Chemo-enzymatic syntheses of drimane-type sesquiterpenes and the fundamental core of hongoquercin meroterpenoid by recombinant squalene-hopene cyclase

Article abstract of DOI:10.1039/c1ob06419c

Squalene-hopene cyclase (SHC) converts squalene (C₃₀) into pentacyclic triterpenes of hopene and hopanol. A linear sesquiterpene, (6E,10E)-2,6,10-trimethyldodeca-2,6,10-triene, underwent cyclization catalyzed by SHC, affording the following six

Full text of DOI:10.1039/c1ob06419c

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2012, 10, 440
www.rsc.org/obc
PAPER
Chemo-enzymatic syntheses of drimane-type sesquiterpenes and the
fundamental core of hongoquercin meroterpenoid by recombinant
squalene–hopene cyclase†
Yukie Yonemura, Takuro Ohyama and Tsutomu Hoshino*
Received 19th August 2011, Accepted 19th September 2011
DOI: 10.1039/c1ob06419c
Squalene–hopene cyclase (SHC) converts squalene (C30) into pentacyclic triterpenes of hopene and
hopanol. A linear sesquiterpene, (6E,10E)-2,6,10-trimethyldodeca-2,6,10-triene, underwent cyclization
catalyzed by SHC, affording the following six bicyclic sesquiterpenes (drimane skeleton) in relatively high
yield (68%): drim-7(8)-ene, drim-8(12)-ene, drim-8(9)-ene, driman-8a-ol, driman-8b-ol, and the novel
sesquiterpene, named quasiclerodane, the skeleton of which is analogous to that of clerodane diterpene.
To extend the scope of the enzymatic syntheses, acyclic sesquiterpenes to which a phenol moiety
was appended were subjected to the enzymatic reaction catalyzed by SHC. The cyclic meroterpene
core present in hongoquercins A and B was successfully prepared. The formation mechanisms of
drimane-type sesquiterpenes and the cyclic meroterpene core of hongoquercins A and B are discussed.
was found in the volatile extracts of Greek tobacco.12 Compound
8 has not yet been found in nature. Therefore, SHC holds great
potential for the syntheses of drimane-type sesquiterpenes from
the readily available farnesol 4.2
Introduction
Enzymatic cyclization reactions of polyolefin substrates have
fascinated chemists for over half a century. The polycyclization
reactions of squalene 1 and (3S)-2,3-oxidosqualene, each having
C30, proceed with complete regio- and stereochemical specificity to
furnish the remarkable structural diversity of triterpene scaffolds.1
We have reported the enzymatic characterization of squalene–
hopene cyclase (SHC) from Alicyclobacillus acidocaldarius, which
yields pentacyclic hopene 2 and hopanol 3 (diplopterol) via a
hopanyl cation (Scheme 1A), through mutagenesis and substrate
analog experiments.1a Remarkably, this enzyme is highly tolerant
to truncated and elongated substrates with carbon numbers of C15–
C35.2–4 Farnesol (C15, 4), for example, proved to be an excellent
substrate to afford cyclic sesquiterpenes in high yield (64%).2
We succeeded in synthesizing drimane-type sequiterpenes 5–7 by
incubating farnesol 4 with SHC (Scheme 1B). Drimenol 55 was
isolated from Bazzania trilobata5 and Diplophylum serrulatum,6
and used as the starting material for the syntheses of cytotoxic
8-epipuupehedione7 and g-polypodatetraene.8 Albicanol 6 was
isolated from the liverworts of Diplophyllum albicans,9 Diplophy-
lum serrulatum6 and Bazzania japonica,10 and employed for the
syntheses of a-polypodatetraene8 and cryptoporic acid A methyl
ester11 with anti-tumor promotion activity. Driman-8,11-diol 7
To extend the scope of enzymatic syntheses of sesquiterpenes by
SHC, substrate 9 (6E,10E)-2,6,10-trimethyldodeca-2,6,10-triene),
lacking the hydroxyl group of 4, was prepared and incubated
with SHC in the effort to synthesize 15–17, which are analogous
to 5–7, but lack a hydroxyl group at the 11-position. Interest-
ingly, in addition to 15–17, novel sesquiterpenes 18 and 20 and
naturally occurring 19 were prepared. In addition, other acyclic
sesquiterpene analogs 13 and 14 containing phenol moiety were
also synthesized and subjected to the enzymatic reaction in order
to synthesize a cyclic meroterpene core, such as hongoquercin
A 10 and hongoquercin B 11, and BE-40644 12 (Scheme 1C).
The C(8)-stereochemistry (S) of 12 is opposite to those of 10
and 11 (R), but their basic cores are the same. Meroterpenoids
are hybrid natural products composed of both terpenoid and
polyketide-derived structures.13 Hongoquercins A and B are
bioactive compounds with antibacterial activity,14 and BE-40664
has a human thioredoxin system inhibitory activity.15 Herein,
we report the first chemo-enzymatic syntheses of drimane-type
terpenes 15–20 from 9 and the cyclic meroterpene cores 21–25
from 13 and 14, in which a phenol moiety is appended to 9.
Department of Applied Biological Chemistry, Faculty of Agriculture, and
Graduate School of Science and Technology, Niigata University, Nishi-ku,
Ikarashi 2-8050, Niigata, 950-2181, Japan. E-mail: hoshitsu@agr.niigata-
u.ac.jp
Results and discussion
Enzymatic reaction of 9 with recombinant SHC
† Electronic supplementary information (ESI) available: synthetic methods
of compounds 9, 13, and 14; GC trace of reaction mixture of compound
14; and spectroscopic data (EIMS and NMR) of the enzymatic products
15–25. See DOI: 10.1039/c1ob06419c
Incubation of 9, prepared from farnesol 4 (ESI, page S3†), with
SHC was conducted as follows. Compound 9 (1 mg), emulsified
with Triton X-100 (20 mg), was incubated with the cell free extract
440 | Org. Biomol. Chem., 2012, 10, 440–446
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
Scheme 1 (A) Cyclization cascade of squalene 1 into hopene 2 and hopanol 3 mediated by squalene–hopene cyclase (SHC). (B) Enzymatic products2
of farnesol 4 and the substrate structure 9 to be tested. (C) Structures of hongoquercins A and B and BE-40644 (cyclic meroterpenes), and those of
substrates 13 and 14 to be tested.
(2 ml) as the enzyme source at pH 6.0 and 60 ◦C for 20 h with
a total reaction volume of 5 ml. Then, 15% KOH/MeOH was
added to the reaction mixture, and the products were extracted
with hexane. Triton X-100 included in the extracted solution was
removed with a short SiO2 column chromatography eluting with
a mixture of hexane and EtOAc (100 : 20). Fig. 1A shows the GC
profile of the product distribution pattern. A full conversion was
not attained – 32% of 9 remained unreacted – but 68% of 9 was
successfully converted into the enzymatic products. A large-scale
incubation of 9 (100 mg) was conducted in order to isolate the
enzymatic products. Non-polar and polar factions were crudely
separated by SiO2 column chromatography; the non-polar fraction
This journal is
The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 440–446 | 441
©

View Article Online
Structures of all the enzymatic products were determined by
1
1H- and 13C NMR spectral analyses, including DEPTs, H–1H
COSY, NOESY, HMQC and HMBC spectra. The molecular
formula of 15, 16, 19 and 20 was determined to be C15H26 by
HREIMS, indicating the unsaturation number of 3. Each of these
products had one double bond, which was revealed by observing
the following chemical shifts in C6D6: dC 121.9 (d) and 135.2 (s)
for 15; 106.3 (t) and 151.5 (s) for 16; 124.3 (s) and 136.2 (s) for
19; and 121.0 (d) and 143.8 (s) for 20, indicating that bicyclic
1
skeletons should be assigned to these products. In the H NMR
spectrum of 15, one doublet Me (3H, dH 0.958, J = 7.2 Hz, Me-
11) and one allylic Me (3H, dH 1.75, br s, Me-12) were found.
Me-11 had HMBC correlations with C-9 (dC 49.03, d), C-10 (dC
36.02, s) and C-8 (dC 135.3, s). Me-12 had HMBC cross peaks
with C-9, C-8 and C-7 (121.9, d); the detailed HMBC analyses
are shown in ESI (page S6).† The a-oriented H-9 was confirmed
from clear NOEs of H-9/Hax-5/Hax-1/Hax-3. These data allowed
us to propose the structure of drim-7(8)-ene for 15 (Scheme 2).
For 16, the presence of one doublet Me (3H, dH 1.04, J = 6.8 Hz,
Me-11) and one vinylidene residue (1H, dH 4.98, brs and 1H, dH
4.79, brs, CH2-12) was confirmed. The clear HMBC correlations
of Me-11/C-9, Me-11/C-10 and Me-11/C-8, as well as those of
H-12/C-9 and H-12/C-7, established the structure of 16 to be
drim-8(12)-ene as depicted in Scheme 2. H-9 was determined to
be a-oriented by the unambiguous NOE between H-5 and H-9.
Fig. 1 (A) GC trace of the reaction mixture obtained by incubating 9 with
the native SHC. The asterisk mark s◦hows the impurity. GC conditions:
column temp., 80–250 ◦C; rate, 5.0 C min-1, injection temp., 290 ◦C;
carrier gas (N2), 0.75 kg cm-2.(B) GC-chromatogram of the non-polar
fraction, which was separated by SiO2 column chromatography. GC
conditions: column temp., 140 ◦C; injection temp., 290 ◦C; carrier gas
(N2), 0.2 kg cm-2.
1
In the H NMR spectrum of 19, two allylic Me groups (Me-11
and Me-12) were found at dH 1.67 (6H, brs). No doublet Me
was detected. The clear HMBC cross peaks of Me-15 (dH 1.12,
3H, s)/C-9 (dC 136.2, s), Me-11/C-10 (dC 38.50, s), Me-12/C-
7 (dC 34.05, t) and Me-12/C-8 ((dC 124.3, s) demonstrated that
the double bond is located at C-8 and C-9. A strong NOE was
observed between Me-14 (dH 0.985, 3H, s) and Me-15. Thus, the
structure of 19 was determined as shown in Scheme 2. Products
17 and 18 are both polar compounds; alcoholic carbons were
found at C-8 (dC 72.25 (s) for 17 and dC 71.84 (s) for 18). For 17,
a strong NOE between Me-15 (dH 0.800, 3H, s) and Me-12 (dH
1.13, 3H, s) was found, but no corresponding NOE was found
for 18, indicating that the C8-stereochemistry is opposite between
was collected by eluting with hexane, and polar fraction was eluted
with a mixture of hexane and EtOAc (100 : 10). Four enzymatic
products (15, 16, 19, and 20) were included in the non-polar
fraction (Fig. 1B). These compounds were successfully separated
and purified using a 5% AgNO3-impregnated SiO2 column eluting
with hexane containing a trace of EtOAc. Two products (17 and
18) included in the polar fraction were isolated by a SiO2 column
(hexane : EtOAc = 100 : 5). Fig. 1A and 1B revealed the following
product distribution: 10% for 15, 4% for 16, 16% for 17, 14% for
18, 4% for 19 and 20% for 20.
Scheme 2 Enzymatic products 15–20 prepared by incubating 9 with the native SHC. The deprotonation reactions from intermediate 26 gave 15, 16 and
19. A water molecule attack to 26 afforded 17 and 18. 1,2-Shifts of hydrides and methyl groups in antiparallel fashion according to the curved arrows
shown in 26 generated novel sesquterpene 20, named quasiclerodane.
442 | Org. Biomol. Chem., 2012, 10, 440–446
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
17 and 18. The structures of 17 and 18 were determined to be
driman-8a-ol and driman-8b-ol, respectively (Scheme 2). For 20,
one allylic Me (Me-13, dH 1.72, 3H, d, J = 1.2 Hz) was found that
had HMBC correlations with C-3 (dC 121.0, d), C-4 (dC 143.9,
s) and C-5 (dC 38.48, s). One doublet Me (Me-12, dH 0.963, 3H,
d, J = 6.8 Hz) had a cross peak with C-9 (dC 36.55, s), which
was further correlated with Me-11 (dH 1.01, 3H, s) and Me-15 (dH
0.809, 3H, s) in the HMBC spectrum. Further detailed NOE and
HMBC analyses (ESI, pages S14, S18, S20 and S21†) revealed an
unnatural novel bicyclic skeleton (SciFinder), as shown in Scheme
2. This sesquiterpene skeleton is analogous to that of clerodane
diterpene. We propose the name quasiclerodane for 20. It should
be noted that novel product 20 was generated in a highest yield
(20%) among all of the products 15–20.
occurred. The detailed analyses of the HMBC data (ESI, page
22†) led to the proposal of the 6/6/6/6-fused tetracyclic structure
of 21. The molecular formula of products 22 and 23 was C21H30O
with an unsaturation number of 7. In addition to a benzene ring
(unsaturation number of 4), one double bond was involved in 22
and 23: dC 122.4 (d) and 135.8 (s) for 22; dC 107.5 (t) and 149.0
(s) for 23, suggesting that 22 and 23 are both tricyclic compounds
containing a benzene ring. In the HMBC spectrum of 22, Me-20
(dH 1.04, 3H, s) and allylic Me-21 (dH 1.77, 3H, s) had a cross peak
with C-9 (dC 54.51, d), and H-5 (dH 1.41, m) and H-9 (dH 2.56,
m) showed a strong NOE between them. The allylic Me-21 had
definitive HMBC cross peaks with C-7 (dC 122.4, d) and C-8 (dC
135.8, s), demonstrating that the double bond is located at C-7 and
C-8. Thus, the whole structure of 22 was determined as shown in
Scheme 3. For 23, protons of a vinylidene moiety were found at dH
4.82 (1H, s) and dH 4.71 (1H, s), which were correlated with C-21
(dC 107.5, t) in the HMQC spectrum. The methylene protons at
21-position and Me-20 (dH 0.820, 3H, s) had definitive HMBC
correlations with C-9 (dC 56.08, d). Detailed analyses of the
HMBC and NOESY spectra allowed us to propose the structure of
23, as shown in Scheme 2. SHC accepts 2,3-oxidosqualene as well
as squalene,16 and so we examined whether 14 also is acceptable
as the substrate. GC analysis of the reaction mixture showed that
only one product was produced (14:24 = ca. 1 : 1, see ESI, page
S5†). Purification by a SiO2 column (hexane : EtOAc = 100 : 05)
and subsequent HPLC (hexane : 2-PrOH = 100 : 0.05) afforded
pure product 24, which was then acetylated with Ac2O/py and
subjected to NMR analyses. An ether bridge was found between
C-8 (dC 76.43, s) and C-13 (dC 154, s). The assignment of C-
8 was confirmed by the HMBC correlation between Me-21 and
C-8. H-3 showed the following splitting pattern (1H, dH 4.70,
dd, J = 12.0, 4.8 Hz), indicating axial-orientation of H-3 and
equatorial-arrangement of OH. An acetyl group (dC 169.8, s;
dH 1.87, 3H, s) was determined to be at C-3 (dC 80.06, s) by
confirming the HMBC correlation of H-3 with the acetyl carbonyl
carbon (dC 169.8, s). No allylic methyl group was found in the 1H
NMR spectrum, indicating that a complete cyclization reaction
had occurred as shown in the structure of 24 in Scheme 3. The
b-oriented hydroxyl group at C-3 of 24 indicates that the (10S)-
10,11-epoxy derivative 14 was stereoselectively converted, but that
the10R-isomer underwent no reaction. Products 21 and 25 (acetate
of 24) are fundamental cores of (+)-hongoquercin A and (+)-
hongoquercin B, respectively. Thus, we succeeded in the syntheses
of these cyclic meroterpene cores by using recombinant SHC.
All the enzymatic products from 9 are listed in Scheme 2.
Drim-8(12)-ene 16 was isolated from Hedychium acuminatum,17
Cistus L.18 Eupatorium odoratum,19 rhizomes of Hedychium spica-
tum,20 Silphium L.21 and so forth. Driman-8a-ol 17 was found
in Greek tobacco and plant culture of Nicotiana silvestris,22,23
and from Nicotiana tabacum.24 Drim-8(9)-ene 19 is a natural
product isolated from Colorado shale oil.25 Product 20 is a novel
compound. Neither drim-7(8)-ene 15 nor driman-8b-ol 18 have
been isolated as natural products (SciFinder). Substrate 9 could
undergo cyclization reactions to give an intermediary bicyclic C8-
cation 26. The subsequent deprotonation reactions of H-7, H-12
and H-9 could afford 15, 16 and 19, respectively. The re-face attack
of a water molecule on the C8-cation gave 17, while the si-face
attack yielded 18. A series of 1,2-shifts of hydrides and methyl
groups in an antiparallel fashion could generate novel compound
Enzymatic reactions of 13 and 14 with recombinant SHC
Substrates 13 and 14, in which an o-hydroxylphenyl ring is
appended to a farnesyl moiety, were prepared (ESI, page S3–S5†)
in order to construct the fundamental skeletons (21 and 24) of
hongoquercins A and B. The incubation conditions of 13 and 14
were the same as that of 9. As shown in Fig. 2, GC analysis of the
incubation mixture of 13 with SHC afforded three product peaks
(with conversion yield of 10%). The yields of products 21, 22 and
23 were 6%, 2.5% and 1.5%, respectively.
Fig. 2 GC trace of the enzymatic reaction obtained by incubating
substrate 13 with SHC. Triton X-100 was removed by a short SiO2 column.
GC conditions: column temp., 190 ◦C; injection temp., 280 ◦C; carrier gas
(N2), 1.0 kg cm-2.
Separation by a SiO2 column (hexane : EtOAc = 100 : 0.05)
afforded two fractions, one containing 21 and the other containing
a mixture of 22 and 23. Products 22 and 23 were separated by a 5%
AgNO3-SiO2 column (hexane : EtOAc = 100 : 8–100 : 20). Products
21–23 were finally purified via normal phase HPLC. The structure
of product 21 was confirmed by NMR analyses, including 1D
and 2D NMR spectral data. In addition to a phenolic carbon (dC
154, s), an alcoholic carbon (dC 76.77, s, C-8) appeared, indicating
that an ether linkage was present between C-13 and C-8 of 21.
The position of the alcoholic carbon (C-8) was confirmed by the
HMBC cross peak between Me-21 (dH 1.22, 3H, s) and C-8. The
allylic methyl protons, which are present in substrate 13, were
not found, indicating that the complete polycyclization reaction
This journal is
The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 440–446 | 443
©

View Article Online
Scheme 3 Enzymatic products 21–25 obtained by the incubations of 13 and 14 with SHC. The tricyclic intermediate 27 could be produced. The
deprotonation reactions from H-7 and H-21 gave 22 and 23, respectively. A nucleophilic attack of phenolic OH to C-8 cation afforded 21 that is a
fundamental skeleton of hongoquercin A and B.
20: H-9 → C-8; Me-11 → C-9; H-5 → C-10; Me-14 → C-5 and the
final elimination reaction of H-3. The skeletons of products 18–
20 were not found in the products obtained by the cyclization of
farnesol 4. This finding suggests that the OH group of 4 interacts
with certain amino acid(s) via hydrogen bonding inside the SHC
cavity, but 9 cannot form the hydrogen bonding. Therefore, the
location of 9 inside the cavity may be different from that of 4,
possibly contributing to the production of 18–20.
have afforded the higher conversion. The epoxide ring of (10S)-14,
folded in a chair conformation,26 could have accessed the acidic
DXDD motif, but that of the (10R)-14, folded in a boat form,26
could not gain proximity to the DXDD motif, resulting in no
reaction. Previously, we reported that a squalene analog with C20
28 was cyclized in 33% yield to give complete cyclization products
29 and 30 (Scheme 4).2 The structure of 28 (C20) has one additional
isoprene unit than is found in 9 (C15). The conversion yield of 9
(68%) was approximately two times that (33%) of 28, indicating
that C15 was more efficiently cyclized than C20, but the reason for
this is not clear at present. Tanaka et al. reported the enzymatic
reaction of squalene analogs (C20) containing an indole ring 3127
or a pyrrole ring 32.28 They succeeded in the syntheses of indole-
or pyrrole-containing cyclic polyprenoid compounds by SHC, but
in very small yields (7.5% for indole-containing polyprenoid;27
and 1.1% for pyrrole-containing polyprenoid28). The enzymatic
products (36–39) of substrates 31 and 32, prepared by SHC,27,28
are described in the ESI (page S30).† The cyclization yield of 28 is
significantly higher than those of 31 and 32. As was discussed in
the comparison of 9 with 13, the low enzymatic yields of substrates
31 and 32 can be explained by 1) limited access of the terminal
SHC could convert substrate 13 to the bicyclic C-8 cation 27.
Phenolic OH group could attack to this cation as a nucleophile,
successfully affording the fundamental core 21 of hongoquercin
A (6/6/6/6-fused tetracycle, where ring C is a tetrahydropyran)
in 6% yield. The proton eliminations of H-7 and Me-21 from
27 gave tricyclic products 22 (2.5%) and 23 (1.5%), respectively.
The total conversion yield was significantly smaller (10%) than
that of phenol-lacking substrate 9 (68%). Presumably, the bulky
phenol moiety may have hindered access to the DXDD motif 1b
that enables initiation of the polycyclization reaction by proton
attack on the terminal double bond and further prevented the
access of the phenol ring to the C-8 cation of 27, and thus, the
cyclization reaction terminated at the bicyclic stage to yield 22 and
23. The conversion ratio of epoxide 14 was significantly higher than
that of 13. Furthermore, the S-form of 14 was exclusively cyclized
to give 24 in a high yield (almost quantitative yield, see the GC
trace shown in ESI, page S5†), but the alternative R-stereoisomer
underwent no reaction (no detectable amount of product with a-
oriented OH). The acetate of 24, i.e. 25, forms the fundamental
framework of hongoquercin B. At present, it is not clear why the
conversion yield of 14 was higher than that of 13, and the reason
for the enantioselective preference is not yet understood. The more
facile access to the DXDD sequence of (10S)-14, compared to that
of 13, may have caused the higher yield of (10S)-14. In other words,
a more favorable orientation of (10S)-14 in the catalytic cavity may
Scheme 4 Structures of polyprenoids 28, 31 and 32 with acyclic C20
isoprenoid chain length. Indole and pyrrole rings are substituted in the case
of 31 and 32, respectively. 28 undergoes a complete cyclization reaction to
give fully cyclized products 29 and 30,2 but the enzymatic reactions of 31
and 32 give incomplete cyclization products (see ESI, page S30†).27,28
444 | Org. Biomol. Chem., 2012, 10, 440–446
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
double bond of 31 and 32 to the DXDD motif, and 2) improper
orientation of 31 and 32 in the reaction cavity caused by the bulky
size of the indole and pyrrole rings; these impediments halted the
polycyclization reactions at the partially cyclized stage (ESI, page
S30†). Thus, the fully cyclized petromindole skeleton 40,29 in which
the indole ring is fused to the 6/6/6/6-fused carbotetracycle (ESI,
page S30†), was not produced.27
In conclusion, we succeeded in the chemo-enzymatic syntheses
of unnatural novel sesquiterpenes 15, 18 and 20 in relatively high
yields, and also of the basic cores of hongoquercins A and B (21
and 25, respectively) by recombinant SHC. As demonstrated here,
the enzymatic reactions of 9, 13 and 14 proceeded also with regio-
and stereochemical specificity as well as that of squalene molecule.
SHC has a great plasticity to tolerate a variety of squalene analogs.
Native or mutated SHC is a promising tool for obtaining further
novel cyclic terpenoid derivatives by chemo-enzymatic syntheses.
sesquiterpene products were evaporated at room temperature
without handling them at reduced pressure.
Syntheses of substrates 9, 13 and 14. The details are described
in the ESI.†
Spectroscopic data of enzymatic products 15–25. Products 16,
17 and 19 are known compounds. The spectroscopic data are
described in the ESI.†
Product 15. 1H NMR (C6D6, 400 MHz) d 0.917 (3H, s, Me-
15), 0.958 (d, J = 7.2 Hz, Me-11), 0.977(3H, s, Me-13), 0.98 (m,
H-1), 1.01 (3H, s, Me-14), 1.25 (ddd, 13.0, 12.8, 3.6 Hz, H-3), 1.29
(dd, J = 11.8, 5.2 Hz, H-5), 1.50 (m, H-2), 1.54 (m, H-3), 1.63 (m,
H-2), 1.75 (3H, br s, Me-12), 1.86 (m, H-1), 1.94 (m, H-9), 1.95 (m,
H-6); 2.05(m, H-6), 5.58 (brs, H-7); 13C NMR (C6D6, 100.6 MHz)
d 11.58 (q, C-11), 13.47 (q, C-15), 19.29 (t, C-2), 21.98 (q, C-12),
22.12 (q, C-14), 24.10 (t, C-6), 33.03 (s, C-4), 33.47 (q, C-13), 36.02
(s, C-10), 39.81 (t, C-1), 42.55 (t, C-3), 49.03 (d, C-9), 50.35 (d, C-
5), 121.9 (d, C-7), 135.3 (s, C-8), the assignments of C-12 and C-14
may be interchangeable due to the close values; GCMS (EI) m/z
206 (M+, 18), 191 (13), 124 (33), 109 (100), 82 (58); HREIMS m/z
206.2067 (calcd for C15H26, 206.2035); [a]2D5 -12.5 (c 0.12, C6H6).
Experimental section
Analytical methods
NMR spectra were mainly recorded in C6D6 on a Bruker DMX
600 or DPX 400 spectrometer, the chemical shifts being relative
to the solvent peak dH 7.28 and dC128.0 ppm as the internal
reference for 1H- and 13C NMR spectra, respectively. Some
synthetic intermediates were measured in CDCl3. The chemical
shifts in CDCl3 solution were given according to the internal
solvent peaks of dH 7.26 and dC 77.0 ppm. To further validate
the proposed structures, the NMR data of some enzymic products
were measured in acetone-d6, the solvent peak being referred to
be dH 2.04, dC 29.8 ppm. GC analyses were done 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 on
a JEOL SX 100 spectrometer under electronic impact at 70 eV with
a DB-1 capillary column (0.32 mm ¥ 30 m), the oven temperature
being elevated according to the methods described in the legends
to Fig. 1 and 2. HR-EIMS was measured by direct inlet system.
Specific rotation values were measured at 25 ◦C with a Horiba
SEPA-300 polarimeter.
Product 18. 1H NMR (acetone d6, 400 MHz) d 0.82 (m, H-1),
0.846 (3H, s, Me-14), 0.862 (3H, s, Me-13), 0.87 (m, H-5), 0.888
(3H, d, J = 6.8 Hz, Me-11), 0.974(3H, s, Me-15), 1.01(1H, q, J =
7.2 Hz, H-9), 1.08(3H, s, Me-12), 1.14 (ddd, J = 13.6, 13.6, 4.4 Hz,
H-3), 1.35 (m, H-2), 1.36 (m, H-3), 1.42 (m, H-6), 1.43(m, H-7),
1.59 (m, H-2), 1.59 (m, H-6), 1.68 (m, H-1), 1.76(m, H-7), 2.69 (br
s, OH); 13C NMR (acetone d6, 100.6 MHz) d 7.88 (q, C-11), 14.88
(q, C-15), 19.25 (t, C-6), 19.31 (t, C-2), 22.19 (q, C-14), 31.41 (q,
C-12), 33.88 (s, C-4), 34.00 (q, C-13), 38.65 (s, C-10), 40.82 (t, C-1),
42.77 (t, C-3), 43.62 (t, C-7), 53.58 (d, C-9), 56.89 (d, C-5), 71.84
(s, C-8), the assignments of C-2 and C-6 may be interchangeable
due to the close values; GCMS (EI) m/z 224 (M+, 3), 206 (18),
191 (38), 137 (28), 109 (100), 95 (37), 82 (43), 81 (37), 67 (35);
HREIMS m/z 206.2022 (M+-H2O, calcd for C15H26, 206.2035);
[a]2D5 -32.1 (c 0.037, C6D6), cf. lit. [a]D -12.3 (CHCl3, concentr.
not described31); Rf value on SiO2 plate = 0.54, when developed
with hexane : EtOAc (100/20).
Product 20. 1H NMR (C6D6, 400 MHz) d 0.809 (3H, s, Me-
15), 0.963 (3H, d, J = 6.8 Hz, Me-12), 1.011 (3H, s, Me-11), 1.107
(3H, s, Me-14), 1.24 (m, H-8), 1.27 (dd, 12.0,1.8 Hz, H-10), 1.27
(m, H-6), 1.28 (m, H-7), 1.42(m, H-7), 1.56 (m, H-1), 1.72 (3H,
d, J = 1.2 Hz, Me-13), 1.72 (m, H-6), 1.75 (m, H-1), 2.14 (2H,
m, H-2), 5.36 (br s, H-3); 13C NMR (C6D6, 100.6 MHz) d 16.64
(q, C-12), 16.72 (q, C-15), 18.27 (q, C-13),, 19.28 ((t, C-1), 19.62
(q, C-14), 27.59 (t, C-2), 28.04 (t, C-7), 29.30 (q, C-11), 30.03 (t,
C-6), 36.55 (s, C-9), 38.48 (s, C-5), 42.69 (d, C-8), 52.91 (d, C-10),
121.0 (d, C-3), 143.8 (s, C-4), the carbon signals of C-12 and C-15
may be interchangeable due to the close values; GCMS (EI) m/z
206 (M+, 23), 191 (37), 163 (100), 136 (39), 123 (62), 107 (97), 95
(85), 81 (62), 69 (58); HREIMS m/z 206.2040 (calcd for C15H26,
206.2035); [a]2D5 -10.7 (c 0.14, C6H6).
Incubation condition and purification of enzymatic products
Standard incubation conditions were performed according to
the published protocols.26,30 The cell-free extracts were prepared
as follows. A one liter culture of E. coli encoding the native
SHC was harvested by centrifugation and to the collected pellets
was added 50 cm3 of citrate buffer solution (pH 6.0), and then
subjected to ultrasonication to disrupt the cells. The supernatant
was used for the incubations after removing the cell debris by
centrifugation. One ml of the supernatant contains ca. 200 mg of
the pure SHC. The products and the unreacted substrate analogs
were extracted three times with a mixture of hexane and EtOAc
(100 : 10). An excess of Triton X-100 detergent was removed by
passing a short SiO2 column eluting with hexane/EtOAc (100 : 20).
Each product was purified by SiO2 column chromatography and
normal phase HPLC, as described in the text. It is likely that
the enzymatic products of 15, 16, 19 and 20 vaporize under
reduced pressure. Therefore, the organic solvents dissolving the
Product 21. 1H NMR (C6D6, 600 MHz) d 0.74 (m, H-1), 0.757
(3H,s, Me-20), 0.863 (3H,s, Me-19), 0.89 (m, H-5), 0.914 (3H,s,
Me-18), 1.14 (m, H-3), 1.216 (3H,s, Me-21), 1.23 (m, H-6), 1.41
(m, H-3), 1.44 (m, H-2), 1.52 (bd, J = 12.2 Hz, H-1), 1.59 (m,
H-9), 1.60 (m, H-6), 1.63 (m, H-2), 1.86 (ddd, J = 13.0, 13.0, 4.0
This journal is
The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 440–446 | 445
©

View Article Online
Hz, H-7);, 2.16 (ddd. J = 12.5, 3.2, 3.2 Hz, H-7), 2.54 (2H,d,9.0
Hz, H-11), 6.99 (m, H-14), 7.16 (bd.J = 7.6 Hz, H-17) 7.18 (2H,
m, H-16 & H-15);13C NMR (C6D6, 150.9 MHz) d 14.89 (q, C-20),
18.73 (t, C-6), 19.87 (t, C-2), 20.95 (q, C-21), 21.66 (q, C-19), 22.55
(t, C-11), 33.15 (s, C-4), 33.45 (q, C-18), 36.77 (s, C-10), 39.07 (t,
C-1), 41.48 (t, C-7), 41.99 (t, C-3), 52.12 (d, C-9), 55.97 (d, C-5),
76.77 (s, C-8), 117.5 (d, C-15), 119.9 (d, C-14), 122.4 (s, C-12),
127.6 (d, C-16), 130.1 (d, C-17), 154.0 (s, C-13). EIMS m/z 298
(M+, 53), 191 (62), 107 (84), 69 (100); HREIMS m/z 298.2286
(calcd for C21H30O 298.2297). [a]2D5 + 49.76 (c = 0.4, CHCl3).
281 (33), 189 (77), 107 (33); HREIMS m/z 356.2357 (calcd for
C23H32O3 356.2351). [a]2D5 + 25.5 (c = 0.05, CHCl3),
Acknowledgements
This work was supported by Grant-in-Aid for Scientific Research
from the Ministry of Education, Culture, Sports, Science and
Technology, Japan (No 18380001).
References
Product 22. 1H NMR (CDCl3, 400 MHz) d 0.998(3H, s, Me-
19), 1.01 (3H, s, Me-18), 1.04 (3H, s, Me-20), 1.17(ddd, J =
3.6,13.2,13.2 Hz, H-1), 1.30 (m, H-2), 1.31(m, H-3), 1.41 (m, H-
5), 1.53 (m, H-3), 1.60 (m, H-2), 1.77(3H, s, Me-21), 1.94(bd, J =
11.6 Hz, H-1), 2.08 (2H, m, H-6), 2.56 (bd, J = 8.0 Hz, H-9), 2.76
(d,15.2 Hz, H-11), 2.93 (m, H-11), 5.57 (bs, H-7), 6.39 (m, H-14),
6.98 (t, J = 7.6 Hz, H-16), 7.08 (t, J = 7.6 Hz, H-15), 7.32 (d, J =
7.6 Hz, H-17);13C NMR (CDCl3, 100.6 MHz) d 14.09 (q, C-20),
19.30 (t, C-2), 22.13 (q, C-19), 22.69 (q, C-21), 24.97 (t, C-6), 26.49
(t, C-11), 33.16 (s, C-4), 33.47 (q, C-18), 37.17 (s, C-10), 39.72 (t,
C-1), 42.58 (t, C-3), 50.45 (d, C-5), 54.51 (d, C-9), 115.3 (d, C-14),
120.7 (d, C-16), 122.4 (d, C-7), 127.7 (d, C-15), 130.2 (d, C-17),
130.3 (s, C-12), 135.8 (s, C-8), 153.8 (s, C-13); EIMS m/z 298 33),
191 (100), 109 (72); HREIMS m/z 298.2292 (calcd for C21H30O
298.2297). Specific optical rotation was not determined due to the
very small amount available.
1 Reviews: (a) T. Hoshino and T. Sato, Chem. Commun., 2002, 291–301;
(b) R. A. Yoder and J. N. Johnston, Chem. Rev., 2005, 105, 4730–4756;
(c) K. U. Wendt, G. E. Schulz, E. J. Corey and D. R. Liu, Angew. Chem.,
Int. Ed., 2000, 39, 2812–2833; (d) M. J. R. Segura, B. E. Jackson and
S. P. T. Matsuda, Nat. Prod. Rep., 2003, 20, 304–317; (e) G. Siedenburg
and D. Jendrossek, Appl. Environ. Microbiol., 2011, 77, 3905–3915.
2 T. Hoshino, Y. Kumai, I. Kudo, S. Nakano and S. Ohashi, Org. Biomol.
Chem., 2004, 2, 2650–2657.
3 T. Hoshino, Y. Kumai and T. Sato, Chem.–Eur. J., 2009, 15, 2091–2100.
4 J. Cheng and T. Hoshino, Org. Biomol. Chem., 2009, 7, 1689–1699.
5 S. Huneck, Z. Naturforsch., 1967, 22b, 462–463.
6 M. Toyota, Y. Ooiso, T. Kusuyama and Y. Asakawa, Phytochemistry,
1994, 35, 1263–1265.
7 V. Armstrong, A. F. Barrero, E. J. Alvarez-Manzaneda, M. Corte´s and
B. Sepu´lveda, J. Nat. Prod., 2003, 66, 1382–1383.
8 M. Kinoshita, M. Ohtsuka, D. Nakamura and H. Akita, Chem. Pharm.
Bull., 2002, 50, 930–934.
9 Y. Ohta, N. H. Andersen and C. B. Liu, Tetrahedron, 1977, 33, 617–628.
10 Y. Asakawa, M. Toyota, A. Ueda, M. Tori and Y. Fukazawa,
Phytochemistry, 1991, 30, 3037–3040.
11 M. Tori, N. Hamada, M. Sono, Y. Sono, M. Ishikawa, K. Nakashima,
T. Hashimoto and Y. Asakawa, Tetrahedron Lett., 2000, 41, 3099–3102.
12 R. M. Carman and W. Craig, Aust. J. Chem., 1971, 24, 361–370.
13 (a) T. Kuzuyama, J. P. Noel and S. B. Richard, Nature, 2005, 435, 983–
987; (b) T. Itoh, K. Tokunaga, Y. Matsuda, I. Fujii, I. Abe, Y. Ebizuka
and T. Kushiro, Nat. Chem., 2010, 2, 858–864.
14 (a) D. M. Roll, J. K. Manning and G. T. Carter, J. Antibiot. (Tokyo),
1998, 51, 635–639; (b) H. Tsujimori, M. Bando and K. Mori, Eur. J.
Org. Chem., 2000, 2, 297–302.
15 K. Torigoe, N. Wakasugi, N. Sakaizumi, T. Ikejima, H. Suzuki, K.
Kojiri and H. Suda, J. Antibiot. (Tokyo), 1996, 49, 314–317.
16 T. Hoshino, K. Shimizu and T. Sato, Angew. Chem., Int. Ed., 2004, 43,
6700–6703.
Product 23. 1H NMR (CDCl3, 600 MHz) d 0.820 (3H, s Me-
20), 0.834 (3H, s, Me-19), 0.885 (3H, s, Me-18), 1.17 (m, H-1), 1.19
(m, H-3), 1.20 (m, H-5), 1.37 (m, H-6), 1.41 (d,J = 12.0 Hz, H-3),
1.52 (m, H-2), 1.63 (m, H-2), 1.76 (m, H-6), 1.91(d, J = 12.0 Hz,
H-1), 2.03 (ddd, J = 4.2,12.7,12.7 Hz), 2.22 (br m, H-9), 2.38 (d, J =
12.7 Hz. H-7), 2.75 (2H, d, J = 8.8 Hz, H-11), 4.71 (s, H-21), 4.82
(s, H-21), 6.72 (d, J = 7.5 Hz, H-14), 6.83 (t, J = 7.5 Hz, H-16),
7.03 (t, J = 7.51 Hz, H-15), 7.10 (d, J = 7.5 Hz, H-17); 13C NMR
(CDCl3, 150.9 MHz) d 19.47 (t, C-2), 21.76 (q, C-19), 23.71 (t,
C-11), 24.48 (t, C-6), 33.64 (q, C-18), 33.67 (s, C-4), 38.29 (t, C-7),
39.20 (t, C-1), 40.29 (s, C-10), 42.21 (t, C-3), 55.74 (d, C-5), 56.08
(d, C-9), 107.5 (t, C-21), 115.2 (d, C-14), 120.5 (d, C-16), 126.6
(d, C-15), 128.3 (C-12), 129.9 (d, C-17), 14.50 (q, C-20), 149.0 (s,
C-8), 153.6 (s, C-13); GCMS (EI) m/z 298 (M+, 50), 191 (65), 137
(48), 107 (100), 81 (50), 69 (62); HREIMS m/z 298.2292 (calcd for
C21H30O 298.2297). Specific optical rotation was not determined
due to a very small amount available.
17 P. Weyerstahl, H. Marschall, S. Schneider and G. C. Subba, Flavour
Fragrance J., 1995, 10, 179–185.
18 J. Paolini, A. Falchi, Y. Quilichini, J-M Desjobert, M-C De Cian, L.
Varesi and J. Costa, Phytochemistry, 2009, 70, 1146–1160.
19 S. Cui, S. Tan, G. Ouyang, S. Jiang and J. Pawliszyn, J. Chromatogr.,
2009, B877, 1901–1906.
20 P. P. Reddy, R. R. Rao, J. Shashidhar, B. S. Sastry, J. M. Rao and K. S.
Babu, Bioorg. Med. Chem. Lett., 2009, 19, 6078–6081.
21 R. Kowalski, Food Chem., 2009, 112, 820–830.
22 J. R. Hlubucek, A. J. Aasen, S-O Almqvist and C. R. Enzell, Acta
Chem. Scand., Ser. B, 1974, B28, 289–294.
23 C. R. Enzell, A. Nasiri, I. Forsblom, E. Johansson, C. Lundverg, J.
Arnarp and L. Bjo¨rk, Acta Chem. Scand., 1995, 49, 375–379.
24 B. Corbier, C. Ehret, E. Giraudi and G. Pelerin, Dev. Food Sci., 1988,
18, 483–494.
25 T. Iida, E. Yoshii, K. Takeshima, T. Nakata, Y. Tani and T. Oishi,
Yakugaku Zasshi, 1980, 100, 915–919.
Product 24 acetate (25). 1H NMR (C6D6, 400 MHz) d 0.687
(3H, s, Me-20), 0.83 (m, H-1), 0.86(m, H-5), 0.931(6H, s, Me-18 &
Me-19)), 1.167 (3H, s, Me-21), 1.37 (m, H-1), 1.47(m, H-9), 1.52
(2H, m, H-6), 1.64 (m, H-2), 1.79 (m, H-7), 1.82(m, H-2), 1.87
(3H, s, Me-23), 2.13(m, H-7), 2.44 (2H, m, H-11), 4.70 (dd, J =
12.0, 4.8 Hz, H-3), 6.99(m, H-14), 7.15(bd. J = 8.0 Hz, H-17),
7.18(2H, m, H-15 &H-16); 13C NMR (C6D6, 100.6 MHz) d 14.88
(q, C-20), 16.78 (q, C-19), 19.36 (t, C-6), 20.77 (q, C-21), 20.81
(q, C-23), 22.49 (t, C-11), 23.84 (t, C-2), 28.02 (q, C-18), 36.30
(s, C-10), 36.80 (t, C-1), 37.76 (s, C-4), 41.22 (t, C-7), 51.59 (d,
C-9), 54.80 (d, C-5), 76.43 (s, C-8), 80.06 (d, C-3), 117.5 (d, C-15),
120.0 (d, C-14), 122.4 (s, C-12), 127.6 (d, C-16), 130.0 (d, C-17),
154.0 (s, C-13), 169.8 (s, C-22): the assignments of C-21 and C-23
may be exchangeable due to the close values. EIMS m/z 356 (100),
26 T. Hoshino, S. Nakano, T. Kondo, T. Sato and A. Miyoshi, Org. Biomol.
Chem., 2004, 2, 1456–1470.
27 H. Tanaka, H. Noguchi and I. Abe, Org. Lett., 2005, 7, 5873–
5876.
28 H. Tanaka, H. Noma, H. Noguchi and I. Abe, Tetrahedron Lett., 2006,
47, 3085–3089.
29 Q. Xiong, X. Zhu, W. K. Wilson, A. Ganesan and S. P. T. Matsuda, J.
Am. Chem. Soc., 2003, 125, 9002–9003.
30 T. Sato and T. Hoshino, Biosci., Biotechnol., Biochem., 1999, 63, 2189–
2198.
31 G. Ohlof and W. Giersch, Croatica Chem. Acta, 1985, 58, 491–509.
446 | Org. Biomol. Chem., 2012, 10, 440–446
This journal is
The Royal Society of Chemistry 2012
©