Scalable Synthesis of the Amber Odorant 9-epi-Ambrox through a Biomimetic Cationic Cyclization/Nucleophilic Bromination Reaction

Article abstract of DOI:10.1021/acs.orglett.6b02266

A novel biomimetic nucleophilic bromocyclization reaction is used in the key step of a new and straightforward synthesis of 9-epi-Ambrox, an organic compound of high interest and value in the context of fragrances. This strategic reaction allows access to 9-epi-Ambrox on a gram scale from a dienyne derivative, easily available from geraniol, following a sequence of seven steps (35% global yield) with just one purification process. Both enantiomers of the molecule were obtained by a challenging enzymatic resolution.

Full text of DOI:10.1021/acs.orglett.6b02266

Letter
pubs.acs.org/OrgLett
Scalable Synthesis of the Amber Odorant 9-epi-Ambrox through a
Biomimetic Cationic Cyclization/Nucleophilic Bromination Reaction
Raquel Fontaneda, Pedro Alonso, Francisco J. Fananas,* and Felix Rodríguez*
́
́
̃
́ ́
Instituto Universitario de Química Organometalica “Enrique Moles”, Universidad de Oviedo, Julian Clavería 8, E-33006 Oviedo,
Spain
S
* Supporting Information
ABSTRACT: A novel biomimetic nucleophilic bromocycliza-
tion reaction is used in the key step of a new and
straightforward synthesis of 9-epi-Ambrox, an organic com-
pound of high interest and value in the context of fragrances.
This strategic reaction allows access to 9-epi-Ambrox on a
gram scale from a dienyne derivative, easily available from
geraniol, following a sequence of seven steps (35% global yield) with just one purification process. Both enantiomers of the
molecule were obtained by a challenging enzymatic resolution.
erpenes constitute the largest and most varied class of
natural products with tens of thousands of members.1
nucleophile. Alkenyl bromide 3 fulfills these prerequisites as it
is also an appropriate reagent for conventional cross-coupling
reactions but could be easily converted into a nucleophile
through a simple bromine−lithium interchange. Herein, we
report not only a biomimetic synthesis of alkenyl bromide 3 but
also its application in the context of terpene synthesis. In
particular, a scalable synthesis of the powerful odorant 9-epi-
Ambrox is presented.
Within the field of perfumery, an important family of
perfumes are the ambergris fragrances.4 In particular, Ambrox
(4; Scheme 1) is typically used as base note of perfume
compositions, and the list of commercial products containing
this compound is very long.5,6 Interestingly, studies of
structure−odor relationships undertaken on different amber-
type compounds showed that the 9-epi isomer 5 is more
powerful than Ambrox (Scheme 1).7 In fact, (−)-9-epi-Ambrox
(5) possesses the strongest scent and lowest threshold
concentration (0.15 ppb) of all the tested molecules. In spite
of that, and surprisingly, this compound hardly earned any
synthetic interest.8 For all these reasons, 9-epi-Ambrox is an
interesting synthetic target not only from an academic but also
from an industrial point of view.
T
They have found wide application in the pharmaceutical,
agrochemical, food, and cosmetic industries. The way the
enzymatic machinery transforms simple starting materials into
intricate terpenes in Nature has always fascinated synthetic
chemists and has served as an inspiration for the development
of beautiful biomimetic cyclization processes.2 In this context,
we have recently reported a new method for the synthesis of
(poly)cyclic alkenyl triflates from simple acyclic precursors.3
For example, from the dienyne derivative 1, we have
constructed the decaline-derived triflate 2, a versatile reagent
for the synthesis of terpenes (Scheme 1).3 Thus, triflate 2 is an
ideal substrate for subsequent well-documented cross-coupling
reactions where the carbon of the C−OTf bond is an
electrophilic position. However, it would be much more
interesting to have at our disposal a reagent similar to 2 that
could behave not only as an electrophilic partner but also as a
Scheme 1. Our Proposal and Previous Work
Our synthetic plan to 9-epi-Ambrox (5) is shown in Scheme
2. First, we considered that the target molecule 5 could be
available through a simple acid-catalyzed cyclization of the
exocyclic alkenol derivative 6. Probably, the most important
challenge in the synthesis of this molecule is the generation of
the stereogenic center at C9 with the appropriate configuration.
In this context, we thought that compound 6 could be derived
from the enol ether 7 through a Claisen rearrangement. The
presence of the axial-methyl group at the bridge carbon atom
(C8) in 7 should force the Claisen rearrangement to occur
through the opposite face to deliver the desired alkyl chain at
Received: July 31, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b02266
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 2. Retrosynthetic Analysis
Scheme 3. Scalable Synthesis of 9-epi-Ambrox
C9 with appropriate configuration.9 Enol ether 7 could be easily
available through the reaction of the alkenyl lithium derivative 8
with formaldehyde 9 (or a related reagent) followed by a formal
alkoxy-group exchange reaction with enol ether 10. Organo-
lithium derivative 8 would be derived from the alkenyl bromide
3. For the synthesis of this bromine-containing decalin 3, we
devised an unprecedented biomimetic cationic cyclization/
nucleophilic bromination reaction of the dienyne 1, easily
available from geraniol (11). Thus, we considered that an acid
(H+) would promote a stereoselective cationic bicyclization
reaction to deliver an alkenyl cation derivative that should be
trapped by a bromide ion (Br−) present in the reaction media.
Our recent studies on the synthesis of alkenyl triflates support
the viability of the proposed cationic cyclization reaction of
dienyne 1.3 At this point, a work published by Johnson and co-
workers more than 40 years ago should also be mentioned.10 In
this interesting work, an alkenyl cation generated through a
biomimetic cyclization reaction is trapped by a chloride ion
coming from the solvent of the reaction (dichloromethane).
Although just a single example is reported, the synthetic
potential of this reaction seems very high. Surprisingly, this
remarkable transformation seems to have passed somewhat
unnoticed by the synthetic organic community through the
years. However, we believed that this initial finding by Johnson
could be further exploited to get interesting alkenyl bromides
such as 3 with potential application in the synthesis of terpenes
and particularly in the context of the synthesis of 9-epi-Ambrox.
As shown in Scheme 3, our synthesis of 9-epi-Ambrox
commenced with the multigram scale preparation of the
dienyne derivative 1 from inexpensive (∼0.1 $/gram) and easily
available geraniol 11 through a known method.11 With starting
dienyne derivative 1 in hand, we initiate our planned synthetic
sequence. The unprecedented crucial biomimetic cyclization
reaction of dienyne 1 to get the decaline derivative 3 required
some optimization in terms of finding the best protic acid,
temperature, and concentration. Finally, we found that reaction
of a 0.1 M solution of dienyne 1 in dibromomethane (as
solvent and source of bromide) with 1 equiv of tetrafluoroboric
acid (HBF4) at room temperature for 1 h resulted in the
formation of the desired alkenyl bromide 3. In order to
demonstrate the scalability of this biomimetic cyclization, we
performed the reaction with 5.6 g of the starting dienyne 1. In
addition, we found that the crude of this reaction could be used
in the next step without purification.
organolithium compound 8. Although the simple addition of
paraformaldehyde led to the formation of the desired allylic
alcohol 12, better results in terms of reliability were obtained
when this transformation was performed through a two-step
sequence involving the initial treatment of organolithium
compound 8 with N,N-dimethylformamide followed by
reduction of the so-formed aldehyde with sodium borohydride.
Without purification, the crude alcohol 12 was treated with
butyl vinyl ether in the presence of 0.5 mol % of palladium(II)
trifluoroacetate and 0.5 mol % of bathophenanthroline (BPhen)
to obtain the desired enol ether 7.12 Again, this product could
be used in the next step without purification. Although the
planned Claisen rearrangement of enol ether 7 could be
performed under conventional heating, a much more efficient
protocol included the use of microwave irradiation conditions.
Thus, when a 1 M solution of enol ether 7 was heated in 1,4-
dioxane at 195 °C under microwave irradiation, we observed
the clean formation of the desired rearranged aldehyde in just 2
h. This result further demonstrates the convenience of
microwave irradiation to improve and speed up some chemical
processes.13 The direct treatment of the crude of this reaction
with sodium borohydride led to alcohol 6. Without purification,
this alcohol 6 was subjected to reaction with tetrafluoroboric
acid (15 mol %) to deliver the final product 5. A simple final
column chromatography over silica gel was enough to obtain
2.6 g of pure ( )-9-epi-Ambrox (5), which exhibited the
expected strong, nice, and rich ambery odor. Importantly, the
global yield of the process from enyne 1 was relatively high
(35%; 86% average yield per step), and it should be noted that
the complete synthetic sequence from enyne 1 to ( )-9-epi-
Ambrox (5) was performed on a multigram scale and without
purification of any intermediate.
This robust synthetic protocol can be easily adapted for the
construction of analogues of 9-epi-Ambrox with potential utility
in the fragrance industry. As an example, we have synthesized
the 12-alkyl-substituted analogues 13 and 14 without problems
and in high global yield (Scheme 4). These compounds were
obtained submitting enol ether 7 to the microwave-promoted
Crude alkenyl bromide 3 was treated with tert-butyllithium
(2.5 equiv) in diethyl ether at −78 °C to obtain the anticipated
B
DOI: 10.1021/acs.orglett.6b02266
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 4. Synthesis of the 9-epi-Ambrox Analogues 13 and
14
Scheme 5. Synthesis of Both Enantiomers of 9-epi-Ambrox
from Racemic Alcohol 6
Claisen rearrangement followed by treatment of the so-formed
aldehyde with methylmagnesium bromide or ethylmagnesium
bromide, respectively. Finally, the corresponding secondary
alcohol analogue to 6 was cyclized to get product 13 or 14 by
the usual treatment with tetrafluoroboric acid (15 mol %).
At this point, it should be noted that studies of structure−
odor relationships performed on Ambrox (4) demonstrated
that the odor threshold of the enantiomer (+)-4 is about 10
times weaker than that of (−)-4.7 In addition, the racemate
( )-4 is only slightly weaker than (−)-4. As far as we know,
this kind of study has not been performed on 9-epi-Ambrox (5).
In fact, the synthesis of the enantiomer (+)-5 has not been
described.14 Having all this in mind, we considered the
possibility of adapting our synthetic route to 9-epi-Ambrox
(5) to obtain both enantiomers of the molecule. Under our
circumstances, having developed a procedure to obtain high
quantities of ( )-9-epi-Ambrox (5), we thought that the best
and shortest option to obtain both enantiomers of this
molecule would be the resolution of the racemic mixture of
an advanced intermediate of our synthetic sequence. In
particular, resolution of alcohol 6, the direct precursor of 9-
epi-Ambrox, seemed ideal. Although attractive, the resolution of
such a molecule where the stereogenic centers are far away
from the reactive site (alcohol) seemed challenging. In fact, we
were aware of the usually low enantioselectivities described in
the literature for lipase-catalyzed asymmetric acylations of
primary alcohols,15 particularly of those compounds such as 6
with remote stereogenic centers.16 Despite these negative
forewarnings, we decided to attempt the enzymatic resolution
of the racemate ( )-6 because it would provide the easiest and
shortest option to get both enantiomers of 9-epi-Ambrox.
Our initial attempts to perform the asymmetric acylation
using typical enzymes such as lipase AK (from Pseudomonas
fluorescens), lipase PS (from Burkholderia cepacia, formerly
called Pseudomonas cepacia lipase), or CAL-B (Candida
antarctica lipase B) were unsuccessful. However, to our delight,
appreciable enantioselectivities were observed with CAL-A
(Candida antarctica lipase A). Thus, after some optimization, we
found that the reaction of racemate ( )-6 with 2 equiv of vinyl
acetate in the presence of CAL-A (covalently attached to dry
acrylic polymer beads) in methyl tert-butyl ether (MTBE) at 10
°C delivered, after 5 h (70% conversion), the acetate 15 (er =
70:30) along with unreacted alcohol (−)-6 with high
enantioselectivity (er = 96:4) (Scheme 5). The simple acid-
promoted cyclization of this alcohol (−)-6 led to (−)-9-epi-
Ambrox [(−)-5].
derivative led to the alcohol (+)-6 (er = 93:7), and finally,
the acid-promoted cyclization of this alcohol led to (+)-9-epi-
Ambrox [(+)-5]. As far as we know, this is the first synthesis of
the (+)-enantiomer of 9-epi-Ambrox. Although a more detailed
study is necessary, it seems that the odor characteristics and
threshold of each enantiomer and the racemic mixture of 9-epi-
Ambrox are slightly different. This could be an interesting way
to create diverse perfume compositions by simply changing the
enantiomeric purity of the molecule.
In conclusion, a new and straightforward synthesis of the
powerful amber odorant 9-epi-Ambrox is described. The key
step of the process is a new cationic cyclization/nucleophilic
bromination reaction. This process, inspired by a reaction
reported more than 40 years ago, is expected to find further
application in the synthesis of other terpenes because the
alkenyl bromides available by this reaction are very versatile
reagents (they can be used as nucleophilic or electrophilic
partners). By means of this new brominative biomimetic
cyclization reaction, 9-epi-Ambrox could be obtained on a gram
scale. A dienyne derivative, easily available from geraniol, was
used as the starting material, and 9-epi-Ambrox was obtained
following a sequence of seven steps (35% global yield) with just
one purification process. We also describe access to both
enantiomers of 9-epi-Ambrox by an enzymatic resolution of an
advanced intermediate of the synthetic route. This resolution is
not only a striking case of remote stereocenter discrimination
but also an unusual example of enantioselective acylation of a
primary alcohol. The application of the strategy here described
in the context of “perfume discovery” by the synthesis of
analogues of 9-epi-Ambrox and the possibility of performing the
synthetic sequence in-flow are interesting issues to consider in
the near future.
To obtain the other enantiomer of the molecule, the acetate
derivative 15 (er = 70:30) was hydrolyzed and the so-formed
alcohol 6 was submitted to a second round of acetylation under
conditions identical to those discussed above (Scheme 5).
Thus, at 35% conversion, acetate 15 could be isolated with high
enantioselectivity (er = 93:7). Hydrolysis of this acetate
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.6b02266.
■
S
C
DOI: 10.1021/acs.orglett.6b02266
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Experimental details and characterization data for all new
Letter
(13) (a) Microwaves in Organic and Medicinal Chemistry; Kappe, C.
O., Stadler, A., Eds.; Wiley−VCH: Weinheim, 2005. (b) Microwaves in
Organic Synthesis; Loupy, A., Ed.; Wiley−VCH: Weinheim, 2002.
(14) Chiral pool based synthesis of 9-epi-Ambrox (see, for example,
refs 7, 8c and 8d) only allow the access to the (−)-enantiomer.
(15) Asymmetric Organic Synthesis with Enzymes; Gotor, V., Alfonso,
I., García-Urdiales, E., Eds.; Wiley−VCH: Weinheim, 2008.
(16) Cunha, R. L. O. R.; Ferreira, E. A.; Oliveira, C. S.; Omori, A. T.
Biotechnol. Adv. 2015, 33, 614−623.
compounds (PDF)
AUTHOR INFORMATION
Corresponding Authors
■
*E-mail: fjfv@uniovi.es.
*E-mail: frodriguez@uniovi.es.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We acknowledge financial support from MINECO-Spain
(Grant No. CTQ2013-41336-P) and MEC (FPU-predoctoral
grant to P.A.). Preliminary studies performed by Laura
́
Martinez (Master’s degree final project, University of Oviedo)
are also acknowledged. We are indebted to Dr Ivan Lavandera,
Dr. Vicente Gotor-Fernandez, and the rest of the Bioorganic
́
́
Chemistry Group of the University of Oviedo for their help and
advice with the enzymatic reactions.
REFERENCES
■
(1) Breitmaier, E. Terpenes: Flavors, Fragrances, Pharmaca, Pher-
omones; Wiley−VCH: Weinheim, 2006.
(2) (a) Yoder, R. A.; Johnston, J. N. Chem. Rev. 2005, 105, 4730−
4756. (b) Brunoldi, E.; Luparia, M.; Porta, A.; Zanoni, G.; Vidari, G.
Curr. Org. Chem. 2006, 10, 2259−2282.
(3) Alonso, P.; Pardo, P.; Galvan
Angew. Chem., Int. Ed. 2015, 54, 15506−15510.
́ ́
, A.; Fananas, F. J.; Rodríguez, F.
̃
(4) Ohloff, G. In Fragance Chemistry: The Science of the Sense of Smell;
Theimer, E. T., Ed.; Academic Press: New York, 1982; p 535.
(5) Ambrox [(−)-4] and Cetalox [( )-4] are trade names
(Firmenich). For the first preparation of Ambrox, see: Stoll, M.;
Hinder, M. Helv. Chim. Acta 1950, 33, 1251−1260.
(6) For some recent representative syntheses of Ambrox, see:
(a) Moulines, J.; Bats, J.-P.; Lamidey, A.-M.; Da Silva, N. Helv. Chim.
Acta 2004, 87, 2695−2705. (b) Xie, Y.-P.; Li, B.-G.; Luo, Y.-G.; Chen,
X.-Z.; Zhang, G.-L. Helv. Chim. Acta 2008, 91, 734−740. (c) Chapuis,
C. Helv. Chim. Acta 2014, 97, 197−214. (d) Rosales, A.; Foley, L. A.
́ ́
R.; Padial, N. M.; Munoz-Bascon, J.; Sancho-Sanz, I.; Roldan-Molina,
̃
́
E.; Pozo-Morales, L.; Irías-Alvarez, A.; Rodríguez-Maecker, R.;
Rodríguez-García, I.; Oltra, J. E. Synlett 2016, 27, 369−374. See
also: (e) Snowden, R. L. Chem. Biodiversity 2008, 5, 958−969 and
references cited therein.
(7) Ohloff, G.; Giersch, W.; Pickenhagen, W.; Furrer, A.; Frei, B.
Helv. Chim. Acta 1985, 68, 2022−2029.
(8) (a) Paquette, L. A.; Maleczka, R. E. J. J. Org. Chem. 1991, 56,
912−913. (b) Snowden, R. L.; Eichenberger, J.-C.; Linder, S. M.;
Sonnay, P.; Vial, C.; Schulte-Elte, K. H. J. J. Org. Chem. 1992, 57, 955−
960. (c) Kutney, J. P.; Cirera, C. Can. J. Chem. 1997, 75, 1136−1150.
(d) Yang, H.-J.; Li, B.-G.; Cai, X.-H.; Qi, H.-Y.; Luo, Y.-G.; Zhang, G.-
L. J. J. Nat. Prod. 2006, 69, 1531−1538.
(9) Related approaches have been used by other authors to obtain
the desired configuration at C9. For a strategy involving a [2,3]-Wittig
rearrangement, see, for example: (a) Oguchi, T.; Watanabe, K.;
Ohkubo, K.; Abe, H.; Katoh, T. Chem. - Eur. J. 2009, 15, 2826−2845.
For a strategy involving an Eschenmoser−Claisen rearrangement, see:
(b) Zhang, F.; Danishefsky, D. Angew. Chem., Int. Ed. 2002, 41, 1434−
1437.
(10) Johnson, W. S.; Gravestock, M. B.; Parry, R. J.; Okorie, D. A. J. J.
Am. Chem. Soc. 1972, 94, 8604−8605.
(11) Compound 1 was synthesized by transforming geraniol into the
corresponding geranyl bromide: Paz, J. L.; Rodrigues, J. A. R. J. J. Braz.
Chem. Soc. 2003, 14, 975−981. Formal coupling of the propargylic
subunit followed: Domingo, V.; Lorenzo, L.; Quilez del Moral, J. F.;
Barrero, A. F. Org. Biomol. Chem. 2013, 11, 559−562.
(12) Bosch, M.; Schlaf, M. J. J. Org. Chem. 2003, 68, 5225−5227.
D
DOI: 10.1021/acs.orglett.6b02266
Org. Lett. XXXX, XXX, XXX−XXX