Zeolite NaY-promoted monocyclization of epoxy polyene terpenes: A unique route for the direct synthesis of incompletely cyclized naturally occurring terpenols

Article abstract of DOI:10.1002/adsc.200800180

A variety of epoxy polyene terpenes cyclize readily by confinement within zeolite NaY to form primarily products of monocyclization. The monocyclization pathway is highly predominant, irrespectively of the side chain of the epoxy terpene, while the monocyclic products possess regioselectively an exomethylenic double bond. The selective monocyclization in the case of epoxyfarnesyl acetate, epoxyfarnesylacetone or 2,3-epoxysqualene, provides a direct route to the synthesis of a variety of natural products, such as elengasidiol, farnesiferols B-D, achilleol A, camelliol C and to four farnesylacetone-derived metabolites isolated from the brown algae Cystophora monoliformis. The optical rotation of achilleol A derived from the cyclization of (S)-2,3-epoxysqualene matches with that of the natural product, thus the absolute configuration of achilleol A was established as 1S,3R. From the mechanistic point of view, the NaY-promoted cyclization of 9,10-epoxygeranylacetone, selectively deuterium labelled at the C-10 methyl group, is >97% stereoselective with respect to the topicity of the gem-dimethyl group. This result is in agreement with a concerted mechanism. Finally, we have proved through labelling experiments, for the first time, that the biomimetic transformation of epoxy polyene terpenes to 2,3,4-trimethylcyclohexanones upon acid catalysis is a highly stereoselective process. Thus, the less hindered gem-methyl group on the epoxide functionality becomes α- to the carbonyl in the final isomerized product.

Full text of DOI:10.1002/adsc.200800180

FULL PAPERS
DOI: 10.1002/adsc.200800180
Zeolite NaY-Promoted Monocyclization of Epoxy Polyene
Terpenes: A Unique Route for the Direct Synthesis of
Incompletely Cyclized Naturally Occurring Terpenols
Constantinos Tsangarakis,a Christos Raptis,a Elias Arkoudis,a
and Manolis Stratakisa,*
a
Department of Chemistry, University of Crete, Voutes 71003 Iraklion, Greece
Fax : (+30)-2810-545-001; phone: (+30)-2810-545-087; e-mail: stratakis@chemistry.uoc.gr
Received: March 24, 2008; Published online: June 23, 2008
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: A variety of epoxy polyene terpenes cy- point of view, the NaY-promoted cyclization of 9,10-
clize readily by confinement within zeolite NaY to epoxygeranylacetone, selectively deuterium labelled
form primarily products of monocyclization. The at the C-10 methyl group, is >97% stereoselective
monocyclization pathway is highly predominant, irre- with respect to the topicity of the gem-dimethyl
spectively of the side chain of the epoxy terpene, group. This result is in agreement with a concerted
while the monocyclic products possess regioselective- mechanism. Finally, we have proved through label-
ly an exomethylenic double bond. The selective ling experiments, for the first time, that the biomim-
monocyclization in the case of epoxyfarnesyl acetate, etic transformation of epoxy polyene terpenes to
epoxyfarnesylacetone or 2,3-epoxysqualene, provides 2,3,4-trimethylcyclohexanones upon acid catalysis is
a direct route to the synthesis of a variety of natural a highly stereoselective process. Thus, the less hin-
products, such as elengasidiol, farnesiferols B–D, dered gem-methyl group on the epoxide functionality
achilleol A, camelliol C and to four farnesylacetone- becomes a- to the carbonyl in the final isomerized
derived metabolites isolated from the brown algae product.
Cystophora monoliformis. The optical rotation of
achilleol A derived from the cyclization of (S)-2,3-
epoxysqualene matches with that of the natural Keywords: green chemistry; heterogeneous catalysis;
product, thus the absolute configuration of achilleol natural products; reaction mechanisms; terpenoids;
A was established as 1S,3R. From the mechanistic zeolites
Introduction
a highly stereoselective manner. On many occasions
organic chemists have attempted to reproduce this
The acid-catalyzed cyclization of epoxy polyenes has fascinating enzymatic reaction using acids as catalysts
attracted the interest of the organic chemists for the with varying degrees of success. Brønsted acids have
past 50 years,[1] especially due to the early discovery very rarely been used as promoters,[4] while super
that this reaction is involved in the biosynthesis of a acids (such as FSO3H)[5] which are superior catalysts
vast number of terpenols. The most known among in terpene cyclization fail to provide cyclization of
them is the oxidosqualene cyclase-catalyzed cycliza- epoxy terpenes, leading to solvolysis products of the
tion of 2,3-epoxysqualene to lanosterol, the biogenetic epoxide, instead. On the other hand, Lewis acids such
precursor of cholesterol and other triterpenoids.[2] The as BF3,[6] SnCl4,[6] ZrCl4,[7] FeCl3,[8] Sc
(OTf)3,
[12]
G
[9]
[10]
cyclase cavity provides an ideal compartment, for the
Bi
(OTf)3, InBr3,[11] and, especially, MeAlCl2 have
N
acyclic epoxy polyene to adopt a suitable conforma- been widely used, often providing efficient polycycli-
tion, so that upon protonation of the epoxide func- zation with good diastereoselectivity. In addition to
tionality by an aspartic acid residue,[3] iterative inter- the cationic cyclization provided by acidic catalysts, a
molecular nucleophilic attacks of the proximal double Ti(III)-promoted free-radical methodology has been
A
bonds to the protonated epoxide or to the intermedi- developed[13] for the cascade polycyclization of epoxy
ate carbocations, provide fused polycyclic products in
Adv. Synth. Catal. 2008, 350, 1587– 1600
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1587

Constantinos Tsangarakis et al.
FULL PAPERS
terpenes, and found significant applications in the syn- Results and Discussion
thesis of natural products.[14]
Recently, we examined the efficiency of acidic Cyclization of Small Epoxy Terpenoids
porous materials such as zeolites, as hosts and cata-
lysts as well, for the cyclization of polyene terpenoids. As a continuation of our studies, we explored the effi-
Our concept was that the zeolite cavity would provide ciency of NaY as a catalyst for the cyclization of
the necessary acidic sites for initiating the cationic epoxy polyene terpenes. The use of acidic zeolites or
cyclization, whereas substrate confinement in a re- other porous materials for the cyclization of epoxy
stricted space[15] would decrease its conformational terpenoids is very limited in the literature. Sen and
mobility and thus provide product(s) stereocontrol. co-workers[20] have used the H-form of the small-pore
The remarkable stereocontrol in the enzyme-cata- zeolites A and ZSM-5 for the cyclization of suitable
lyzed cyclization of terpenes is primarily attributed to silyl-substituted epoxy polyenes. A moderate to good
the suitable substrate conformation by confinement product selectivity was achieved. It was postulated
within the restricted environment of the enzyme that cyclization initiates at the opening of the zeolite
cavity. Zeolites are mixed aluminosilicates that cata- pores, as the size of the substrates does not allow
lyze reactions[16] by confining the substrates within them to diffuse into the interior of the cages. By con-
“active site” cavities, and have found significant appli- trast to the small pore zeolites A and ZSM-5, we an-
cations in heterogeneous photocatalysis.[17] We have ticipated complete adsorption (confinement) of epoxy
chosen faujasites (zeolites MY, M=cation) as cata- polyene terpenes possessing up to 30 carbon atoms
lysts. They possess a 3D framework which consists of (triterpenes) within NaY supercages, as it well known
supercages, whose interior diameter is approximately that the faujasite cavities can host even steroids.[21] In-
13 Š. The supercages are interconnected by four itially, the cyclization of the small acyclic terpenoids
“windows” which are tetrahedrally distributed around 3, 5a–d and 10a, b shown in Table 1 was examined.[22]
each cage and have a pore entrance diameter of The reactions were carried out by adding a hexane so-
around 7.5 Š. Such cage dimensions are ideal to host lution containing a suitable amount of the epoxy ter-
relatively large organic molecules, such as higher ter- penoid (see discussion below) to the dry NaY, and the
penes. Zeolite NaY was our first choice, as it is a very resulting slurry was left reacting for a certain period
mild acidic catalyst due to a limited amount of intra- of time (Table 1).
zeolite Brønsted sites.[18] We were pleased to find[19]
Hexane was the proper reaction solvent as due to
that small acyclic terpenes undergo a clean cyclization its non-polar nature it allows the facile diffusion of
reaction under essentially extremely mild conditions. the epoxides into the polar zeolite cages. That was
For example, geranyl acetate (1) forms at the initial clearly evident by GC examination of the supernatant
reaction stages primarily g-cyclogeranyl acetate (1a) hexane which revealed the absence of any reacting
which on prolonged reaction time isomerizes to a-cy- epoxide or product(s) during the course of the reac-
clogeranyl acetate (1b, Scheme 1). The most remark- tion or at the end. This implies a complete substrate
able finding of those initial studies was that geranyl- confinement within the zeolite pores, under the low
acetone (2) and its geometrical isomer (nerylacetone) substrate loading levels used in our experiments. Sub-
provide a direct one-pot highly diastereoselective cas- sequently, the zeolite was filtered and treated with
cade to a-ambrinol (2c), a valuable compound in the methanol to extract the adsorbed products (see Ex-
fragrance industry, through the intermediate forma- perimental Section).
tion, but not isolable, g-cyclogeranylacetone (2a).
Scheme 1. Zeolite NaY-promoted cyclization of small terpenoids.
1588
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1587– 1600

Zeolite NaY-Promoted Monocyclization of Epoxy Polyene Terpenes
FULL PAPERS
Table 1. Cyclization of small acyclic epoxy terpenoids promoted by zeolite NaY.
Substrate
X
Reaction time [min]
Yield
Products (relative ratio)
exo/endo[a]
3
10
10
10
15
35
45
78%
76%
92%
83%
78%
85%
4
-
5a
5b
5c
10a
10b
OH
CH2COMe
CH2CH
OAc
CH2COMe
6a/7a/8a=70/20/10
6b/7b/8b=70/25/5
8/1
12/1
4/1
8/1
9/1
(OAc)Me[b]
6c[b]/7c[b]/8c[b]/9c[c] =55/22/8/15
11a/12a/13a/4=40/20/10/30
11b/12b/7b/13b=70/6/2/22
[a]
We define as exo the regioisomer for which the double bond is exocyclic (disubstituted), and endo that in which the
double bond is endocyclic (trisubstituted).
Mixture of two diastereomers in ratio 1/1.
[b]
[c]
Mixture of two diastereomers in a ratio ~3/1.
We were gratified to find that, with the exception geranyl-derivatives 5a, b per 1 g of dry NaY), the re-
of 6,7-epoxygeranyl acetate (3), NaY promotes the actions went to completion on heating at 708C for 1
fast cyclization of the epoxy terpenoids to form ste- hour. The combined product yields vary from 76–
reoselectively and regioselectively mainly the exome- 92%. The isomerization of the predominantly formed,
thylenic cyclohexanols 6a–c or 11a, b (Scheme 2). The in the cases of regioisomeric products 6 or 11, exo- to
stereochemistry of the products depends on the con- the endo-double bond occurs slowly under the reac-
figuration of the adjacent nucleophilic double bond. tion conditions. Thus, prolonged intrazeolite treat-
Thus, alkenes with the (E)-configuration afford cis ment of the epoxides (1–2 h) results in a drop of the
cyclization products (6), while those with (Z), afford ratio exo/endo double bond isomers from a typical 8–
trans (11). The bicyclic ethers (7 or 12) and the allylic 12/1 value to approximately 3–6/1.
alcohols (8 or 13) are formed as side-products. The
From the practical point of view, the cyclization
stereochemistry of these products depends again on procedure describe herein is extremely simple, abso-
the configuration of the nucleophilic double bond. To lutely mild, and environmentally friendly, as toxic and
ensure complete reactant adsorption within the zeo- difficult to handle Lewis acids are avoided, no aque-
lite cavities, low loading levels (~0.1 mmol of sub- ous work-up is necessary, while the yields are quite
strate per 1 g of dry NaY) were used. While the ger- acceptable. Furthermore, this methodology provides a
anyl derivatives (3, 5a–c) react smoothly at ambient powerful and direct route to a series of synthetically
temperature, going to completion within 10–15 min, useful exomethylenic cyclohexanols and bicyclic
the neryl derivatives (10a, b) require up to 45 min re- ethers. For instance, cyclized products 6a,[23] 7a,[23] and
action time. At higher loading levels (1 mmol of the 6b[13] have been used as valuable blocks in terpenoid
Adv. Synth. Catal. 2008, 350, 1587– 1600
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1589

Constantinos Tsangarakis et al.
FULL PAPERS
Another interesting feature of the results presented
in Table 1, is the formation of the 2,3,4-trimethylcy-
clohexanone derivative 9c in 15% relative yield, ob-
served only in the specific case of epoxide 5c. Com-
pound 9c appears as a mixture of 2 diastereomers in a
relative ratio ~3/1, and was characterized by trans-
forming it to the known diketone 16,[27] through PCC
oxidation of its hydrolysis product, alcohol 15
(Scheme 2). The relative stereochemistry on the ste-
reogenic center of the side chain for the diastereo-
mers 9c could not be assigned. Compounds possessing
the 2,3,4-trimethylcyclohexanone moiety are well
known to arise as by-products in the Lewis acid-cata-
lyzed cyclization of epoxy polyene terpenes,[28]
through the widely postulated mechanism shown in
Scheme 2. It is also notable that several natural prod-
ucts such as cordiaquinone B,[29] striatenone,[30] and as-
cochlorin[31] bear the characteristic 2,3,4-trimethylcy-
clohexanone moiety, indicative that their biosynthesis
occurs through such a delicate pathway (1,2-succesive
Scheme 2. Structural characterization of by-product 9c, and
a plausible mechanism for its formation.
synthesis. It is important to note that the regioselec- shifts of two hydrides and one methyl group).
tive formation of exomethylenic cyclized products,
such as 6a–c or 11a, b, under acidic conditions, can be
achieved only by using silicon as a controlling ele- Cyclization of Higher Epoxy Polyene Terpenes
ment, in the cyclization of terpenes bearing a trialkyl-
silyl group on the allylic methyl group where termina- In order to prove the worth of our new cyclization
tion of the cyclization is pursued.[24] This methodolo- protocol in organic synthesis, we extended our stud-
gy, however, requires additional synthetic steps for ies[22] to the examination of the cyclization of higher
the introduction of the trialkylsilyl group on the ter- epoxy polyene terpenes, with 10,11-epoxyfarnesyl ace-
pene backbone.
tate (17) being the first target. The intrazeolite reac-
It was somewhat surprising that while 6,7-epoxyger- tion of 17 went to completion after 15 min to afford
aniol (5a) cyclizes to form mainly 6a, the correspond- selectively the monocyclization products 18 and 19,
ing acetate 3 does not afford any cyclization products. the bicyclic ether 20, and the allylic alcohol 21 (analo-
Instead, (4E,6E)-2,6-dimethylocta-4,6-dien-3-one (4) gous to the by-products 8 or 13 of Table 1), in a rela-
was produced in 78% isolated yield. Dienone 4 along tive ratio 18/19/20/21=51/6/30/13, and in 75% com-
with its (4E,6Z)-geometrical isomer are known[25] to bined isolated yield (Scheme 3). The very important
arise upon treatment of linalool oxide with 25% result of this reaction was that only monocyclized
H2SO4, or through the acid-catalyzed isomerization of products were formed, despite the presence of addi-
6,7-epoxy myrcene (14). In our case, formation of 4 tional double bond that could provide access to prod-
from 3 can be envisioned to have occurred through ucts of dicyclization, as occurs upon treatment of 17
an epoxide to ketone isomerization in combination with Lewis acids.[32] The non-separable mixture of the
with an elimination of the acetate functionality and regioisomeric alcohols 18 and 19 was easily isolated
subsequent diene isomerization. At this stage, how- from the crude reaction mixture by column chroma-
ever, we do not want to speculate as to why we see tography, and hydrolyzed almost quantitatively, using
this switch in mode of reaction upon going from 5a to K2CO3 in methanol, to afford the racemic sesquiterpe-
its acetoxy analogue 3, especially in the light of our nol elegansidiol (22),[33] along with its minor endocy-
observation that 14 does not form any 2 upon NaY clic isomer 23 (ratio 22/23~8/1). There are three
treatment, but primarily the isomeric cyclization prod- known syntheses of elegansidiol so far,[34] however,
ucts 14a and 14b[26] in a ~3/1 relative ratio, and in applying multiple steps compared to our approach
46% isolated yield after column chromatography.
and by using less or non-bioinspired approaches. Ele-
1590
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1587– 1600

Zeolite NaY-Promoted Monocyclization of Epoxy Polyene Terpenes
FULL PAPERS
Scheme 3. Monocyclization of 10,11-epoxyfarnesyl acetate (17) promoted by NaY: synthesis of elegansidiol and farnesiferols
B–D.
gansidiol might be directly expected from the cycliza- mides 25 and 26. Finally, alkylation of the potassium
tion of 10,11-epoxyfarnesol (24). Epoxide 24 (pre- salt of 7-hydroxycoumarin (umbelliferone) with the
pared by hydrolysis of 17) was treated with NaY allylic bromides 25 and 26 yielded an inseparable mix-
under identical conditions to 17. Due to the relative ture of farnesiferol B (27) and farnesiferol D (28) in a
lability of the allylic alcohol under the acidic zeolite relative ratio of ~8/1, and in 87% isolated yield from
conditions, substantial dehydration occurred either on 18/19 (over 3 steps). In addition, the analogous 3-step
the reacting epoxy alcohol or on the products. The de- transformation to the minor bicyclic ether 20 fur-
hydration products rather undergo several skeletal re- nished farnesiferol C (31) in 86% overall yield.
arrangements/isomerizations, thus generating a com-
The clean and primary formation of monocyclized
plex reaction mixture. Nevertheless, an approximately products in the case epoxy farnesyl acetate was en-
15–20% of elegansidiol was seen in the crude reaction couraging, since a variety of naturally occurring, in-
mixture after 10 min reaction time (22/23 ~7/1), thus, completely cyclized terpenols, among them nassauvir-
this direct approach to elegansidiol suffers from poor evolutin A,[36] achilleol A,[37] or its double bond
yield.
isomer camelliol C,[38] could derive from the incom-
The closely related structure of the products 18 and plete cyclization of suitable epoxy polyene precursors
19 with farnesiferols B–D[27,35] prompted us to achieve under NaY treatment conditions. Thus, for a further
their synthesis (Scheme 3). Thus, reaction of the re- exploration, we turned our attention to a mixture of
gioisomeric mixture 22 and 23 with PBr3 (1.1 equiv.) farnesylacetone-derived metabolites co-isolated[39]
afforded selectively the corresponding allylic bro- from the brown algae Cystophora monoliformis, as
Adv. Synth. Catal. 2008, 350, 1587– 1600
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1591

Constantinos Tsangarakis et al.
FULL PAPERS
Scheme 4. NaY-promoted cyclization of epoxy trans,trans-farnesylacetone (32).
they could readily derive from the intrazeolite cycliza- tion of new unidentified triterpene alcohols (most
tion of 13,14-epoxy-trans,trans-farnesylacetone (32).[40] probably dicyclization or polycyclization products) in
Treatment of 32 with NaY, afforded a mixture of 33, ~10–15% relative yield, with accompanying decrease
34, 35 and 36 in 78% isolated yield, and in a relative of the relative yield of the achilleol A/camelliol C
ratio of 53/7/35/5, respectively (Scheme 4). We were fraction. Under those conditions, the relative yield of
pleased to realize that all intrazeolite products 33–36, achilleol A/camelliol C was ca. 30%, while their ratio
along with trans,trans-farnesylacetone and epoxide 32, 38/39 became almost 1/1. Recently, Matsuda and co-
belong to the family of those metabolites. The major workers reported[44] that 2,3-epoxysqualene affords
33 and 34 species (g- and a-cystophorol, respective- very slowly products of cyclization on treatment with
ly)[41] were co-isolated after column chromatography the slightly acidic silica gel (0.02% of 2,3-epoxysqua-
as an inseparable mixture, while the minor 35 and 36 lene undergoes cyclization during column chromatog-
species were also isolated, and their NMR spectro- raphy purification). This SiO2-derived product mix-
scopic data matched with those of the natural prod- ture consists primarily of polycyclized structures, with
ucts. This result provides another example for the achilleol A and camelliol C, not detected among
unique ability of NaY as a powerful and highly selec- them. In our opinion, the change in product selectivi-
tive cyclization catalyst towards the synthesis of natu- ty for the cyclization of 2,3-epoxysqualene reflects the
rally occurring compounds, employing a simple and different environment where cyclization takes place
biomimetic strategy.
(within cavities for NaY, and on the surface for SiO2).
The methodology presented herein is a fast route
Finally, we extended the scope and limitation of the
zeolite-promoted cyclization, by studying the reaction to the synthesis of achilleol A and camelliol C, essen-
of the triterpene 2,3-epoxysqualene (37). Reaction of tially within a two-step process from squalene. These
37 (0.1 mmol) with 1 g activated NaY yielded after triterpene alcohols (as a mixture) have been prepared
10 min a mixture of achilleol A/camelliol C (38 and by Barrero and co-workers[45] applying a multi-step
39, respectively), the bicyclic ether 40[42] and the natu- sequence using geranylacetone as the starting materi-
rally occurring allylic alcohol 41.[43] The relative ratio al. On treatment of 2,3-epoxysqualene with Lewis
(38+39)/40/41 was 55/40/15 and the combined isolat- acids mainly tricyclic products are formed,[46] as well
ed yield 72% (Scheme 5). Achilleol A[37] and camel- as, in the presence of Ti(III) species[13a] (radical-initi-
G
liol C[38] arose as an inseparable mixture in a relative ated cyclization). Additionally, by means of biochemi-
ratio of ~4/1. They were easily separated from the cal engineering, through specific mutation of the
rest of the crude reaction mixture by flash chromatog- amino acid sequence of the cyclases, it was possible to
raphy. Prolonged treatment of 2,3-epoxysqualene alter the typical 2,3-oxidosqualene to lanosterol route
within NaY (1 hour, 208C) result to the decrease of with formation of partially cyclized terpenes such as
the overall yield to 58%, and to the additional forma- achilleol A and camelliol C.[47] The current results in-
1592
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1587– 1600

Zeolite NaY-Promoted Monocyclization of Epoxy Polyene Terpenes
FULL PAPERS
Scheme 5. Cyclization of 2,3-epoxysqualene (37) promoted by NaY.
dicate that NaY may be applied as a general mono-
To establish the unknown absolute configurations
cyclization catalyst for epoxy polyene terpenoids re- of achilleol A and camelliol C, we prepared the opti-
gardless of if the side chain is tethered to the epoxy cally active (S)-2,3-oxidosqualene (S-37)[48] in 94% er,
geranyl moiety. To the best of our knowledge, zeolite by asymmetric dihydroxylation of squalene,[49] to form
NaY is the only acidic catalyst that provides selectively the (R)-2,3-diol 42 in 12% isolated yield, followed by
products of monocyclization with epoxy polyene ter- treatment of the diol with tosyl chloride under basic
penes, as any Lewis acid gives rise to products of poly- conditions (Scheme 6). The NaY-promoted cyclization
cyclization.
of (S)-37 afforded the expected reaction products
Scheme 6. Absolute configurations of (À)-achilleol A and (À)-camelliol C.
Adv. Synth. Catal. 2008, 350, 1587– 1600
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1593

Constantinos Tsangarakis et al.
FULL PAPERS
shown in Scheme 5. The achilleol A/camelliol C frac- Mechanistic Studies
tion was isolated (ratio achilleol A/camelliol C=4/1),
and showed a specific rotation of [a]D: À11.58 (c 0.4 a) gem-Dimethyl Group Topicity during the
in CHCl3). This establishes the absolute configuration Cyclization
of the natural products as (1S,3R) for achilleol A[50]
and (1S,5R) for camelliol C. The specific rotation of The enzyme-catalyzed cyclization of epoxy polyene
the naturally occurring achilleol A is [a]D: À10.98 (c terpenes is widely accepted to occur through a con-
0.9 CHCl3),[37] while for camelliol C, [a]D: À12.98 (c certed mechanism. For example, Hoshino and co-
0.2 in CHCl3).[38a] The similarity in the absolute con- workers[51] studied the squalene-hopene cyclase-cata-
figuration of the naturally occurring achilleol A and lyzed cyclization of a specifically deuterium-labelled
camelliol C with that on the A ring of lanosterol, 2,3-epoxysqualene, and found >97% topicity in terms
clearly indicates that both are incompletely cycliza- of the gem-methyl group on the product, 3b-hydroxy-
tion products of (S)-2,3-epoxysqualene, the biogenetic hopene. This stereochemical outcome provides strong
precursor of lanosterol and related polycyclized triter- evidence for a concerted reaction. Regarding the re-
pene alcohols.
action under Lewis acid conditions, van Tamelen[52]
The selective formation of monocyclized exomethy- came to similar conclusions, by studying the cycliza-
lenic regioisomers in all epoxy terpenoids studied tion of a specifically deuterium-labelled epoxy farne-
throughout this manuscript could be envisioned to syl acetate. Finally, Corey and Staas[53] proposed a
occur through a kinetically-driven deprotonation of concerted mechanism on the basis of kinetic data,
the monocyclized carbocation (Scheme 7), by the using a protic acid (dichloroacetic acid) as a catalyst,
À À
À À
oxygen atom of the Si O Si or Si O Al moieties in to mimic the Brønsted nature of the acidic initiator in
the interior walls of the zeolite cages. The hydrogen cyclases (aspartic acid).
atoms on the methyl group (Ha) are more accessible
To shed light on the intrazeolite reaction mecha-
for abstraction by the “zeolite wallꢁ relative to the Hb nism we prepared the deuterium labelled epoxy gera-
hydrogen atoms on the cyclohexyl ring, primarily due nylacetone 47 in 93% diastereomeric purity. The syn-
to steric reasons. Thus, the zeolite cavity can be con- thesis was accomplished in four trivial steps as shown
sidered as a bi-functional catalyst (acidic initiation in Scheme 8, from geraniol-8,8,8-d3 (43) that was
and basic deprotonation). The isomerization of the readily available from our previous studies.[19b] Thus,
exo to the endo regio-isomer is a slow process within 43 was transformed to the labelled bromide 44, which
the timescale of our experiments (ca. 10 min), and be- alkylated ethyl acetoacetate, and the produced 45 was
comes significant only on prolonged reaction time saponified/decarboxylated to form the geranylace-
(more than an hour). Finally, we point out that, the tone-d3 (46).
fast deprotonation of the monocyclized carbocation
Treatment of 47 with NaY gave the expected cy-
prevents further nucleophilic attack by any adjacent clized products 48 and 49 (Scheme 9) in a relative
double bonds to form product(s) of polycyclization.
Scheme 7. Mechanistic arguments on the selective and regio-
selective monocyclization of epoxy polyene terpenoids pro-
moted by NaY.
Scheme 8. Synthesis of the deuterium-labelled epoxy
geranylacetone 47.
G
1594
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1587– 1600

Zeolite NaY-Promoted Monocyclization of Epoxy Polyene Terpenes
FULL PAPERS
Scheme 9. Stereochemistry for the cyclization of 47@NaY, and the proposed mechanism.
ratio of ~1/3, for which the presence of one diastereo- b) Stereochemistry in the Transformation of Epoxy
1
mer was clearly evident by H NMR in terms of the Polyene Trpenes to 2,3,4-Timethylcyclohexanones
gem-dimethyl group topicity. More specifically, for
both 48 and 49, it was established through nOe ex- Having in our hands the deuterium-labeled epoxide
periments (see below the resonances of the gem- 47, we undertook an examination of the stereochemis-
methyls for the perprotio 6b and 7b) that the CD3 try regarding the formation of the 2,3,4-trimethylcy-
group is cis-arranged to the tertiary hydrogen atom clohexanone derivative 9c (minor pathway) upon
on the stereogenic center bearing the side chain treatment of the epoxide 5c with NaY (Scheme 2).
À
( CH2CH2COCH3).
Thus, 47 was reduced to the alcohol 50 as an equimo-
lar mixture of two diastereomers (with respect to the
epoxide and hydroxy functionalities). Subsequently,
50 was acetylated to form 51. The epoxy acetate 51
was isolated as a mixture of two equimolar diastereo-
mers with respect to the epoxide and hydroxy func-
tionalities, and had 93% stereochemical purity in
terms of the gem-dimethyl moiety. Upon zeolite treat-
ment, we isolated the minor trimethylcyclohexanone
derivative 52 shown in Scheme 10.
1
From the H NMR it was obvious that the a-methyl
This stereochemical outcome is more likely in group with respect to the carbonyl functionality was
agreement with a concerted cyclization mechanism the deuterated one, while the methyl group on the
through the transition state shown in Scheme 9. The tertiary carbon atom was the protonated one (see
possibility that the open carbocation CI (Scheme 9) is Supporting Information). This result clearly estab-
formed upon protonation of the epoxide (PE, lishes the so far proposed,[27,28,54,55] but never proven,
Scheme 9), which cyclizes in an accompanying step, stereochemical rearrangements during the isomeriza-
cannot be excluded. Yet, it requires that the activa- tion of an epoxide to 2,3,4-trimethylcyclohexanone
tion energy of the second cyclization step, be at least under acid catalysis. Among the two stereoselectively
3 kcalmolÀ1 lower compared to the barrier to rotation disposed gem-methyl groups of the intermediate cy-
À
around the C(OH) Ccarbocation bond.
clized carbocation (see also Scheme 9), the protonat-
ed methyl group which is axially oriented in a
pseudo-chair conformation shifts, giving rise to the
stereoselective formation of 52 (Scheme 11). Similar
stereochemical results (selective shift of the CH3
group relative to the CD3) were obtained upon Lewis
Adv. Synth. Catal. 2008, 350, 1587– 1600
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1595

Constantinos Tsangarakis et al.
FULL PAPERS
presented herein is extremely mild, environmentally
friendly, does not require an extractive aqueous work-
up, and gives yields that are quite acceptable. Finally,
through deuterium-labelling experiments we propose
a concerted mechanism for the zeolite-promoted cy-
clization reaction, while for the first time, we unravel-
led the mechanistic aspects of the proposed, but
never proven, mechanism for the transformation of
epoxy polyene terpenes to 2,3,4-trimethylcyclohexa-
nones under acid catalysis. We believe that solid acids
will provide in the future a more practical and envi-
ronmentally benign methodology for several organic
transformations. The environmentally benign nature
of the zeolites and other porous solid acids[56] justify
further exploration towards new frontiers in organic
chemistry.
Scheme 10. Synthesis of the labeled epoxide 51, and the ste-
reochemistry of the by-product 52 isolated after treatment
with NaY.
acid treatment of 51 (SnCl4, CH2Cl2, 08C), with for- Experimental Section
mation of 52 in ca. 10% relative yield (dr~3/1). This
Nuclear magnetic resonance spectra were obtained on a
result indicates an identical mechanism for the forma-
tion of trimethylcyclohexanones either under homo-
geneous or heterogeneous conditions (zeolite confine-
ment).
500 MHz instrument. Isomeric purities were determined by
1H NMR and by GC analysis on a 60 m HP-5 capillary
column. All spectra reported herein were taken in CDCl3.
The synthesis of compounds 3, 4, 5a, b, 6a, b, 7a, b, 10a, b,
11a, b, 12a, b, 17, 18, 19, 20, 22, 23, 27, 28, 29, 30 and 31
have been described in the Supporting Information of ref.[22]
The general method for the synthesis of epoxy terpenoids,
as well as, the spectroscopic data for substrates 5c, 14, 32
and 37, are presented in the Supporting Information Sec-
tion.
Conclusions
In conclusion, we presented herein a novel methodol-
ogy for the direct synthesis of incompletely cyclized
terpenols upon treatment of epoxy polyene terpe-
noids with the slightly acidic zeolite NaY. All epox-
ides studied yielded primarily monocyclic products
possessing selectively an exomethylenic double bond.
This methodology is unique, as Lewis acids provide
products of polycyclization, and allows the fast and ef-
ficient synthesis of the naturally occurring elegansi-
diol, farnesiferols B–D, g-cystophorol, achilleol A and
camelliol C. Additionally, the absolute configuration
of (À)-achilleol A was established. The methodology
Intrazeolite Cyclization of the Epoxy Terpenes
To a slurry containing 1 g of NaY (dried at 1208C at least
for 6 h under vacuum of 10À4 torr) in 5 mL hexane, were
added approximately 50 mg of the epoxy polyene and the
heterogeneous mixture was stirred at room temperature for
10–15 min for the geranyl, farnesyl and squalene derivatives,
or for 30–45 min for the neryl derivatives (10a, b). After
that period, the reaction mixture was filtered, and the fil-
trate was retained. The solid material was further washed
Scheme 11. Specific shift of the CH3 relative to the CD3 group during the formation of 52.
1596
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1587– 1600

Zeolite NaY-Promoted Monocyclization of Epoxy Polyene Terpenes
FULL PAPERS
with 2”10 mL of methanol for 30 min each time, and then
filtered again. The combined solvent extracts were evaporat-
ed to afford the cyclized products in 70–90% yield. For
larger scale experiments, 1 mmol of geranyl derivatives was
used per 1 g of dry NaY, and the reaction was complete
after 1 hour by heating to 708C.
13.0 Hz, J2 =5.0 Hz, 1H), 1.86 (m, 1H), 1.77 (m, 1H), 1.09
(s, 3H), 1.01 (s, 3H).
From Epoxide 32
(E)-8-[(1,3-cis)-3-Hydroxy-2,2-dimethyl-6-methylenecyclo-
hexyl]-6-methyloct-5-en-2-one (33):[39a] 1H NMR: d=5.04 (t,
J=6.5 Hz, 1H), 4.86 (s, 1H), 4.57(s, 1H), 3.40 (dd, J1 =
9.5 Hz, J2 =4.0 Hz, 1H), 2.44 (t, J=7.0 Hz, 2H), 1.80–2.35
(m, 7H), 2.13 (s, 3H), 1.60 (s, 3H), 1.45–1.70 (m, 4H), 1.02
(s, 3H), 0.71 (s, 3H); 13C NMR: 209.0, 147.2, 136.8, 122.5,
108.4, 77.2, 50.9, 43.8, 40.5, 38.5, 32.9, 32.2, 29.8, 25.9, 23.7,
22.4, 16.0, 15.2.
Spectroscopic Data of the Intrazeolite Products
From Epoxide 5c
4-(3-Hydroxy-2,2-dimethyl-6-methylenecyclohexyl)butan-2-
yl acetate (6c-exo, mixture of
2 diastereomers ~1/1):
(E)-8-[(1,5-cis)-5-Hydroxy-2,6,6-trimethylcyclohex-2-enyl]-
6-methyloct-5-en-2-one (34):[39a] 1H NMR (characteristic ab-
sorptions): d=5.22 (br. s, 1H), 5.07(t, J=7.0 Hz, 1H), 3.44
(dd, J1 =9.5 Hz, J2 =6.0 Hz, 1H), 0.95 (s, 3H), 0.82 (s, 3H).
(E)-6-Methyl-8-{(1R*,2S*,4S*)-1,3,3-trimethyl-7-oxa-
1H NMR: d=4.88 (m, 1H), 4.87(s, 1H), 4.59 (s, 1H), 3.41
(dd, J1 =4.0 Hz, J2 =10.0 Hz, 1H), 2.33 (m, 1H), 2.03 (s,
3H), 1.99 (m, 1H), 1.84 (m, 1H), 1.72–1.47 (m, 6H), 1.21 (d,
J=4.0 Hz, 3H), 1.03 (s, 3H), 1.02 (s, 3H of the second dia-
stereomer), 0.73 (s, 3H), 0.72 (s, 3H of the second diastereo-
mer).
bicyclo
[2.2.1]heptan-2-yl}oct-5-en-2-one (35):[39a] 1H NMR:
G
d=5.06 (t, J=6.5 Hz, 1H), 3.71 (d, J=5.5 Hz, 1H), 2.46 (t,
J=7.0 Hz, 2H), 2.26 (m, 2H), 2.13 (s, 3H), 1.90 (m, 3H),
1.65 (m, 1H), 1.61 (s, 3H), 1.33–1.49 (m, 2H), 1.29 (s, 3H),
1.14 (m, 1H), 1.04 (s, 3H), 1.00 (s, 3H); 13C NMR: d=208.8,
136.7, 122.6, 86.7, 86.1, 55.2, 45.2, 43.8, 39.8, 39.0, 29.9, 26.1,
26.1, 25.8, 23.4, 22.4, 18.9, 15.9.
4-(5-Hydroxy-2,6,6-trimethylcyclohex-2-enyl)butan-2-yl
acetate (6c-endo, mixture of
2 diastereomers ~1/1):
1H NMR (characteristic absorptions): d=5.25 (br. s, 1H),
4.90 (m, 1H), 3.46 (dd, J1 =6.0 Hz, J2 =8.0 Hz, 1H), 2.03 (s,
3H), 1.22 (d, J=6.5 Hz, 3H), 0.97(s, 3H), 0.98 (s, 3H of the
second diastereomer), 0.83 (s, 3H), 0.82 (s, 3H of the second
diastereomer).
A
trien-2-one (36):[39a] 1H NMR: d=5.13 (t, J=6.5 Hz, 1H),
5.07(t, J=6.5 Hz, 1H), 4.94 (t, J=1.0 Hz, 1H), 4.84 (t, J=
1.0 Hz, 1H), 4.04 (m, 1H), 2.46 (t, J=6.5 Hz, 1H), 2.27(m,
2H), 2.12 (s, 3H), 1.94–2.09 (m, 6H), 1.73 (s, 3H), 1.60–1.67
(m, 2H), 1.61 (br. s, 6H); 13C NMR: d=208.9, 147.6, 136.2,
134.7, 124.6, 122.7, 110.9, 75.5, 43.7, 39.6, 35.7, 33.3, 29.9,
26.4, 22.5, 17.7, 16.0.
4-(1,3,3-Trimethyl-7-oxa-bicyclo[2.2.1]heptan-2-yl)butan-2-
A
yl acetate (7c-endo, mixture of 2 diastereomers ~1/1):
1H NMR: d=4.86 (m, 1H), 3.72 (d, J=5.5 Hz, 1H), 2.03 (s,
3H), 1.71–1.43 (m, 7H), 1.37 (m, 1H), 1.31 (s, 3H), 1.30 (s,
3H of the second diastereomer), 1.22 (d, J=6.5 Hz, 3H),
1.17(m, 1H), 1.04 (s, 3H), 1.03 (s, 3H of the second diaste-
reomer), 1.00 (s, 3H), 0.99 (s, 3H (s, 3H of the second dia-
stereomer).
4-(1,2,6-Trimethyl-3-oxocyclohexyl)butan-2-yl acetate (9c,
1
mixture of 2 diastereomers ~4/1): H NMR (major diaste-
From 2,3-Epoxysqualene (37)
reomer): d=4.87(m, 1H), 2.43 (m, 1H), 2.34 (m, 2H), 2.04
(s, 3H), 1.95 (m, 1H), 1.85 (m, 1H), 1.63 (m, 1H), 1.51–1.31
(m, 4H), 1.25 (d, J=6.0 Hz, 3H), 0.93 (d, J=7.0 Hz, 3H),
0.88 (d, J=7.0 Hz, 3H), 0.59 (s, 3H); 13C NMR (major dia-
stereomer): d=213.7, 170.7, 71.1, 50.4, 43.2, 41.6, 36.2, 32.3,
A
3,8,12,16-tetramethylheptadeca-3,7,11,15-tetraenyl]cyclohex-
anol (38,[37] achilleol A): 1H NMR: d=5.10 (m, 4H), 4.86
(br. s, 1H), 4.59 (br. s, 1H), 3.41 (dd, J1 =9.5 Hz, J2 =4.0 Hz,
1H), 2.32 (td, J1 =13.0 Hz, J2 =5.0 Hz, 1H), 1.95–2.10 (m,
16H), 1.55–1.75 (m, 4H), 1.67 (s, 3H), 1.58 (s, 12H), 1.01 (s,
3H), 0.70 (s, 3H); 13C NMR: d=147.4, 135.6, 135.3, 135.1,
131.4, 124.6, 124.5, 124.5, 124.4, 108.5, 77.4, 51.0, 40.7, 39.9,
39.9, 38.8, 33.3, 32.4, 28.5, 28.4, 26.9, 26.8, 26.0, 25.8, 23.9,
17.8, 16.3, 16.2, 16.2, 15.6.
1
30.9, 28.8, 21.4, 20.0, 15.0, 14.3, 7.4; H NMR (minor diaste-
reomer, characteristic absorptions): d=0.91 (d, J=7.0 Hz,
3H), 0.90 (d, J=7.0 Hz, 3H); 13C NMR (minor diastereo-
mer, characteristic absorptions): d=71.2, 36.1, 19.9, 15.4,
14.9, 7.5.
AHCTREUNG
From Epoxide 14 (Myrcene Oxide)
1
camelliol C): H NMR (selected absorptions): d=5.22 (br. s,
1H), 5.10–5.15 (m, 4H), 3.45 (t, J=7.0 Hz, 1H), 0.96 (s,
3H), 0.82 (s, 3H).
6,6-Dimethyl-4-vinylcyclohex-3-enol (14a):[26] 1H NMR: d=
6.38 (dd, J1 =16.0 Hz, J2 =10.0 Hz, 1H), 5.59 (br. s, 1H),
5.09 (d, J=16.0 Hz, 1H), 4.93 (d, J=10.0 Hz, 1H), 3.56 (m,
1H), 2.44 (br. d, J=11.0 Hz, 1H), 2.18 (br. d, J=11.0 Hz,
1H), 2.15 (d, J=13.0 Hz, 1H), 1.97(d, J=13.0 Hz, 1H),
0.99 (s, 3H), 0.95 (s, 3H).
(1R*,2S*,4S*)-1,3,3-Trimethyl-2-[(3E,7E,11E)-3,8,12,16-
tetramethylheptadeca-3,7,11,15-tetraenyl]-7-oxa-bicyclo-
[2.2.1]heptane (40): H NMR: d=5.11 (m, 4H), 3.71 (d, J=
5.0 Hz, 1H), 1.85–2.12 (m, 14H), 1.67(s, 3H), 1.59 (br. s,
12H), 1.35–1.50 (m, 6H), 1.32 (s, 3H), 1.17(m, 1H), 1.04 (s,
3H), 1.01 (s, 3H); 13C NMR: d=135.4, 135.2, 134.9, 131.2,
124.5, 124.3, 86.7, 86.1, 55.3, 45.3, 41.3, 39.8, 39.7, 39.0, 28.2,
26.8, 26.7, 26.7, 26.2, 26.1, 25.8, 25.7, 23.4, 18.8, 17.7, 16.0.
1
A
2,2-Dimethyl-4-vinylcyclohex-3-enol (14b):[26] 1H NMR:
d=6.35 (dd, J1 =16.0 Hz, J2 =10.0 Hz, 1H), 5.41 (s, 1H),
5.09 (d, J=16.0 Hz, 1H), 4.91 (d, J=10.0 Hz, 1H), 3.59 (m,
1H), 2.29 (td, J1 =13.0 Hz, J2 =5.0 Hz, 1H), 2.19 (td, J1 =
Adv. Synth. Catal. 2008, 350, 1587– 1600
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1597

Constantinos Tsangarakis et al.
FULL PAPERS
(6E,10E,14E,18E)-2,6,10,15,19,23-Hexamethyltetracosa-
1,6,10,14,18,22-hexaen-3-ol (41):[43] 1H NMR: d=5.08 (m,
6H), 4.93 (br. s, 1H), 4.82 (br. s, 1H), 4.02 (br. t, J=6.5 Hz),
1.92–2.12 (m, 18H), 1.70 (br. s, 6H), 1.66 (br. s, 3H), 1.61
(m, 2H), 1.59 (br. s, 15H).
4 days, the compete transformation of diol to the (S)-2,3-ep-
oxysqualene was realized. Purification was accomplished by
column chromatography (hexane/ethyl acetate=10/1) to iso-
late 72 mg of epoxide (62% yield).
Synthesis of the Deuterium-Labelled Epoxy
Geranylacetone 47
3-(3-Hydroxybutyl)-2,3,4-trimethylcyclohexanone (15,
mixture of 2 diastereomers ~3/1)
To a solution of geraniol-8,8,8-d3, 43 (0.4 g, 2.55 mmol) in
5 mL dry diethyl ether were added at
0
oC, 0.1 mL
The keto acetate 9c (11 mg) was hydrolyzed with K2CO3
(2 equiv.) in methanol to form 15 (89% yield) as a mixture
(1.0 mmol) of PBr3. After 30 min a saturated solution of
NaHCO3 was added and the resulting solution was extracted
with Et2O. The ether layer was dried and the solvent was re-
moved under vacuum to afford geranyl bromide-8,8,8-d3
(44) in 96% yield. 1H NMR of 44: d=5.53 (t, J=7.5 Hz,
1H), 5.07(t, J=6.5 Hz, 1H), 4.03 (d, J=7.5 Hz, 2H), 2.08
(m, 4H), 1.73 (s, 3H), 1.60 (s, 3H). Subsequently, the la-
belled bromide 44 (0.51 g, 2.3 mmol), alkylated the anion of
ethyl acetoacetate formed upon treatment of ethyl acetoace-
tate (0.35 mL, 2.78 mmol) with 0.38 g K2CO3 (2.78 mmol) in
10 mL dry acetone. The reaction was complete after 12 h at
1
of two diastereomers in a ratio of ~3/1. H NMR (major dia-
stereomer): d=3.77 (m, 1H), 2.45 (m, 1H), 2.35 (m, 2H),
1.98 (m, 1H), 1.85 (m, 1H), 1.62 (m, 1H), 1.45 (m, 2H),
1.37–1.20 (m, 3H), 1.24 (d, J=7.0 Hz, 3H), 0.93 (d, J=
7.0 Hz, 3H), 0.89 (d, J=6.5 Hz, 3H), 0.60 (s, 3H); 13C NMR
(major diastereomer): d=213.9, 68.5, 50.5, 43.3, 41.6, 36.2,
1
32.9, 32.0, 31.0, 23.7, 15.4, 15.0, 7.5; H NMR (minor diaste-
reomer, characteristic absorptions): d=0.92 (d, J=7.0 Hz,
3H), 0.91 (d, J=7.0 Hz, 3H); 13C NMR (minor diastereo-
mer, characteristic absorptions): d=68.6, 33.0, 15.4, 15.0,
7.6.
o
65 C. The ethyl a-geranyl-d3 acetoacetate 45 was isolated in
92% yield. 1H NMR of 45: d=5.04 (m, 2H), 4.19 (q, J=
7.0 Hz, 2H), 3.43 (t, J=6.5 Hz, 1H), 2.55 (t, J=7.0 Hz, 2H),
2.22 (s, 3H), 2.04 (m, 2H), 1.97(t, J=7.0 Hz, 2H), 1.63 (s,
3H), 1.59 (s, 3H), 1.27(t, J=7.0 Hz, 3H); 13C NMR of 45:
d=203.1, 169.6, 138.3, 131.5, 124.0, 119.7, 61.3, 59.8, 39.6,
29.1, 26.9, 25.6, 17.7, 16.1, 14.1. By heating an alkaline
(1.6M NaOH) ethanolic solution containing 0.72 g of the
keto ester 45 for 3 h at 708C, the saponification/decarboxy-
lation was complete. The reaction mixture was extracted
with Et2O and the organic residue was chromatographed
using hexane/ethyl acxetate=10/1 as eluant to afford gera-
nylacetone-d3 (46) in 88% yield. From the examination of
the NMR spectrum we concluded that 46 was 93% geomet-
rically pure in terms of the labelling topicity on the terminal
double bond. 1H NMR of 46: d=5.07(t, J=7.0 Hz, 2H),
2.45 (t, J=7.0 Hz, 2H), 2.27 (m, 2H), 2.13 (s, 3H), 2.06 (m,
2H), 1.97(t, J=7.0 Hz, 2H), 1.61 (s, 3H), 1.59 (s, 3H);
13C NMR of 46: d=208.9, 136.4, 131.3, 124.2, 122.5, 43.8,
39.7, 30.0, 29.6, 22.5, 17.6, 16.0. Finally, geranylacetone-d3
(46) was epoxidized in 61% isolated yield according to the
general procedure for the synthesis of epoxy terpenoids
(NBS/H2O, K2CO3) presented earlier, to form the labelled
(2R*,3S*,4R*)-2,3,4-Trimethyl-3-(3-oxobutyl)cyclo-
hexanone (16, single diastereomer)
The keto alcohol 15 (9 mg) was oxidized with 1.5 equiv. of
PCC in CH2Cl2 at room temperature to afford after 3 h the
known diketone 16[27] as a single diastereomer. Purification
was performed by passing the concentrated reaction residue
through a short pad of celite using diethyl ether as eluant
1
(92% yield). H NMR: d=2.46 (m, 1H), 2.35 (m, 4H), 2.19
(s, 3H), 1.88 (m, 2H), 1.61 (m, 3H), 0.92 (d, J=7.0 Hz, 3H),
0.89 (d, J=6.5 Hz, 3H), 0.62 (s, 3H); 13C NMR: d=213.0,
208.4, 50.5, 43.0, 41.5, 37.1, 36.4, 30.8, 30.6, 30.1, 15.2, 15.1,
7.6.
(S)-2,3-Epoxysqualene [(S)-37]
In one-necked flask were placed 1.0 g (2.43 mmol) of squa-
lene, 0.1 g (5%) of (DHQD)2-PHAL, 2.4 g (7.3 mmol) of
K3Fe(CN)6, 1.0 g (7.3 mmol) of K2CO3, 0.23 g (2.43 mmol)
of CH3SO2NH2, 0.25 mL of OsO4 solution (2.5 wt% in t-
BuOH), and 20 mL of t-BuOH. The mixture was stirred at
08C for 48 h. HPLC analysis revealed 60% conversion of
squalene. Then it was diluted with 10 mL of CH2Cl2 and
washed with a solution of 2 N NaOH. The organic residue
was carefully chromatographed with hexane/ethyl acetate=
10/1 to separate the more polar 2,3-diol from the less polar
mixture of the internal 6,7- and 10,11-diols.[57] The desired
(R)-2,3-dihydroxysqualene (42)[49] was isolated in 12% yield
and in 94% er. The enantiomeric purity was determined by
converting 42 to the mono-MTPA ester and subsequent
1H NMR analysis.[58] 1H NMR of 42: d=5.19 (t, J=6.5 Hz,
1H), 5.08–5.15 (m, 4H), 3.35 (dd, J1 =10.5 Hz, J2 =2.0 Hz,
1H), 2.24 (m, 1H), 1.96–2.11 (m, 18H), 1.68 (s, 3H), 1.62 (s,
3H), 1.60 (br. s, 12H), 1.20 (s, 3H), 1.15 (s, 3H); 13C NMR
of 42: d=135.2, 135.0, 134.9, 134.8, 131.3, 125.2, 124.5, 124.4,
124.3, 78.3, 73.0, 39.8, 39.7, 39.7, 36.8, 29.7, 28.3, 26.8, 26.7,
26.6, 26.4, 25.7, 23.3, 17.7, 16.1, 16.0, 15.9. The (R)-42 diol
(120 mg, 0.27mmol) was treated with TsCl (73 mg,
0.4 mmol), Et3N (80 mL, 0.55 mmol) and DMAP (67mg,
0.55 mmol) in dichloromethane at room temperature. After
1
epoxide 47. H NMR of 47: d=5.12 (t, J=7.0 Hz, 1H), 2.68
(t, J=7.0 Hz, 1H), 2.46 (t, J=7.0 Hz, 2H), 2.27 (m, 2H),
2.16 (m, 1H), 2.13 (s, 3H), 2.07(m, 1H), 1.63 (s, 3H), 1.61
(m, 2H), 1.25 (s, 3H); 13C NMR of 47: d=208.6, 135.4,
123.2, 64.0, 58.3, 43.5, 36.2, 29.9, 27.3, 22.3, 18.7, 15.9.
Synthesis of the Deuterium-Labelled Epoxide 51
To a solution of epoxide 47 (0.26 g, 1.22 mmol) in dry THF
were slowly added at 08C 0.32 mL (0.32 mmol) LiAlH4 (1M
in THF). After 30 min the reaction mixture was quenched
with H2O. Following an extractive work-up, the labeled
epoxy alcohol 50 was isolated in 93% yield, as an equimolar
mixture of two diastereomers, with respect to the epoxide
1
and hydroxy functionalities. H NMR of 50: d=5.17(t, J=
7.0 Hz, 1H), 3.76 (m, 1H), 2.67 (t, J=6.5 Hz, 1H), 2.08 (m,
4H), 1.68 (br. s, 1H), 1.63–1.57(m, 2H), 1.61 (s, 3H), 1.48
(m, 2H), 1.23 (s, 3H), 1.17(d, J=6.5 Hz, 3H); 13C NMR of
50: d=134.7, 124.6, 67.8, 64.1, 58.2, 39.1, 36.4, 27.4, 24.4,
1598
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1587– 1600

Zeolite NaY-Promoted Monocyclization of Epoxy Polyene Terpenes
FULL PAPERS
23.6, 18.7, 16.0. Subsequently, 50 was acetylated in 95%
yield by reaction with 0.17mL acetic anhydride
Ça, B. Bazdi, R. Robles, J. M. Cuerva, J. E. Oltra, Adv.
Synth. Catal. 2008, 350, 571.
(1.81 mmol), in the presence of 0.25 g K2CO3 (1.81 mmol)
and DMAP (0.02 g) and using ethyl acetate, as solvent.
1H NMR of 51: d=5.15 (t, J=7.0 Hz, 1H), 4.88 (m, 1H),
2.69 (t, J=6.0 Hz, 1H), 2.14 (m, 1H), 2.09 (m, 1H), 2.03 (s,
3H), 2.02 (m, 2H), 1.65 (m, 3H), 1.50 (m, 1H), 1.26 (s, 3H),
1.21 (d, J=6.5 Hz, 3H); 13C NMR of 51: d=170.7, 134.8,
123.9, 70.6, 64.0, 58.2, 36.3, 35.9, 27.4, 23.9, 21.4, 19.9, 18.7,
15.9.
[15] For the influence of restricted spaces on product selec-
tivity see: a) N. J. Turro, X. Lei, C.-C. Cheng, J. Am.
Chem. Soc. 1985, 107, 5824; b) J. Sivaguru, A. Natara-
jan, L. S. Kaanumalle, J. Shailaja, S. Uppili, A. Joy, V.
Ramamurthy, Acc. Chem. Res. 2003, 36, 509; c) T. S.
Koblenz, J. Wassenaar, J. N. H. Reek, Chem. Soc. Rev.
2008, 37, 247.
[16] a) S. E. Sen, S. M. Smith, K. A. Sullivan, Tetrahedron
1999, 55, 12657; b) N. J. Turro, Acc. Chem. Res. 2000,
33, 637; c) V. Ramamurthy, P. Lakshminarasimhan,
C. P. Grey, L. J. Johnston, Chem. Commun. 1998, 2411.
[17] a) A. Corma, H. Garcia, Chem. Commun. 2004, 1443;
b) M. Alvaro, J. F. Cabeza, D. Fabuel, A. Corma, H.
Garcia, Chem. Eur. J. 2007, 13, 3733; c) J. Shailaja, J.
Sivaguru, S. Uppili, A. Joy, V. Ramamurthy, Micropo-
rous Mesoporous Mater. 2001, 48, 319; d) M. Stratakis,
Top. Curr. Synth. 2005, 2, 281; e) M. Alvaro, E. Carbon-
ell, V. Fornꢂs, H. Garcꢃa, ChemPhysChem 2006, 7, 200.
[18] a) G. Busca, Chem. Rev. 2007, 107, 5366; b) V. J. Rao,
D. L. Perlstein, R. J. Robbins, P. H. Lakshminarasim-
han, H.-M. Kao, C. P. Grey, V. Ramamurthy, V. Chem.
Commun. 1998, 269.
[19] a) C. Tsangarakis, M. Stratakis, Adv. Synth. Catal. 2005,
347, 1280; b) C. Tsangarakis, M. Stratakis, Eur. J. Org.
Chem. 2006, 4435; c) C. Tsangarakis, I. N. Lykakis, M.
Stratakis, J. Org. Chem. 2008, 73, 2905.
[20] a) S. E. Sen, Y. Z. Zhang, S. L. Roach, J. Org. Chem.
1996, 61, 9534; b) S. E. Sen, Y. Z. Zhang, S. M. Smith,
J. Org. Chem. 1998, 63, 4459.
[21] V. J. Rao, S. R. Uppili, D. R. Corbin, S. Schwarz, S. R.
Lustig, V. Ramamurthy, J. Am. Chem. Soc. 1998, 120,
2480.
Acknowledgements
This work was partially supported financially by the Greek
Ministry of Education (Bꢀ EPEAEK, Pythagoras II). EA
thanks IKY for a 3 years fellowship.
References
[1] a) W. S. Johnson, A fifty-year love affair with organic
chemistry; American Chemical Society: Washington
DC, 1998; b) R. A. Yoder, J. N. Johnston, Chem. Rev.
2005, 105, 4730; c) K. U. Wendt, G. E. Schulz, E. J.
Corey, D. R. Liu, Angew. Chem. 2000, 112, 2930;
Angew. Chem. Int. Ed. 2000, 39, 2812.
[2] a) E. E. van Tamelen, Acc. Chem. Res. 1975, 8, 152;
b) A. Eschenmoser, D. Arigoni, Helv. Chim. Acta 2005,
88, 3011; c) I. Abe, Nat. Prod. Rep. 2007, 24, 1311.
[3] E. J. Corey, D. D. Staas, J. Am. Chem. Soc. 1998, 120,
3526.
[4] E. J. Corey, H. Cheng, C. H. Baker, S. P. T. Matsuda, X.
Li, X. Song, J. Am. Chem. Soc. 1997, 119, 1289.
[5] P. F. Vlad, Pure Appl. Chem. 1993, 65, 1329.
[6] E. E. van Tamelen, A. Storni, E. J. Hessler, M.
Schwartz, J. Am. Chem. Soc. 1963, 85, 3295.
[22] For a preliminary study see: C. Tsangarakis, C.; E. Ar-
koudis, C. Raptis, M. Stratakis, Org. Lett. 2007, 9, 583.
[23] A. F. Barrero, E. J. Alvarez-Manzaneda, P. L. Palomino,
Tetrahedron 1994, 50, 13239.
[24] a) Y. Mi, J. V. Schreiber, E. J. Corey, J. Am. Chem. Soc.
2002, 124, 11290; b) L. Petterson, T. Frejd, J. Chem.
Soc. Perkin Trans. 1 2001, 789.
[25] T.-L. Ho, S.-H. Liu, Liebigs Ann. Chem. 1983, 761.
[26] Y. Shinya, K. Takeo, Nihon Yukag. 1996, 45, 1271.
[27] E. E. van Tamelen, R. G. Nadeau, Bioorg. Chem. 1982,
11, 197.
[28] T. Hoshino, Y. Sakai, Tetrahedron Lett. 2001, 42, 7319.
[29] L. W. Bieber, I. Messana, S. C. N. Lins, A. A. da Silva
Filho, A. A. Chiappeta, J. F. De Mello, Phytochemistry
1990, 29, 1955.
[7] G. Vidari, S. Beszant, J. E. Merabet, M. Bovolenta, G.
Zanoni, Tetrahedron Lett. 2002, 43, 2687.
[8] S. E. Sen, S. L. Roach, S. M. Smith, Y. Z. Zhang, Tetra-
hedron Lett. 1998, 39, 3969.
[9] M. Bogenstatter, A. Limberg, L. E. Overman, A. L.
Tomasi, J. Am. Chem. Soc. 1999, 121, 12206.
[10] J. R. Lacey, P. W. Anzalone, C. M. Duncan, M. J. Hack-
ert, R. S. Mohan, Tetrahedron Lett. 2005, 46, 8507.
[11] B. M. Smith, E. J. Skellam, S. J. Oxley, A. E. Graham,
Org. Biomol. Chem. 2007, 5, 1979.
[12] a) E. J. Corey, M. Sodeoka, Tetrahedron Lett. 1991, 32,
7005; b) A. X. Huang, Z. M. Xiong, E. J. Corey, J. Am.
Chem. Soc. 1999, 121, 9999.
[30] M. Toyota, F. Nagashima, Y. Asakawa, Phytochemistry
1989, 28, 1661.
[31] K. Mori, S. Takechi, Tetrahedron Lett. 1985, 26, 3049.
[32] The BF3-catalyzed cyclization of 17 gives a mixture of
products, with a low yield of products of bi-cyclization.
A minor fraction of three regio-isomeric products of
monocyclization is also formed: E. E. van Tamelen, A.
Storni, E. J. Hessler, M. A. Schwartz, Bioorg. Chem.
1982, 11, 133.
[33] A. F. Barrero, E. J. Alvarez-Manzaneda, M. M. Herra-
dor, R. Alvarez-Manzaneda, J. Quilez, R. Chahboun, P.
Linares, A. Rivas, Tetrahedron Lett. 1999, 40, 8273.
[13] a) J. Justicia, A. Rosales, E. Bunuel, J. L. Oller-Lopez,
N. Valdivia, A. Haidour, J. E. Oltra, A. F. Barrero, D. J.
Cardenas, J. M. Cuerva, Chem. Eur. J. 2004, 10, 17 7 8;
b) J. Justicia, J. L. Oller-Lopez, A. G. Campana, J. E.
Oltra, J. M. Cuerva, E. Bunuel, D. J. Cardenas, J. Am.
Chem. Soc. 2005, 127, 14911.
[14] a) A. F. Barrero, J. F. Q. del Moral, E. M. Sanchez, J. F.
Arteaga, Eur. J. Org. Chem. 2006, 10, 1627; b) J. M.
Cuerva, J. Justicia, J. L. Oller-Lopez, J. E. Oltra, Top.
Curr. Chem. 2006, 264, 63; c) J. Justicia, A. G. Campa-
Adv. Synth. Catal. 2008, 350, 1587– 1600
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1599

Constantinos Tsangarakis et al.
FULL PAPERS
[34] a) L. Cao, J. Sun, X. Wang, R. Zhu, H. Shi, F. Hu, Tet-
rahedron 2007, 63, 5036; b) G. Audran, J. M. Galano,
H. Monti, Eur. J. Org. Chem. 2001, 12, 2293; c) A. F.
Barrero, E. J. Alvarez-Manzaneda, R. Chahboun, A. R.
Rivas, P. L. Palomino, Tetrahedron 2000, 56, 6099.
[35] a) L. Caglioti, H. Naef, D. Arigoni, O. Jeger, Helv.
Chim. Acta 1959, 42, 2557; b) M. I. Nassar, Pharmazie
1994, 49, 542; c) M. Iranshahi, G. Amin, A. Shafiee,
Pharm. Biol. 2004, 42, 440.
[44] S. Lodeiro, Q. Xiong, W. K. Wilson, M. D. Kolesnikova,
C. S. Onak, S. P. T. Matsuda, J. Am. Chem. Soc. 2007,
129, 11213.
[45] A. F. Barrero, J. M. Cuerva, E. J. Alvarez-Manzaneda,
J. E. Oltra, R. Chahboun, Tetrahedron Lett. 2002, 43,
2793.
[46] E. E. van Tamelen, J. Willet, M. Schwartz, R. Nadeau,
J. Am. Chem. Soc. 1966, 88, 5937.
[47] a) Q. B Xiong, W. K. Wilson, S. P. T. Matsuda, Angew.
Chem. 2006, 118, 1307; Angew. Chem. Int. Ed. 2006, 45,
1285; b) T. Hoshino, K. Shimizu, T. Sato, Angew.
Chem. 2004, 116, 6868; Angew. Chem. Int. Ed. 2004, 43,
6700; c) T. K. Wu, J. H. Griffin, Biochemistry 2002, 41,
8238; d) S. P. T. Matsuda, L. B. Darr, E. A. Hart,
J. B. R. Herrera, K. E. McCann, M. M. Meyer, J. H.
Pang, H. G. Schepmann, Org. Lett. 2000, 2, 2261;
e) B. M. Joubert, L. Hua, S. P. T. Matsuda, Org. Lett.
2000, 2, 339.
[36] M. Bittner, J. Jakupovic, F. Bohlmann, M. Silva, Phyto-
chemistry 1988, 27, 3845.
[37] A. F. Barrero, E. J. Alvarez-Manzaneda, R. Alvarez-
Manzaneda, Tetrahedron Lett. 1989, 30, 3351.
[38] a) T. Akihisa, K. Arai, Y. Kimura, K. Koike,
W. C. M. C. Kokke, T. Shibata, T. Nikaido, J. Nat. Prod.
1999, 62, 265; b) M. D. Kolensikova, W. K. Wilson,
D. A. Lynch, A. C. Obermeyer, S. P. T. Matsuda, Org.
Lett. 2007, 9, 5223.
[39] a) B. N. Ravi, P. T. Murphy, R. O. Lidgard, R. G.
Warren, R. J. Wells, Aust. J. Chem. 1982, 35, 17 1; b) V.
Amico, Phytochemistry 1995, 39, 1257.
[48] L. D’Accolti, C. Annese, C. Fusco, Tetrahedron Lett.
2005, 46, 8459.
[49] G. A. Crispino, K. B. Sharpless, Tetrahedron Lett. 1992,
33, 4273.
[40] 13,14-Epoxy-trans,trans-farnesylacetone (32) was pre-
pared in 64% yield via epoxidation of trans,trans-farne-
sylacetone upon reaction with NBS in H2O/THF, fol-
lowed by treatment of the resulting bromohydrin with
K2CO3 in MeOH. trans,trans-Farnesylacetone was syn-
thesized in turn by alkylation of ethyl acetoacetate with
trans,trans-farnesyl bromide followed by hydrolysis/de-
carboxylation of the resulting ethyl a-farnesylated ace-
toacetate.
[41] Our designation for the naturally occurring isomeric
keto alcohols 33 and 34, is g- and a-cystophorol, re-
spectively, which were given no chemical descriptor by
the isolation chemists (ref.[39a]).
[42] Y. Kitahara, T. Kato, M. Kishi, Chem. Pharm. Bull.
1968, 16, 2216.
[43] a) L. De Napoli, E. Fattorusso, S. Magno, L. Mayol,
Phytochemistry 1982, 21, 782; b) C. A. N. Catalan, C. S.
Heluani, C. Kotowicz, T. E. Gedris, W. Herz, Phyto-
chemistry 2003, 64, 625.
[50] After this manuscript had been submitted, Barrero and
co-workers reported the enantioselective synthesis of
(À)-achilleol A, whose absolute configuration is in
aggrement with our assignment: J. F. Arteaga, V. Domi-
ngo, J. F. Quilez del Moral, A. F. Barrero, Org. Lett.
2008, 10, 1723.
[51] T. Hoshino, S. Nakano, T. Kondo, T. Sato, A. Miyoshi,
Org. Biomol. Chem. 2004, 2, 1456.
[52] E. E. van Tamelen, D. R. James, J. Am. Chem. Soc.
1977, 99, 950.
[53] E. J. Corey, D. D. Staas, J. Am. Chem. Soc. 1998, 120,
3526.
[54] E. E. van Tamelen, R. M. Coates, Bioorg. Chem. 1982,
11, 171.
[55] R. Hunter, G. Mellows, Tetrahedron Lett. 1978, 5051.
[56] a) J. H. Clark, Acc. Chem. Res. 2002, 35, 791; b) K.
Kaneda, Synlett 2007, 999.
[57] T. Hoshino, Y. Yonemura, T. Abe, Y. Sugino, Org.
Biomol. Chem. 2007, 5, 792.
[58] E. J. Corey, J. Zhang, Org. Lett. 2001, 3, 3211.
1600
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1587– 1600