me-31;H 2.20, 2H, m, H-26). These
intense proton signals are accompanied by similar proton signal patterns but with lower
intensities by ca. 1/1.5; the lower signals were as follows: H 0.983 (3H, s, Me-24), 1.08
(3H, s, Me-23), and H 1.14 (3H, t, J=5.2 Hz, Me-31), and 2.23 (2H, m, H-27) for the Et
group (see Fig. S8.2B). Thus, we were able to differentiate the 1H-NMR signals of 35
(higher signals) from those of 34 (lower signals). Methylidene protons (H 4.99, s; 4.82,
s: H-25), which were correlated with C-25 (C 109.3, t) in the HSQC spectrum and were
not distinguishable between 35 and 34, showed clear cross peaks with C-6 (C 149.5, s,
for both 35 and 34), C-1 (C 32.8, t, for both 35 and 34) and C-5 (C 53.97, d, for 35; C
54.16, d, for 34). The 13C-NMR spectrum (expanded) is shown in Figs. S8.3B and
S8.3C. Both Me-23 and Me-24 of 35 and 34 had strong HMBC correlation with C-3 (C
36.6, t, for both 35 and 34). The COSY and HOHAHA spectra revealed the 1H-1H spin
couplings of H-1/H-2/H-3, as shown in Fig. S8.9. Further, the 1H-1H spin coupling
networks H-5/H-7/H-8 were found from the HOHAHA spectrum. Observation of the
strong HMBC correlation between Me-26 (H 1.77, s) and C-8 (C 35.6, t), which was
found only for 34, indicated that 34 has a Me group at C-9 (C 135.8, s). Thus, 34 and
35 are monocyclic compounds, and the complete structures of 35 and 34 are as depicted
in Fig. 2. The detailed assignments are shown in Fig. S8.9. As described below, the Et
group of product 35 is positioned at C-9 (C 141.1, s), based on the cyclization pathway
(Scheme 4).
9
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
The GC-EIMS spectrum of the mixture of products 36 and 38 is shown in Fig.
S9.1. The two products showed similar fragment patterns, which is suggestive of the
similar structures of these two products. The yield of 38 was significantly higher than
that of 36. The ratio was estimated to be ca. 4:1 by integration of the Me-25 signals in
the 1H-NMR spectrum (600 MHz, CDCl3, see Fig. S9.2A and B), and thus, we could
easily differentiate the signals of 38 from those of 36. Three aliphatic Me protons and
four vinylic Me protons were found in the 1H-NMR spectrum of 38 (see Fig. S9.2B): H
0.615 (3H, s, Me-25), 0.786 (3H, s, Me-24), 0.867 (3H, s, Me-23); H 1.56 (3H, s, Me-
27), 1.60 (6H, s, Me-28 and Me-30), 1.68 (3H, s, Me-29), but the Et signal of substrate
20 was not observed. Instead, a doublet Me proton signal (H 1.65, d, J=6.6 Hz, Me-31),
which was correlated with the olefinic proton (H 5.04, 1H, q, H-26) in the COSY
spectrum, appeared, definitively indicating the presence of a CH3CH= group in 38. The
following 1H-1H spin coupling networks were revealed by analyzing the COSY and
HOHAHA spectra: H-5 (H 1.08, dd, J=8.4, 2.0 Hz)/H-6 (H 1.20, m; 1.70, m)/H-7 (H
1.56, m; 2.78, m). Olefinic H-26 had distinct HMBC cross peaks for C-7 (C 30.4, t) and
C-9 (C 57.0, d). Me-25 also showed a strong HMBC contour peak for C-9. The
expanded spectrum of the 13C-NMR is shown in Fig. S9. 3B. Thus, 38 is a bicyclic
compound as shown in Fig. 2 and Fig. S9.9. In contrast, 36 showed the presence of
methylidene protons (H 4.82, s; 4.53, s, CH2-26) with ca. one-fourth of the intensity of
H-26 of 38. Furthermore, 36 had an olefinic Et group (H 0.948, t, J=7.8 Hz, Me-31 and
H 1.99, m, CH2-27). Thus, the structure of 36 could be proposed as shown in Fig. 2 and
Fig. S9.9. The detailed assignments are depicted in Fig. S9.9.
The 1H-NMR spectrum (600 MHz, CDCl3, see Fig. S10.2) showed that the isolated
37 was a single product (not a mixture of two products, as observed for 34+35 and for
36+38). Three olefinic Me groups were found:H 1.68 (3H, s, Me-29), 1.604 (3H, s,
Me-30) and 1.559 (3H, s, Me-28). Three aliphatic Me groups were observed:H 0.905
(3H, s, Me-24), 0.869 (3H, s, Me-23) and 0.734 (3H, s, Me-25). In the HMBC spectrum
(Fig. S10.8), the following correlations were observed: Me-23/Me-24/C-3 (C 42.8,
t)/C-4 (C 32.9, s)/C-5 (C 50.7, d) and Me-25/C-1(C 40.4, t)/C-5/C-9 (C 58.8, d)/C-10
(C 34.9, s). Olefinic proton H-7 (H 5.20, 1H, brs) had a distinct HMBC contour with
C-9, and further, this proton clearly showed the following COSY and HOHAHA
correlations: H-7/H-6 (H 1.88, 1H, m; 2.08, m)/H-5 (H 1.22, 1H, m). The H-9 (H
1.95, m) also had HOHAHA cross peaks for H-11 and H-12. These data suggest that the
double bond is situated at C-7 and C-8. The proton signals of an Et moiety were found:
H 0.772 (3H, t, J=7.4 Hz) for Me-31; H 1.35 (1H, m) andH 1.52 (1H, m) for H-26.
However, Me-31 had strong HMBC correlations with C-13 (C 50.7, s) and C-26 (C
10
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
30.4, t). The chemical shift of C-13 definitively demonstrates that the Et group of 37 is
not placed on the double bond that is seen in substrate 20. The doublet Me group (H
0.823, 3H, d, J=6.7 Hz, Me-27) showed a clear HMBC cross peak for C-13.
Consequently, product 37 is a 6,6,5-fused tricyclic compound with a double bond at C-7
and C-8; the double-bond position was further established by definitive HMBC contour
of H-26 with C-8 (C 147.1, s). Distinct NOEs were found for H-5/H-9 and H-7/H-14
(H 1.38, 1H, m), indicating that the Et group is -oriented, and the side chain is -
configured. Therefore, the complete structure of 37 is as depicted in Fig. 2 and Fig.
S10.9.
Products 39 and 40 were not distinguishable by the GC analysis (Fig. 1), and thus,
the yield ratio of 39 and 40 was not determined from the GC chromatogram. The 1H-
NMR spectrum of a mixture of 39 and 40 (600 MHz, CDCl3) showed that the yield ratio
was ca. 1:1 (Fig. 11.2B). Two Et groups were found in a 1:1 ratio; the proton signals of
H 0.996 (3H, t, J=7.8 Hz, Me-31) and H 2.02 (2H, m, H-27) were observed for the Et
group of 39; H 1.01 (3H, t, J=7.2 Hz, Me-31) and 2.03 (2H, m, H-26) were assigned to
the Et group of 40. A pair of three aliphatic Me groups were also observed in a 1:1 ratio:
H 0.871 (s, Me-24), 0.849 (s, Me-23) and 0.740 (s, Me-25) for 39; H 0.878 (s), 0.857
(s, Me-23) and 0.741 (s, Me-25) for 40, but these assignments may be exchangeable
between 39 and 40. The signals of Me-23, Me-24 and Me-25 were assigned by HSQC
and HMBC analyses (Figs. S11.7 and S11.8), as depicted in Fig. S11.9. The HMBC
cross peaks from Me-23, Me-24 and Me-25 also enabled the unambiguous assignment
of C-1, C-3, C-5 and C-9. The COSY and HOHAHA analyses (Fig. S11.4 and S11.5)
revealed the presence of the following cross peaks: H-5/H-6/H-7 (H 5.38, brs); H-7 is
an olefinic proton. Definitive NOEs of H-7/Me-26 (H 1.71, s) for 39 and H-7/Me-31
(H 1.01, t, J=7.2 Hz) for 40 (Fig. S11.6) gave important information for differentiating
39 from 40, that is, Me-26 of 39 and the Et group of 40 are positioned at C-8, which
was further supported by the distinct HMBC correlations of Me-26/C-7 (C 122.1, d)/C-
8 (C 135.6, s)/C-9 (C 54.4, d) for 39 and CH2-26 (H 2.03, m, 2H)/C-7(C 119.7, d)/C-8
(C 141.2, s) for 40. Me-25 of 39 and 40 had strong HMBC correlations with C-9 of 39
and 40. Consequently, 39 and 40 were determined to be bicyclic products as shown in
Fig. 2. Structural differences between 39 and 36 and between 40 and 38 were found
only at the double-bond positions.
Fortunately, product 41 was successfully isolated in a pure state, albeit at a low
yield (Fig. 1). SiO2 TLC chromatography showed that 41 was a highly polar compound;
the Rf value was ca. 0.4 with the mixed solvent (hexane/EtOAc,100:10), but no
movement was observed with hexane (100%), while products 34 and 35 showed Rf
11
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
values of ca. 0.6 with hexane (100%). The 1H-NMR spectrum of 41 (600 MHz, C6D6,
see Fig. S12.2) showed the presence of three aliphatic Me and four olefinic Me groups:
H 0.872 (3H, s, Me-24), 1.06 (3H, s, Me-23), 1.20 (3H, s, Me-25), 1.69 (3H, s, Me-30),
1.74 (3H, s, Me-28), 1.81 (3H, s, Me-29) and 1.85 (3H, s, Me-26). In addition, the
presence of an Et group was observed: H 1.12 (3H, t, J=7.4 Hz, Me-31) and 2.19 (2H,
q, J=7.4 Hz, H-27), indicating that 41 is a monocyclic compound and that the Et group
is situated on the double bond of the squalene backbone. The 13C-NMR spectrum (150
MHz, C6D6, see Fig. S12.3) indicated that a quaternary alcoholic carbon (C 73.5, s, C-
6) is present in 41. Me-25 showed distinct HMBC cross peaks for C-1 (C 43.9, t), C-5
(C 56.9, d) and C-6. Me-23 and Me-24 had strong HMBC correlations with C-3 (C
41.8, t), C-4 (C 35.6, s) and C-5. The H-5 (H 1.22, 1H, m) was assigned by the HSQC
cross peak from C-5. In the COSY and HOHAHA spectra (Fig. S12.4 and S12.5B), the
following 1H-1H connectivity was found: H-5/H-7 (H 1.56, 1H, m; 1.85, 1H, m)/H-8
(H 2.33, 1H, m; 2.46, 1H, m). Strong HMBC cross peaks for Me-26/C-8 (C 43.3, t)
and Me-26/C-9 (C 136.5, s) provided important information that the substituent at C-9
is an Me group and not an Et group, as depicted in Fig. 2 and Fig. S12.9. The detailed
analyses, including 2D NMR analyses, are shown in Fig. S12.9.
Production yield of 42 was also trivial (Fig. 1). Product 42 was a highly polar
compound (same Rf value as 41 by SiO2 TLC), indicating that 42, like 41, has a
hydroxyl group. In spite of careful purification, the isolation of 42 was, unfortunately,
unsuccessful; sample loss occurred during several purification steps, leading to a low
yield. However, the EIMS spectrum of 42 was identical to that of 41, including the
fragment patterns (Fig. S13), which suggests that the structure of 42 is as shown in Fig.
2. This idea is further supported by the finding that the EIMS of 34 was identical to that
of 35, although the structure of 34 is different from that of 35 with respect to the Et-
substituent position. Moreover, taking the cyclization mechanism into consideration, 42
has Et and Me groups at C-9 and C-14, respectively, while the structure of 41 was
established by the NMR analyses to have Et and Me groups at C-14 and C-9,
respectively, as described above (see Fig. 2).
Discussion
Cyclization pathways of 19a, 19b, 20a and 20b by AaSHC
As shown in Fig. 2, no cyclization products from 19a was observed, or if they were
present they were below the limit of detection, thus indicating that the introduction of a
larger Et group at C-6 prevented the polycyclization reaction. In contrast, the analog
12
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
that was Et-substituted at C-19 (19b) underwent cyclization reactions to yield a wide
spectrum of mono-, bi-, tetra- and pentacyclic products. The cyclization mechanisms are
depicted in Scheme 4. The cyclization reaction of 19b halted at the monocyclic cation
(4’), the structure being analogous to cation 4. The proton elimination from Me-25 of 4’
(path c) afforded product 28. The same deprotonation reactions of cation 4’ (path c)
from 20a and 20b gave products 35 and 34, respectively. Attack by a water molecule on
cation 4’ yielded 41 and 42 (path d). Further cyclization of 4’ furnished the bicyclic
cation 5’ (path e), which was structurally similar to cation 5, which underwent
deprotonation at Me-26 (path f), yielding products 30 and 36 from 19b and 20b,
respectively. Deprotonation of H-7 of cation 5’ afforded 32 and 39 from 19b and 20b,
respectively. Analog 20a, with the bulky Et group at C-10, also underwent further
cyclization to give the bicyclic cation 5’’. Deprotonation of the Et group (path f) and
that of H-7 (path g) generated products 38 and 40, respectively. Further cyclization of
5’’ afforded the 6,6,5-fused tricyclic cation 6’’ (path h). The successive 1,2-shift
reactions occurred as follows (path i): H-13cation C-14, Et at C-8  C-13, and
deprotonation of H-7. The Et and H-7 are arranged in anti-parallel fashion, thus the
deprotonation reaction occurs from H-7, but not from H-7. Notably, the tricyclic
product 37 was generated only from 20a and not from 20b, that is, 20b yielded only
mono- and bicyclic products. Therefore, the bulky Et substituent at C-10 is, to an extent,
better tolerated than that at C-15; the hydrophobic interaction between the alkyl group
and the binding site at C-10 would be more robust compared with that at C-15, leading
to the formation of the tricyclic 37. However, the C-10 recognition site is not large
enough to fully accommodate the bulkier Et group, thus mono- and bicyclic products
were generated at relatively higher yields than the tricyclic product. The cyclization
reactions specific to 19b, which were not found for 20a and 20b, are depicted in
Scheme 4D and 4E. Substrate 19b could be converted to the tetracyclic cation 8’
through a series of ring-forming reactions: 4’ 5’  6’ (like cation 6)7’ (like cation
7). The 1,2-hydride shift of H-17 to cation C-20 followed by the deprotonation of H-13
generated the tetracyclic product 31. The tetracyclic cation 8’ underwent a normal
polycyclization pathway, as shown in Scheme 1, yielding hopene homolog 33. It is
notable that the production of the tetracyclic dammarane scaffold 31 and the pentacyclic
hopene skeleton 33 were not observed in the cyclization reactions of 20a and 20b. The
placement of the bulkier Et group at C-19 of 1, to an extent, affected the polycyclization
result, leading to the generation of the tetracyclic 31 in addition to the pentacyclic
product 33, and had far-reaching effects on the reaction pathway and halted the
polycyclization reactions at the premature mono- and bicyclic stages (28─30 and 32). In
13
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
contrast, the substitution of the large group at C-10 and C-15 (20a and 20b) completely
terminated the polycyclization reaction at the abortive cyclization stages (mono- to
tricycle 34─42). The Et-substituent at C-19 perturbed the normal polycyclization
reactions and had a strong effect on the early reaction stage(s) of the polycyclization
reaction. The Me-29 binding site is not large enough to fully accommodate the Et-
substituent, possibly leading to the slightly inclined geometry of 19b around the D-ring
formation site. This altered geometry may have caused inappropriate positioning of the
substrate head of 19b (A- and A/B-ring formation sites), but the hydrophobic interaction
between the binding site and the Me-29 group are substantially robust, and thus, fully
cyclized product 33 could be generated to some extent. Production of 29 was also
limited to the reaction of 19b. As depicted in Scheme 4E, 19b was folded in the boat
structure 43 to fulfil the product stereochemistry of the A-ring. Proton attack on the
terminal double bond could yield the monocyclic cation 44 (path k). A 1,2-hydride shift
from H-5 to C-6 cation and Me-23 shift from C-4 to C-5 followed by the
deprotonation of H-3 (path l) gave rise to the (1,5R,6R)-trimethylcyclohexene ring 29
(IUPAC numbering). Given that 19b is folded in a chair conformation (path m), the
cyclization product would lead to the (1,5R,6S)-trimethylcyclohexene ring (IUPAC
numbering), the stereochemistry of which is different from product 29. The improper
arrangement of the substrate head of 19b in the reaction cavity would have a far-
reaching effect on the cyclization pathway and would lead to an abnormal boat
conformation during A-ring formation.
<Scheme 4>
Steric bulks of the Me-binding sites at C-6, C-10, C-15 and C-19
The yields obtained from the cyclization of 1, 19a, 19b, 20a and 20b by the native
AaSHC were 100%, 0%, 18%, 12% and 9.0% (Fig. 3A), respectively, and the yield
ratios were as follows: monocycle 7.2%, bicycle 7.5%, tetracycle 1.2% and pentacycle
2.4% (total yield 18%) for 19b; monocycle 4.8%, bicycle 4.8%, tricycle 2.6% (total
yield 12%) for 20a; and monocycle 7.1% and bicycle 1.9% (total yield 9.0%) for 20b;
the absence of a reaction for 19a indicated that the large Et-substituent at C-6
completely prevented the cyclization reaction. In contrast, a less-bulky hydrogen atom
at C-6 (11a) afforded the fully cyclized hopene skeleton 12 without any abortive
cyclization product (Scheme 2).[14] Thus, the recognition site at C-6 is noted to be
compact (Fig. 3B). Substitution of the Et-group at C-19 (19b) afforded a broad
spectrum of cyclization products, including the final product 33, indicating that
bulkiness at C-19 can be tolerated to a certain extent. However, the tolerance is low, as
14
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
evidenced by the markedly lower conversion (18%) than that of the genuine substrate 1
and by the high production yields of mono- and bicycles (see Fig. 3A). In contrast, 29-
norsqualene 11b underwent normal cyclization to yield hopane scaffolds 13 and 14
(Scheme 2A). [14] Therefore, the recognition site of position C-19 is somewhat loosely
packed but is not large enough to fully accommodate the Et group. The recognition sites
at positions C-10 and C-15 are more tightly packed than that at C-19, as no tetra- and/or
pentacyclic product was found, and instead, only mono-, bi- and tricyclic products were
found, which were produced at the early cyclization stage. Furthermore, the individual
yields of 20a and 20b were somewhat lower than the yield of 19b. However, the
replacement of the less bulky hydrogen atom at C-15 (12b) afforded the normal
cyclization hopene homolog 17 and the unusual folding to the boat conformation
afforded 18 (opposite stereochemistry at position C-21, see Scheme 2). Thus, the Me-
binding site at C-15 is not large enough to fully accommodate the Et group. The Me-
recognition site at C-10 is also tightly packed but is somewhat less compact than that at
C-15, as tricyclic 37 was generated as one of the enzymatic products. As described in
the Introduction section, 27-norsqualene 12a underwent abnormal conformational
folding, as shown in Scheme 2A, yielding unprecedented cyclization products 15 and
16.[15] Norsqualene 12a lacks the Me-27 group at C-10. The recognition site at C-10
strongly captures Me-27 in 1, but the Me-27 group is missing in norsqualene 12a. The
Me-27 recognition site incorrectly accommodates the Me-28 instead of the hydrogen
atom at C-10, leading to an abnormal folding conformation, as shown in Scheme 2A.
This strong hydrophobic interaction between the Me-27 binding site and the Me-27 of 1
led to the normal all-chair folding conformation shown in Scheme 1. However, the cleft
for the Me-27 binding (recognition) site is not large enough to completely accommodate
the Et group, and thus, the polycyclization of 20a is halted at the tricyclic 37 stage. The
cleft size of the binding site of Me-28 at C-15 is not larger than that of Me-27 of 1, and
the hydrophobic interaction is not robust; thus, the Et-substituent of 20b could not be
accommodated, and no tricyclic product was generated.
<Fig. 3>
Comparison of cyclization pathways between AaSHC, and lanosterol and -amyrin
synthases
We have reported the pathways of cyclization of Et-substituted substrate analogs by
hog-liver lanosterol synthase (HlLAS) [20-23] and Euphorbia tirucalli -amyrin synthase
(EtAS).[16, 17, 24, 25] Fig. S14 summarizes the cyclization reactions of ethyl-substituted
oxidosqualenes (OXSQs) by HlLAS. The compound 10-EthylOXSQ was converted to
15
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
four different products: methylenecyclohexane skeleton 43 (monocycle produced from a
chair structure), cyclohexanone scaffold 44 (monocycle from a boat form), a 6,6,5-fused
tricycle 45 and a 6,6,6,5-fused lanosterol homolog 46 in a ratio of ca. 1:1:1:1. In the
case of the15-ethylOXSQ substrate, a 6,6,5-fused tricycle 47 and a lanosterol homolog
48 were produced in a ratio of 28:72. The compound 19-EthylOXSQ was transformed
only to lanosterol homolog 49. Notably, monocyclic products were produced only from
10-ethylOXSQ and not from 15-ethylOXSQ. In contrast, monocyclic products were
generated in high yields from both 10-ethylsqualene (20a) and 15-ethylsqualene (20b)
by reaction with AaSHC, indicating that the Me-recognition sites at C-10 and C-15 of
AaSHC are more compact than those of HlLAS. Furthermore, in the case of AaSHC-
mediated reactions, tricycle 37 was produced from 10-ethylsqualene 20a, but little or no
tricycle was obtained from the reaction of 15-ethylsqualene 20b, indicating that the Me-
binding site at C-10 of AaSHC is somewhat less compact than that at C-15, unlike that
of HlLAS. This product pattern obtained using AaSHC is clearly distinct from that
obtained using HlLAS. Comparison of Fig. 3B with Fig. S14.2 provides information
about the difference in cleft sizes between AaSHC and HlLAS.
EtAS afforded no product from the substrates 6-, 10-, and 15-ethylOXSQs (Fig.
S15.1).[17] This finding strongly suggests that the cleft sizes at the methyl-binding sites
of these substrates are not large enough to accommodate the ethyl group, and thus, the
ethyl-substituted OXSQs cannot be folded into a chair/chair/chair conformation in the
reaction cavity. This finding indicates that the methyl-recognition sites at C-6, C-10 and
C-15 of EtAS are tightly packed, that is, the reaction cavity size responsible for the
early reaction stages (A/B/C-ring formation) is severely compact, and this tailored
reaction cavity firmly folds the substrate OXSQ into a chair/chair/chair
conformation.[17] In contrast, 19-ethylOXSQ afforded the -amyrin homolog 50
(pentacycle) and tetracycle 49 in a ratio of 1:3 (Fig. S15.1) with a conversion yield of
ca. 70% relative to that of the original substrate OXSQ,[24] demonstrating that the
reaction cavity (the methyl-binding site at C-19) is somewhat loosely packed (Fig.
S15.2), and thus, the larger ethyl group can be partially accommodated to generate the
tetra- and the fully cyclized (-amyrin) scaffolds in a ratio of 3:1.
Conclusions
In the present study conducted with AaSHC, we revealed that the reaction cavities
around positions C-6, C-10 and C-15 are relatively smaller than the reaction cavity at C-
19, and thus, the polycyclization reactions were terminated mainly at the mono- and
bicyclic stages. By comparing the polycyclization outcomes of ethyl-substituted
16
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
substrates using AaSHC, HlLAS or EtAS, we concluded that the promiscuity for the
steric bulk at C-6, C-10 and C-15 of squalene or OXSQ increases in the following
order: EtAS<AaSHC<HlLAS. Notably, the reaction cavities responsible for the early
reaction stages (A/B/C-ring formation) are significantly compact, irrespective of the
type of triterpene cyclase. This tightly packed reaction domain could impel each of
acyclic substrates into a required folding conformation: a chair/chair/chair conformation
by EtAS and AaSHC; and a chair/boat/chair conformation by HlLAS. Once these
required conformations are organized in the reaction cavities around the A/B/C-ring
formation site, the polyolefin cyclization occurs quickly and efficiently without the
accumulation of mono-, bi- and/or tricyclic ring systems. In contrast, the reaction
cavities for the later reaction stages (D/E-ring formation) are more loosely packed for
all three cyclases. It is possible that this large cavity for the later reaction stages allows
movement of the substrates within the reaction cavity, and thus, the large size of the
cavity is likely to be one of the factors responsible for the structural diversity of the
triterpene scaffolds generated; however, electronic properties, steric effects and/or
hydrophobicity of the active site residues of triterpene cyclases are also crucial factors
for determining the polycyclization outcomes.[2] An exceptionally large reaction domain
can lead to largely unrestricted motion of OXSQ and to perturbation of either the chair
or boat structure. Tetracyclic triterpene scaffolds such as lanosterol, cycloartenol and
dammarane would be generated by the exceptionally loosely packing around the E-ring
formation site. On the other hand, pentacyclic skeletons (hopene and plant triterpenes)
would be produced by the reaction cavity at the E-ring formation site, which is
relatively narrow compared to the cavity of the tetracycle-producing triterpene
synthases, thus allowing folding into either the chair or boat conformation around the E-
ring formation sites. We have reported the promiscuity of EtAS for steric sizes at C-23
of OXSQ. EtAS can tolerate the large ethyl group at both 23E-and 23Z-positions to
afford high yields (ca 40─50% of the original substrate) of only the final products (-
amyrin skeletons), indicating that the catalytic domain around the E-ring formation site
is somewhat loosely packed. [16, 25] However, AaSHC-mediated reactions of squalene
analogs bearing an ethyl group at C-2 (E- or Z-configured Et at initiation site) or C-23
(E- or Z-configured Et at terminal site) remain unpublished. These results will be
reported in due course to provide deeper insight into the effect of the bulkiness of the
squalene substrate on polycyclization by AaSHC. The X-ray crystallographic structures
of the two triterpene cyclases AaSHC[2a, 3] and human lanosterol synthase[26] have been
elucidated, and the cyclization mechanisms have been well-studied by mutation of the
active site residues; [2b, 2f, 2j] however, further substrate analog experiments are required
17
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
for better understanding the molecular recognition between the triterpene synthases and
the substrates.
Experimental Section
Analytical Methods
NMR spectra of the enzymatic products were recorded in CDCl3 or C6D6 on a Bruker
DMX 600 and DPX 400 spectrometers, the chemical shifts being given in ppm relative
to the CDCl3 solvent peak δH =7.26 and δC=77.0 ppm or to the C6D6 solvent peak δH
=7.28 and δC=128.0 ppm as the internal reference for 1H- and 13C NMR spectra. The
coupling constants J are given in Hz. GC analyses were done on a Shimadzu GC2014
chromatograph equipped with a flame ionization detector (a DB-1 capillary column,
30m x 0.32 mm x 0.25 m; J&W Scientific. Inc.). GC-MS spectra were on a JEOL SX
100 or a JEOL JMS-Q1000 GC K9 instrument equipped with a ZB-5ms capillary
column (30m x 0.25 mm x 0.25 m; Zebron) by using the EI mode operated at 70 eV.
High resolution-mass spectrometry (HR-MS) was performed on a JMS-T100GCV using
electron ionization (EI) mode. HPLC was carried out with Hitachi L-1700 (pump) and
L-7405 (UV detector), the HPLC peaks having been monitored at 210 or 214 nm.
Determination of the specific rotation values was conducted at 25C with a Horiba
SEPA-300 polarimeter (cuvette length: 10 cm) and the weights were measured with a
microbalance, and the samples were dissolved in 1.0 mL of CHCl3.
Syntheses of substrates 19 and 20
The experiments for the preparation of 19 are reported in the preceding paper.[17]
The detailed experiments for the chemical synthesis of 20 are described in the
Supporting Information, see: Fig. S1.
Incubation experiments with purified AaSHC and the cell-free extract
The Escherichia coli (DE3) encoding pET26b-SHC and pET3a-SHC was cultured
under the conditions reported in the previous papers.[18, 19] The detailed incubation
conditions are described in the Text (see Fig. 1).
Spectroscopic data of products 28─41.
EIMS and the 1H- and 13C-NMR spectra including 2D NMR and the assignments of
1H- and 13C-signals are described in the Supporting information Figs. S2─S13.
Products 28, 29, 30, 31, 32, 33, 37 and 41 were isolated in a pure state, thus the
complete assignments of 1H- and 13C-NMR signals of these products were determined
18
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
as follows.
Product 28. []D25 = +18 (c = 0.27, CHCl3). 1H NMR (400 MHz, CDCl3): =0.833 (s,
3H, 24-Me), 0.911 (s, 3H, 23-Me), 0.959 (t, 3H, J=7.6 Hz, Me-31), 1.22 (m, 1H, 3-H),
1.43 (m, 1H, 7-H ), 1.47 (m, 1H, 3-H), 1.53 (m, 2H, 2-H), 1.56 (m, 1H, 7-H), 1.591 (s,
3H, 26-Me), 1.606 (s, 6H, 27-Me and 30-Me), 1.683 (s, 3H, 29-Me), 1.69 (m, 1H, 5-H),
1.75 (m, 1H, 8-H), 1.94 (m, 1H, 8-H), 1.98 (m, 2H, 19-H), 2.00 (m, 3H, 1-H and 15-H),
2.02 (m, 6H, 11-H, 16-H and 20-H), 2.03 (m, 2H, 28-H), 2.05 (m, 1H, 1-H), 2.08 (m,
2H, 12-H), 4.54 (s, 1H, 25-H), 4.75 (s, 1H, 25-H), 5.15-5.06 (m, 4H, 10-H, 13-H, 17-H,
21-H) ppm. The assignment of 11-H and 12-H may be exchangeable. 13C NMR (100
MHz, CDCl3): =13.2 (q, C-31), 16.1 (q, 2C, C-26 and C-27), 17.7 (q, C-30), 23.2 (t, C-
28), 23.8 (t, C-2), 24.8 (t, C-7), 25.7 (q, C-29), 26.2 (q, C-24), 26.4 (t, C-12), 27.0 (t, C-
11), 28.3 (t, 2C, C-16 and C-20), 28.4 (q, C-23), 32.5 (t, C-1), 34.9 (s, C-4), 36.4 (t, C-
3), 36.5 (t, C-19), 38.2 (t, C-8), 40.1 (t, C-15), 53.6 (d, C-5), 108.8 (t, C-25), 123.8 (d,
C-17), 124.0 (d, C-10), 124.3 (d, C-21), 124.5 (d, C-13), 131.2 (s, C-22), 135.1 (s, C-
14), 135.7 (s, C-9), 140.8 (s, C-18), 149.4 (s, C-6) ppm. The assignment of C-11 and C-
12 and that of C-13 and C-17 may be exchangeable. MS (EI) see Fig. S2.1 in the
Supporting Information. HRMS (EI): m/z: calcd. 424.4069 for C31H52, found: 424.4073.
Product 29. []D25 = - 6 (c = 0.09, CHCl3). 1H NMR (600 MHz, C6D6): =0.979 (d, 3H,
J=6.8 Hz, Me-25), 0.979 (t, 3H, J=7.5 Hz), 1.01 (s, 3H, 24-Me), 1.53 (m, 2H, 1-H)
1.698 (s, 3H, 30-Me), 1.70 (m, 2H, 7-H), 1.747 (s, 3H, 27-Me), 1.766 (s, 3H, 26-Me),
1.791(s, 3H, 23-Me), 1.810 (s, 3H, 29-Me), 1.87 (m, 1H, 6-H), 1.93 (m, 8-H), 2.03 (m,
1H, 2-H), 2.10 (m, 1H, 2-H), 2.16 (mm 8-H), 2.20 (m, 2H, 28-H), 2.23 (m, 2H, 15-H),
2.25 (m, 2H, 19-H), 2.28 (m, 4H, 16-H and 20-H), 2.32 (m, 2H, 11-H), 2.35 (m, 2H, 12-
H), 5.38 (t, 2H, J=6.7 Hz, 17-H and 21-H), 5.47 (s, 2H, 10-H and 13-H), 5.59 (s, 1H, 3-
H) ppm. The assignments of 11-H and 12-H may be exchangeable. 13C NMR (150
MHz, C6D6): =13.5 (q, C-31), 16.0 (q, C-25), 16.2 (q, C-27), 16.4 (q, C-26), 17.7 (q,
C-30), 19.4 (q, C-23), 21.2 (q, C-24), 23.6 (t, C-28), 25.8 (q, C-29), 25.9 (t, C-2), 26.8
(t, C-12), 27.4 (t, C-11), 27.5 (t, C-1), 28.8 (t, 2C, C-16 and C-20), 33.5 (d, C-6), 34.7 (t,
C-8), 35.7 (t, C-7), 37.0 (t, C-19), 40.6 (t, C-15), 40.7 (s, C-5), 124.1 (d, C-10), 124.4
(d, C-17), 124.5 (t, C-3), 124.9 (d, C-13), 125.0 (d, C-21), 131.0 (s, C-22), 135.2 (s, C-
14), 136.2 (s, C-9), 139.7 (s, C-4), 140.9 (s, C-18) ppm. The assignments of C-1, C-11
and C-12 may be exchangeable. MS (EI): see Fig. S3.1 in the Supporting Information.
HRMS (EI): m/z: calcd. 424.4069 for C31H52, found: 424.4054.
Product 30. []D25 = +25 (c = 0.186, CHCl3). 1H NMR (400 MHz, CDCl3): =0.666 (s,
3H, 25-Me), 0.800 (s, 3H, 24-Me), 0.868 (s, 3H, 23-Me), 0.963 (t, 3H, J=7.6 Hz, 31-
Me), 1.00 (m, 1-H), 1.08 (dd, 1H, J=12.8, 2.8Hz, 5-H), 1.18 (ddd, 1H, J=13.0, 13.0, 4.1
19
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
Hz, H-3), 1.32 (m, 1H, 6-H), 1.37 (m, 2H, 2-H and11-H), 1.38 (m, 1H, 3-H), 1.48 (m,
1H, 11-H), 1.567 (s, 3H, 27-Me), 1.57 (m, 2-H), 1.604 (s, 3H, 30-Me), 1.61 (m, 1H, 9-
H), 1.64 (m, 1-H), 1.683 (s, 3H, 29-Me), 1.71 (m, 1H, 6-H), 1.82 (m, 1H, 12-H), 1.97
(m, 1H, 7-H), 1.99 (m, 4H, 15-H and 19-H), 2.03 (m, 2H, 28-H), 2.04 (m, 4H, 16-H and
20-H), 2.08 (m, 1H, 12-H), 2.38 (m, 1H, 7-H), 4.54 (s, 1H, 26-H), 4.82 (s, 1H, 26-H),
5.08 (m, 1H, 17-H), 5.12 (m, 1H, 21-H), 5.13 (m, 1H, 13-H) ppm. 13C NMR (100 MHz,
CDCl3): =13.3 (q, C-31), 14.5 (q, C-25), 16.0 (q, C-27), 17.7 (q, C-30), 19.4 (t, C-2),
21.7 (q, C-24), 23.2 (t, C-28), 23.8 (t, C-11), 24.5 (t, C-6), 25.7 (q, C-29), 26.4 (t, C-20)
26.9 (t, C-16), 27.0 (t, C-12), 33.6 (q, 2C, C-4 and C-23), 36.6 (t, C-19), 38.4 (t, C-7),
39.1 (t, C-1), 39.6 (s, C-10), 40.1 (t, C-15), 42.2 (t, C-3), 55.6 (d, C-5), 56.2 (d, C-9),
106.1 (t, C-26), 123.9 (d, C-17), 124.5 (d, C-21), 125.1 (d, C-13), 131.2 (s, C-22), 134.9
(s, C-14), 140.8 (s, C-18), 148.8 (s, C-8) ppm. The assignments of C-12, C-16 and C-
200 may be exchangeable. MS (EI): see Fig. S4.1 in the Supporting Information. HRMS
(EI): m/z: calcd. 424.4069 for C31H52, found: 424.4060.
Product 31. []D25 = - 41 (c = 0.11, CHCl3). 1H NMR (600 MHz, C6D6): =0.91 (m, 1H,
1-H), 0.940 (m, 1H, 5-H), 0.959 (t, 3H, J=7.1 Hz, Me-31), 0.972 (s, 3H, 29-Me), 0.997
(s, 3H, 19-Me), 1.038 (s, 3H, 28-Me), 1.122 (s, 3H, 18-Me), 1.29 (m, 1H, 3-H), 1.312
(s, 3H, 30-Me), 1.40 (m, 1H, 11-H), 1.42 (m, 1H, 6-H), 1.44 (m, 1H, 15-H), 1.46 (m,
1H, 21-H), 1.48 (m, 1H, 7-H), 1.50 (m, 1H, 3-H), 1.51 (m, 1H, 21-H), 1.52 (m, 1H, 6-
H), 1.58 (m, 2H, 22-H), 1.60 (m, 1H, 9-H), 1.66 (m, 1H, 2-H), 1.67 (m, 1H, 7-H), 1.68
(m, 1H, 11-H), 1.73 (m, 1H, 2-H), 1.741 (s, 3H, 27-Me), 1.78 (m, 1H, 1-H), 1.839 (s,
3H, 26-Me), 2.03 (m, 1H, 12-H), 2.08 (m, 1H, 23-H), 2.11 (m, 1H, 15-H), 2.27 (m, 1H,
16-H), 2.30 (m, 1H, 23-H), 2.35 (m, 1H, 16-H), 2.42 (m, 1H, 20-H), 2.59 (dd, 1H,
J=14.0, 3.4 Hz, 12-H), 5.44 (bs, 1H, 24-H) ppm. 13C NMR (150 MHz, C6D6): =12.6
(q, C-31),
16.6 (q, C-19), 17.8 (q, 2C, C-18 and C-27), 18.9 (t, C-6), 19.1 (t, C-2), 21.9 (q, C-29),
22.0 (t, C-11), 23.2 (q, C-30), 23.6 (t, C-12), 25.9 (q, C-26), 27.1 (t, C-23), 27.7 (t, C-
21), 29.1 (t, C-16), 30.9 (t, C-15), 33.5 (s, C-4), 34.5 (t, C-22), 35.8 (t, C-7), 38.0 (s, C-
10), 38.7 (q, C-28), 39.8 (d, C-20), 40.9 (t, C-1), 41.3 (s, C-8), 42.3 (t, C-3), 52.1 (d, C-
9), 57.1 (s, C-14), 57.3 (d, C-5), 125.6 (d, C-24), 130.8 (s, C-25), 132.7 (s, C-17), 142.0
(s, C-13) ppm. The assignments of C-2 and C-6 and those of C-18 and C-27 may be
exchangeable. MS (EI): see Fig. S5.1 in the Supporting Information. HRMS (EI): m/z:
calcd. 424.4069 for C31H52, found: 424.4083.
Product 32. []D25 = +12 (c = 0.11, CHCl3). 1H NMR (400 MHz, CDCl3): =0.741 (s,
3H, 25-Me), 0.852 (s, 3H, 23-Me), 0.874 (s, 3H, 24-Me), 0.96 (m, 1H, 1-H), 0.980 (t,
J=6.8 Hz, 3H, 31-Me), 1.16 (m, 1H, 3-H), 1.17 (dd, J=12.8, 4.8Hz, 1H, 5-H), 1.21 (m,
20
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
1H, 11-H), 1.29 (m, 1H, 3-H), 1.45 (m, 2H, 2-H and 11-H), 1.51 (m, 1H, 2-H), 1.604 (s,
3H, 30-Me), 1.614 (s, 3H, 27-Me), 1.63 (m, 1H, 9-H), 1.683 (s, 3H, 29-Me), 1.710 (bs,
3H, 26-Me), 1.84 (m, 1H, 1-H), 1.85 (m, 1H, 6-H), 1.94 (m, 1H, 6-H), 1.97 (m, 1H, 12-
H), 2.00 (m, 4H, 15-H and 19-H), 2.03 (m, 4H, 16-H and 28-H), 2.08 (m, 2H, 20-H)
2.17 (m, 1H, 12-H), 5.08 (m, 1H, 17-H), 5.13 (m, 2H, 13-H and 21-H), 5.38 (bs, 1H, 7-
H) ppm. The assignments of 16-H and 20-H may be exchangeable. 13C NMR (100
MHz, CDCl3): =13.3 (q, C-31), 13.6 (q, C-25), 16.2 (q, C-27), 17.7 (q, C-30), 18.8 (t,
C-2), 21.9 (q, C-24), 22.2 (q, C-26), 23.2 (t, C-28), 23.8 (t, C-6), 25.7 (q, C-29), 26.4 (t,
C-20), 27.3 (t, 2C, C-11 and C-16), 30.2 (t, C-12), 33.0 (s, C-4), 33.2 (q, C-23), 36.6 (t,
C-19), 36.7 (s, C-10), 39.2 (t, C-1), 40.1 (t, C-15), 42.4 (t, C-3), 50.2 (d, C-5), 54.2 (d,
C-9), 122.0 (d, C-7), 123.8 (d, C-17), 124.3 (d, C-13), 124.5 (d, C-21), 131.2 (s, C-22),
135.0 (s, C-14), 135.6 (s, C-8), 140.9 (s, C-18) ppm. The assignments of C-16 and C-20
may be exchangeable. MS (EI): see Fig. S6.1 in the Supporting Information. HRMS
(EI): m/z: calcd. 424.4069 for C31H52, found: 424.4039.
Product 33. []D25 = + 16 (c = 0.121, CHCl3). 1H NMR (600 MHz, C6D6): =0.87 (bd,
J=11.4 Hz, 1H, 5-H), 0.89 (m, 1H, 1-H), 0.93 (m, 1H, 19-H), 0.977 (s, 3H, 25-Me)
0.997 (s, 3H, 24-Me), 1.021 (t, J=7.4 Hz, 3H, 31-Me), 1.046 (s, 3H, 23-Me), 1.101 (s,
3H, 26-Me), 1.187 (s, 3H, 27-Me), 1.29 (m, 1H, 3-H), 1.32 (m, 1H, 7-H), 1.35 (m, 2H,
11-H and 15-H), 1.38 (m, 1H, 9-H), 1.45 (m, 1H, 17-H), 1.46 (m, 2H, 6-H and 15-H),
1.50 (m, 2H, 3-H and 13-H), 1.52 (m, 1H, 6-H), 1.57 (m, 1H, 28-H), 1.62 (m, 2H, 7-H
and 11-H), 1.64 (m, 2H, 2-H and 16-H), 1.71 (m, 2H, 12-H), 1.73 (m, 1H, 2-H), 1.76
(m, 1H, 16-H), 1.77 (m, 2H, 1-H and 19-H), 1.80 (m, 1H, 28-H), 1.87 (m, 1H, 20-H)
1.912 (s, 3H, 30-Me), 1.98 (m, 1H, 20-H), 2.72 (m, 1H, 21-H), 5.07 (s, 1H, 29-H), 5.11
(s, 1H, 29-H) ppm. 13C NMR (150 MHz, C6D6): =10.6 (q, C-31), 16.1 (q, C-25)
17.3 (q, C-26), 17.6 (q, C-27), 19.1 (t, 2C, C-2 and C-6), 20.0 (t, C-28), 21.5 (t, C-16),
21.8 (q, C-24 and t, C-11), 25.7 (q, C-30), 26.5 (t, C-12), 28.1 (t, C-20), 33.4 (s, C-4),
33.6 (t, C-7 and q, C-23), 34.3 (t, C-15), 36.9 (t, C-19), 37.6 (s, C-10), 40.6 (t, C-1),
42.4 (t, C-3), 42.5 (s, 2C, C-8 and C-14), 46.3 (d, C-21), 47.4 (s, C-18), 50.9 (d, C-9),
51.3 (d, C-13), 56.3 (d, C-17), 56.5 (d, C-5), 109.9 (t, C-29), 148.9 (s, C-22) ppm. MS
(EI): see Fig. S7.1 in the Supporting Information. HRMS (EI): m/z: calcd. 424.4069 for
C31H52, found: 424.4061.
Product 37. []D25 = + 49 (c = 0.06, CHCl3). 1H NMR (600 MHz, CDCl3): =0.734 (s,
3H, 25-Me), 0.772 (t, J=7.4Hz, 31-Me), 0.823 (d, J=6.7Hz, 27-Me), 0.869 (s, 3H, 23-
Me), 0.905 (s, 3H, 24-Me), 0.91 (m, 1H, 15-H), 1.05 (m, 1H, 1-H ), 1.18 (m, 1H, 3-H),
1.20 (m, 1H, 11-H), 1.22 (m, 1H, 5-H), 1.33 (m, 1H, 12-H), 1.35 (m, 1H, 26-H), 1.38
(m, 1H, 14-H), 1.40 (m, 1H, 2-H), 1.42 (m, 1H, 3-H), 1.48 (m, 1H, 11-H), 1.51 (m, 2H,
21
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
12-H and 15-H), 1.52 (m, 1H, 26-H), 1.55 (m, 1H, 2-H), 1.599 (s, 3H, 28-Me), 1.60 (m,
1H, 1-H ), 1.604 (s, 3H, 30-Me), 1.683 (s, 3H, 29-Me), 1.85 (m, 1H, 16-H), 1.88 (m,
1H, 6-H), 1.95 (m, 1H, 9-H), 1.97 (m, 2H, 19-H), 2.06 (m, 1H, 16-H), 2.07 (m, 2H, 20-
H), 2.08 (m, 1H, 6-H), 5.11 (m, 2H, 17-H and 21-H), 5.20 (bs, 1H, 7-H) ppm. 13C NMR
(150 MHz, CDCl3): =9.11 (q, C-31), 13.6 (q, C-25), 14.6 (q, C-27), 16.0 (q, C-28),
17.7 (q, C-30), 18.9 (t, C-2), 21.7 (q, C-24), 23.6 (t, C-6), 24.2 (t, C-11), 25.7 (q, C-29),
26.7 (t, C-16), 26.8 (t, C-20), 30.4 (t, C-26), 31.9 (t, C-12), 31.9 (t, C-15), 32.9 (s, C-4),
33.4 (q, C-23), 34.9 (s, C-10), 39.8 (t, C-19), 39.9 (d, C-14), 40.4 (t, C-1), 42.8 (t, C-3),
50.7 (d, C-5), 50.7 (s, C-13), 58.8 (d, C-9), 115.7 (d, C-7), 124.5 (d, C-21), 125.1 (d, C-
17), 131.2 (s, C-22), 134.6 (s, C-18), 147.1 (s, C-8) ppm. The assignments of C-12 and
C-15 and those of C-16 and C-20 may be exchangeable. MS (EI): see Fig. S10.1 in the
Supporting Information. HRMS (EI): m/z: calcd. 424.4069 for C31H52, found: 424.4059.
Product 41. []D25 = + 46 (c = 0.02, CHCl3). 1H NMR (600 MHz, C6D6): =0.872 (s,
3H, 24-Me), 1.056 (s, 3H, 23-Me), 1.122 (t, J=7.4 Hz, 3H, 31-Me), 1.202 (s, 3H, 25-
Me), 1.21 (m, 1H, 3-H), 1.22 (m, 1H, 5-H), 1.30 (ddd, J=12.2,12.2, 3.7 Hz, 1H, 1-H),
1.36 (m, 1H, 3-H), 1.42 (m, 1H, 2-H), 1.52 (m, 1H, 2-H), 1.56 (m, 1H, 7-H), 1.694 (s,
3H, 30-Me), 1.72 (m, 1H, 1-H), 1.745 (s, 3H, 28-Me), 1.808 (s, 3H, 29-Me), 1.848 (s,
3H, 26-Me), 1.85 (m, 1H, 7-H), 2.19 (q, J=7.4Hz, 2H, 27-H), 2.23 (m, 2H, 19-H), 2.26
(m, 2H, 15-H), 2.30 (m, 4H, 11-H and 12-H), 2.30 (m, 4H, 16-H and 20-H), 2.33 (m,
1H, 8-H), 2.46 (m, 1H, 8-H), 5.38 (t, J=6.9 Hz, 1H, 21-H), 5.43 (m, 1H, 17-H), 5.45 (m,
1H, 13-H), 5.54 (bs, 10-H) ppm. The assignments of 13-H and 17-H may be
exchangeable. 13C NMR (150 MHz, C6D6): =13.4 (q, C-31), 16.1 (q, C-28), 16.3 (q, C-
26), 17.7 (q, C-30), 20.8 (t, C-2), 21.6 (q, C-24), 23.6 (q, 2C, C-25 and C-27), 25.4 (t,
C-7), 25.8 (q, C-29), 27.2 (t, C-20), 27.4 (t, C-16), 28.4 (t, C-11), 29.0 (t, C-12), 32.9 (q,
C-23), 35.6 (s, C-4), 37.0 (t, C-15), 40.2 (t, C-19), 41.8 (t, C-3), 43.4 (t, C-8), 43.95 (t,
C-1), 56.9 (d, C-5), 73.5 (s, C-6), 124.5 (d, C-13), 124.6 (d, C-10), 124.9 (d, 2C, C-17
and C-21), 131.0 (s, C-22), 134.9 (s, C-18), 136.5 (s, C-9), 141.1 (s, C-14) ppm. The
assignments of C-11 and C-12 and those of C-16 and C-20 may be exchangeable. MS
(EI): see Fig. S12.1 in the Supporting Information; the molecular ion m/z 442 (M+) was
not observed, but m/z 424 (M+-H2O) was clearly observed. HRMS (EI): m/z: calcd.
424.4069 for C31H52, found: 424.4065.
Other products. As described in Text, the separation of a mixture of 34 and 35, that of
36 and 38, and that of 39 and 40 failed. Therefore, the NMR spectra were provided as a
mixture of the two products. The chemical shifts of the 1H- and 13C-signals, which were
differentiated between 34 and 35, between 36 and 38 and between 39 and 40, are
described in Fig. S8.9, Fig. S9.9 and Fig. S11.9, respectively.
22
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
Acknowledgements
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: Alkene; cyclization, enzyme catalysis; polycycle; terpenoids
23
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
References
[1] a) R. B. Woodward and K. Bloch, J. Am. Chem. Soc., 1953, 75, 2023-2024. b) A.
Eschenmoser, L. Ruzicka, O. Jeger and D. Arigoni, Helv. Chim. Acta 1955, 38,
1890–1904. c) J. W. Cornforth, R. H. Cornforth and C. Donniger, G. Popjak, Y.
Shimizu, S. Ichii, E. Forchielli and E. Caspi, J. Am. Chem. Soc., 1965, 87, 3224-
3228.
[2] For the reviews of squalene and oxidosqualene cyclases, see: a) K. U. Wendt, G. E.
Schulz, E. J. Corey and D. R. Liu, Angew. Chem., Int. Ed. 2000, 39, 2812–2833. b)
T. Hoshino and T. Sato, Chem. Commun. 2002, 291–301. c) R. A. Yoder and J. N.
Johnston, Chem. Rev. 2005, 105, 4730–4756. d) M. J. R. Segura, B. E. Jackson and
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 and T. T. Wang, Chem.
Rec. 2008, 8, 302–325. g) W. 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.
[3] K. U. Wendt, K. Poralla and G. E. Schulz, Science, 1997, 277, 1811-1815.
[4] (a) T. Sato and T. Hoshino, Biosci., Biotechnol., Biochem., 1999, 63, 2189–2198; (b)
T. Dang and G. D. Prestwich, Chem. Biol., 2000, 7, 643–649.
[5] T. Hoshino and T. Sato, Chem. Commun., 1999, 2005–2006.
[6] C. Pale-Grosdemange, T. Merkofer, M. Rohmer and K. Poralla, Tetrahedron Lett.,
1999, 40, 6009–6012.
[7] T. Hoshino, M. Kouda, T. Abe and S. Ohashi, Biosci. Biotechnol. Biochem., 1999,
63, 2038–2041.
[8] C. Pale-Grosdemange, C. Feil, M. Rohmer and K. Poralla, Angew. Chem., Int. Ed.,
1998, 37, 2237–2240.
[9] T. Sato, T. Abe and T. Hoshino, Chem. Commun., 1998, 2617–2618.
[10] T. Merkofer, C. Pale-Grosdemange, K. U. Wendt, M. Rohmer and K. Poralla,
Tetrahedron Lett., 1999, 40, 2121–2124.
[11] T. Hoshino, T. Abe and M. Kouda, Chem. Commun., 2000, 441–442.
[12] (a) T. Hoshino, M. Kouda, T. Abe and T. Sato, Chem. Commun., 2000, 1485–1486;
(b) Related paper: S. Schmitz, C. Füll, T. Glaser, K. Albert and K. Poralla,
Tetrahedron Lett., 2001, 42, 883–885.
[13] T. Hoshino, S. Nakano, T. Kondo, T. Sato and A. Miyoshi, Org. Biomol. Chem.,
2004, 2, 1456-1470.
24
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
[14] S. Nakano, S. Ohashi and T. Hoshino, Org. Biomol. Chem., 2004, 2, 2012-2022.
[15] T. Hoshino and S. Ohashi , Org. Lett., 2002, 4, 2553-2556.
[16] T. Hoshino, Y. Miyahara, M. Hanaoka, K. Takahashi and I. Kaneko, Chem. Eur. J.,
2015, 21, 15769 – 15784.
[17] Y. Terasawa, Y. Sasaki, Y. Yamaguchi, K. Takahashi and T. Hoshino, Eur. J. Org.
Chem., 2017, 287-295.
[18] M. Morikubo, Y. Fukuda, K. Ohtake, N. Shinya, D. Kiga, K. Sakamoto, M.
Asanuma, H. Hirota, S. Yokoyama and T. Hoshino, J. Am. Chem. Soc., 2006, 128,
13148-13194.
[19] a) T. Sato, Y. Kanai and T. Hoshino, Biosci. Biotechnol. Biochem., 1998, 62, 407-
411. b) T. Sato and T. Hoshino, Biosci. Biotechnol. Biochem., 1999, 63, 2189-2198.
c) T. Sato, S. Sasahara, T. Yamakami and T. Hoshino, Biosci. Biotechnol. Biochem.,
2002, 66, 1660-1670.
[20] T. Hoshino and Y. Sakai, Tetrahedron Lett., 2001, 42, 7319-7323.
[21] T. Hoshino, E. Ishibashi, J. Chem. Soc., Chem. Commun., 1995, 2401-2402.
[22] T. Hoshino and Y. Sakai, Chem. Commun., 1998, 1591-1592.
[23] T. Hoshino, A. Chiba and N. Abe, Chem. Eur. J., 2012, 18, 13108-13116.
[24] T. Hoshino, Y. Yamaguchi, K. Takahashi and R. Ito, Org. Lett., 2014, 16, 3548-
3551.
[25] I. Kaneko and T. Hoshino, J. Org. Chem., 2016, 81, 6657-6671.
[26] R. Tohma, T. Schulz-Gasch, B. D’Archy, J. Benz, J. Arbi, H. Dehmlow, M.
Henning, M. Stihle and A. Ruf, Nature, 2004, 432,118-122.
25
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
Figures and Schemes
Scheme 1. Polycyclization pathway of squalene 1 to pentacyclic-hopene 2 and hopanol
3 proposed based on the abortive cyclization products afforded by the site-directed
variants.
26
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
Scheme 2. (A) Cyclization pathway of norsqualene analogs 11 and 12. Two pathways a
and b are possible, as the isopropylidene moieties, which are necessary for initiation of
the cyclization reaction,[13} are present at the both termini. (B) Structures of Et-
substituted 19 and 20. Two names are proposed for both 19 and 20 based on the two
different directions for initiating the polycyclization reactions: 6-ethylsqualene 19a, 19-
ethylsqualene 19b, 10-ethylsqualene 20a and 15-ethylsqualene 20b.
27
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
Scheme 3. Schemes for the syntheses of ethylated analogs 19 and 20.
28
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800010
European Journal of Organic Chemistry
2
(A): Squalene (1)
1
3
30
19
32
33
× 8
(B): 6(19)-Et analog (19)
28
31
29
*
(C): 10(15)-Et analog (20)
20
37
38
39, 40
36
41 42
35
34
× 4
26
30
34
38
42
46
50
54
58
Retention time (min)
Fig. 1. Gas chromatograms of the reaction mixture obtained by separately incubating 1
(A), 19 (B) and 20 (C) with wild-type AaSHC. Triton X-100 was removed with a short
SiO2 column by eluting with a mixture of hexane and EtOAc (100:10). Identical
incubation conditions were used to compare each of the product distribution patterns:
substrate 200 g, pET26b-SHC (5 g, purified with a Ni+-NTA column), Triton X-100
(0.2%) in sodium citrate buffer (pH 6.0, 60 mM), total volume 2.0 mL, incubated at
55C for 6 h. GC conditions: J & D, DB-1, capillary column (length 30 m, I.D. 0.32
mm, film thickness 0.25 m); injection temperature 300C; elevated temperature
210─270C (rate 1C/min). The 1H-NMR spectra of the peaks (marked with the symbol
*) indicated a mixture of materials, and thus, the isolation of some products was
unsuccessful due to the marginal production of each product.
29
This article is protected by copyright. All rights reserved.