Enantioselective total synthesis of (-)-laurenditerpenol

Article abstract of DOI:10.1021/ol302106y

A highly convergent total synthesis of (-)-laurenditerpenol has been accomplished through an organolithium to aldehyde nucleophilic addition. Preparation of the prerequisite key intermediates in optically pure form was based on an improved, short, and efficient synthesis of wine lactone from (S)-limonene and Corey's catalytic enantioselective Diels-Alder reaction of 2,5-dimethyl furan with diethyl fumarate.

Full text of DOI:10.1021/ol302106y

ORGANIC
LETTERS
2012
Vol. 14, No. 17
4666–4669
Enantioselective Total Synthesis of
(ꢀ)-Laurenditerpenol
Emmanuel N. Pitsinos,* Nikolaos Athinaios, and Veroniki P. Vidali
NCSR “Demokritos”, P.O. Box 60228, GR-153 10 Ag. Paraskevi, Athens, Greece
pitsinos@chem.demokritos.gr
Received July 30, 2012
ABSTRACT
A highly convergent total synthesis of (ꢀ)-laurenditerpenol has been accomplished through an organolithium to aldehyde nucleophilic addition.
Preparation of the prerequisite key intermediates in optically pure form was based on an improved, short, and efficient synthesis of “wine lactone”
from (S)-limonene and Corey’s catalytic enantioselective DielsꢀAlder reaction of 2,5-dimethyl furan with diethyl fumarate.
Hypoxia-inducible factor-1 (HIF-1) small molecule
inhibitors are considered potential new anticancer drug
leads that could exploit the hypoxic microenvironment of
developing solid tumors.1 Laurenditerpenol (1, Figure 1)
was isolated from the marine alga Laurencia intricata
as a potent and selective inhibitor of HIF-1 in breast
tumor cells (IC50 0.4 μM). However, only the absolute
configuration at C(1) and the relative configuration of
Figure 1. (ꢀ)-Laurenditerpenol, 1.
the 7-oxabicylco[2.1.1]heptane ring could be established
route have been prepared in a stereocontrolled manner.5
We report herein an enantioselective total synthesis of
laurenditerpenol.6
through spectroscopic studies.2 Full elucidation of the
absolute stereostructure relied on the stereocontrolled
synthesis of several diastereomeric pairs and comparison
of spectroscopic data, as well as biological activities, of the
purified diastereomers with those of the natural product.3
The total synthesis of racemic laurenditerpenol has been
accomplished,4 and two key intermediates invoked by this
The presence of two isolated ring systems invites the
retrosynthetic disconnection of the C(8)ꢀC(9) bond to
dissect the molecule in two precursors of similar size and
complexity (Scheme 1). In previous synthetic efforts retro-
synthetic addition of either a double bond or activating
group(s) at this locality was required before this structu-
rally simplifying retrosynthetic transform could be ap-
plied. Thus, the former approach has been adopted for
the preparation of simplified analogues (through an
olefin cross-metathesis),6 as well as for the total synthesis
of 1 in racemic form (via 2 through a JuliaꢀKocienski
olefination).4 However as this synthesis has illustrated, the
(1) (a) Xia, Y.; Choi, H.-K.; Lee, K. Eur. J. Med. Chem. 2012, 49,
24–40. (b) Melillo, G. Cancer Metastasis Rev. 2007, 26, 341–352.
(c) Belozerov, V. E.; Van Meir, E. G. Anti-Cancer Drugs 2005, 16,
901–909. (d) Park, J.-W.; Chun, Y.-S.; Kim, M.-S. J. Pharmacol. Sci.
2004, 94, 221–232.
(2) Mohammed, K. A.; Hossain, C. F.; Zhang, L.; Bruick, R. K.;
Zhou, Y.-D.; Nagle, D. G. J. Nat. Prod. 2004, 67, 2002–2007.
(3) Chittiboyina, A. G.; Kumar, G. M.; Carvalho, P. B.; Liu, Y.;
Zhou, Y.-D.; Nagle, D. G.; Avery, M. A. J. Med. Chem. 2007, 50, 6299–
6302.
(4) (a) Jung, M. E.; Im, G.-Y. J. Tetrahedron Lett. 2008, 49, 4962–
4964. (b) Jung, M. E.; Im, G.-Y. J. J. Org. Chem. 2009, 74, 8739–8753.
(5) Mukherjee, S.; Scopton, A. P.; Corey, E. J. Org. Lett. 2010, 12,
1836–1838.
(6) For prior exploratory synthetic studies from our group, see:
Athinaios, N.; Kazantzis, A.; Putzker, K.; Lewis, J.; Pitsinos, E. N.
Lett. Org. Chem. 2009, 6, 269–271.
r
10.1021/ol302106y
Published on Web 08/20/2012
2012 American Chemical Society

Scheme 1. Retrosynthetic Plan for (ꢀ)-Laurenditerpenol 1
Scheme 2. Synthesis of the C(9)ꢀC(15) Fragment
lack of stereocontrol in the olefination step diminishes
overall efficiency. On the other hand, the latter approach,
in the form of alkylation of a sulfone with an allylic bromide
to provide 3, has been exploited for the establishment of
the absolute stereostructure of 1.3 Unfortunately, this route
forfeits stereocontrol at C(7).
In order to overcome the above-mentioned drawbacks
we aimed to establish the complete carbon framework of 1
through either a SuzukiꢀMiyaura or Negishi-type alkylꢀ
alkyl cross-coupling7 between coupling partners 4 and 5.
Alcohol 6 could serve as the source of the left-part coupling
partner. This alcohol was anticipated to be derived from
“wine lactone” 7 that, in turn, is known to be accessible
from (S)-limonene.8 Preparation of the right-part coupling
partner 5 would exploit the anticipated exo-facial selectiv-
ity9 in monohydrolysis of the known dicarboxylate 9.10
Indeed, this dicarboxylate was prepared as previously
described through a catalytic enantioselective DielsꢀAlder
reaction of diethyl fumarate (10) and 2,5-dimethyl-furan
(11) and subsequent hydrogenation of the cycloadduct.10
Upon hydrolysis with LiOH in aqueous ethanol, an in-
separable mixture of the two monoacids 12a and 12b
was obtained (Scheme 2). Subsequent reduction with
timing of subsequent planned transformations. Thus, al-
cohol 13a (ee = 92% based on HPLC analysis using a
chiral column) was protected as the corresponding TBS-
ether (14), and the remaining carboxylate wasreduced with
DIBAL-H at ꢀ78 °C to provide in 90% overall yield the
monoprotected diol 15. Subsequent treatment with diphe-
nyl disulfide in the presence of tri-n-butylphosphine11
furnished thioether 16 in 95% yield. Then, reduction with
Raney-Ni followed by acidic hydrolysis of the silylether
provided alcohol 18, which was converted to the corre-
sponding iodide 19 (ee = 92%; 10 steps, 37% overall yield
from diethyl fumarate).
Scheme 3. AlkylꢀAlkyl Cross-Coupling Studies
BH3 THF led to isolation of the corresponding alcohols
3
in 74% and 13% yield respectively from 9. The major
product was alcohol 13a. Thus it was revealed that, in the
case of 7-oxabicylco[2.1.1]heptane diester 9, monohydro-
lysis hadproceededwithendo-selectivity instead of the exo-
one previously observed in related bicylco[2.1.1]heptanes.9
The only consequence of this observed reversal of selectiv-
ity, be it unexpected and intriguing, was a reversal in the
(7) (a) Chemler, S. R.; Trauner, D.; Danishefsky, S. J. Angew. Chem.,
Int. Ed. 2001, 40, 4544–4568. (b) Jana, R.; Pathak, T. P.; Sigman, M. S.
Chem. Rev. 2011, 111, 1417–1492.
(8) Guth, H. Helv. Chim. Acta 1996, 79, 1559–1571.
(9) (a) Niwayama, S.; Hiraga, Y. Tetrahedron Lett. 2003, 44, 8567–
8570. (b) Niwayama, S.; Cho, H.; Zabet-Moghaddam, M.; Whittlesey,
B. R. J. Org. Chem. 2010, 75, 3775–3780.
(10) Liu, D.; Canales, E.; Corey, E. J. J. Am. Chem. Soc. 2007, 129,
1498–1499.
(11) Nakagawa, I.; Hata, T. Tetrahedron Lett. 1975, 16, 1409–1412.
Org. Lett., Vol. 14, No. 17, 2012
4667

With iodide 19 in hand, we proceeded to investigate
the feasibility of the planned alkylꢀalkyl cross-coupling
employing the conditions that Fu et al. have developed for
the efficient palladium-catalyzed SuzukiꢀMiyaura cross-
coupling of alkyl bromides with alkyl boranes.12 Thus,
when the readily available8 borane 20 was used as the
alkylating agent, the expected mixture of diastereomeric
cross-coupling products 21 with racemic iodide13 19 was
obtained albeit in low yield (Scheme 3). Borate 23 (R = H)
is presumed to be the actual alkylating species in this
transformation,12 and alkyl borates have been employed
successfully in palladium-catalyzed cross-coupling reac-
tions with alkenyl halides.14 Since generation of an alkyl-
borane more relevant to our plans (4, R1 = BR2; Scheme 1),
through hydroboration, was anticipated to provide a mix-
ture of C(7) epimers,8 use of borate 23 (R = Me) was
attempted as an alternative to borane 20. However,
this exchange failed to furnish any of the desired cross-
coupling products with iodide 19 under otherwise identical
conditions. Equally disappointing were the results ob-
tained using borate 23 (R = Me), iodide 19, and nickel15
or copper16 based catalysts. Moreover, Fu’s conditions did
not promote the coupling of iodide 2217 with the borate
derived from 19 upon lithiation and quenching with
9-MeO-BBN.
In parallel to the above-mentioned cross-coupling stu-
dies, preparation of the C(1)ꢀC(8) fragment had begun
(Scheme 4). It is known8 that treatment of acid 24 with
PDC and t-BuOOH in benzene provides in one step and
25% yield a 1:1 mixture of lactones 28a and 28b. This
procedure however was deemed inappropriate for the large
scale preparation of lactone 28a. Thus, an alternative,
more reliable, and efficient procedure was sought.18 To
this end, methyl ester 25, that was obtained in three steps
as a 1:1 mixture of diastereomers from (S)-limonene,8 was
converted to a mixture of diastereomeric epoxides 26. The
aim was to exploit the organoselenium based method
developed by Sharpless and Lauer19 for their conversion
to allylic alcohols 27. However, upon nucleophilic opening
of the mixture of epoxides 26 by the phenylselenide anion
prepared in situ from diphenyldiselenide/sodium borohy-
dride in methanol and subsequent treatment with hydro-
gen peroxide, direct formation of lactones 28a and 28b as a
1:1 mixture was observed. This mixture could be chroma-
tographically separated and the undesired epimer (28b)
gave a new mixture of 28a/28b (>4:1) upon treatment with
t-BuOK in t-BuOH/THF.18c Thus, all material could be
converted to the desired epimer 28a.
Scheme 4. Synthesis of the C(1)ꢀC(8) Fragment
Conversion of (()-28a to alcohol (()-29 through a four-
step sequence has been described for the synthesis of
racemic laurenditerpenol4 and was used for the prepara-
tion of (ꢀ)-29 from (þ)-28a. However, efficient conversion
of this alcohol or its mesylate derivative to the correspond-
ing iodide could not be achieved. Thus, further explora-
tions along the alkylꢀalkyl cross-coupling approach (i.e.,
evaluation of Negishi-type cross-coupling conditions7b)
were discouraged, and in conjunction with the disappoint-
ing results of the SuzukiꢀMiyaura alkylꢀalkyl cross-
coupling studies (vide supra), we were forced to recon-
sider how to join the fully functionalized C(1)ꢀC(8) and
C(9)ꢀC(15) fragments.
Addition of an organometal species derived from iodide
19 to aldehyde 30, which is readily available from alcohol
29 upon oxidation of the latter with TPAP/NMO,4 ap-
pearedasanattractivealternativecouplingmethodtoward
the fully functionalized carbon framework of 1. Such an
approach, if successful, would fulfill the key requirements
set forth in our original retrosynthetic analysis: (a) to
circumvent the drawbacks associated with the formation
of a C(8)ꢀC(9) double bond and (b) to maintain control of
the C(7) stereocenter. With both iodide 19 and aldehyde 30
in hand, temptation to test this approach overruled the
alarming report that lithiation of iodide 19 and quenching
€
(12) Netherton, M. R.; Dai, C.; Neuschutz, K.; Fu, G. C. J. Am.
Chem. Soc. 2001, 123, 10099–10100.
(13) Racemic iodide 19 was prepared from racemic 9. The latter was
obtained after hydrogenation of the HfCl4-mediated DielsꢀAlder ad-
duct of dimethylfuran and diethyl fumarate as described by: Hayashi,
Y.; Nakamura, M.; Nakao, S.; Inoue, T.; Shoji, M. Angew. Chem., Int.
Ed. 2002, 41, 4079–4082.
(17) Duvold, T.; Francis, G. W.; Papaioannou, D. Tetrahedron Lett.
1995, 36, 3153–3156.
(18) For other syntheses of “wine lactone”, see: (a) Bergner, E. J.;
Helmchen, G. Eur. J. Org. Chem. 2000, 419–423. (b) Chavan, S. P.;
Kharul, R. K.; Sharma, A. K.; Chavan, S. P. Tetrahedron: Asymmetry
2001, 12, 2985–2988. (c) Serra, S.; Fuganti, C. Helv. Chim. Acta 2004, 87,
2100–2109.
(14) (a) Soderquist, J. A.; Matos, K.; Rane, A.; Ramos, J. Tetrahe-
€
dron Lett. 1995, 36, 2401–2402. (b) Furstner, A.; Seidel, G. Tetrahedron
1995, 51, 11165–11176.
(15) Saito, B.; Fu, G. C. J. Am. Chem. Soc. 2007, 129, 9602–9603.
(16) Yang, C.-T.; Zhang, Z.-Q.; Liu, Y.-C.; Liu, L. Angew. Chem.,
Int. Ed. 2011, 50, 3904–3907.
(19) Sharpless, K. B.; Lauer, R. F. J. Am. Chem. Soc. 1973, 95, 2697–
2699.
4668
Org. Lett., Vol. 14, No. 17, 2012

2-propanesulfonates with lithium triethylborohydride.20
Initial attempts to reduce the corresponding xanthates 32
with Bu3SnH/AIBN at 110 °C led to the formation of an
inseparable 4:1 mixture of a byproduct that was tentatively
assigned structure 33 and 34. Running the reaction at lower
temperature (80 °C) improved the ratio to 2:1, but the
desired deoxygenated product 34 still remained the minor
component. Finally, replacement of AIBN with Et3B as a
radical initiator allowed the reaction to proceed at ambient
temperature21 leading to preferential formation of the
radical reduction product 34 over the radical intramolecular
cyclization one 33 (33:34 = 1:2). Subsequent treatment
of this mixture with TBAF cleaved the silyl protective
groups allowing, after careful chromatography, isolation
of (ꢀ)-laurenditerpenol (1) and alcohol 35.
Scheme 5
In summary, an improved, short, and more efficient
preparation of (3R,3aR,7aS)-“wine lactone” from readily
available (S)-limonene secured in optically pure form
aldehyde 30 (fully functionalized C(1)ꢀC(8) segment; 11
steps, 5% overall yield) while a previously reported cata-
lytic asymmetric DielsꢀAlder reaction between diethyl
fumarate and 2,5-dimethyl furan was exploited for the
efficient preparation of iodide 19 in 92% ee (fully functio-
nalized C(9)ꢀC(15) segment; 10 steps, 37% overall yield).
The feasibility of an alternative coupling of these two key
intermediates was demonstrated allowing (in 31% yield, in
four steps) the total synthesis of (ꢀ)-laurenditerpenol (1).
witha one-carbon electrophile had failed toyield any ofthe
desired product.4b
Without any optimization, lithiation of 19 followed by
slow addition at ꢀ100 °C of a solution of aldehyde 30 in
diethyl ether led to the formation of alcohols 31 in 30%
yield as a 3:1 mixture of C(8) epimers, along with 45% of
recovered aldehyde 30 (Scheme 5). Subsequent removal of
the redundant hydroxyl group proved to be a nontrivial
transformation. Thus, attempted reductive cleavage of the
mesylate derivatives of alcohols 31 with lithium aluminum
hydride or lithium triethylborohydride led to recovery of the
original alcohols while formation of complex product mix-
tures was observed upon treatment of the corresponding
Acknowledgment. This study was performed within the
frame of COST Action CM0804 “Chemical Biology with
Natural Products”. N.A. thanks NCSR “D” for a PhD
scholarship and Prof. A. Yiotakis for being the academic
responsible for his PhD thesis. Prof. E. Theodorakis and
Dr. M. Dakanali (Department of Chemistry, UCSD) are
gratefully acknowledged for their assistance in obtaining
HR-MS spectra for some of the synthesized compounds.
Supporting Information Available. Spectral data and
experimental procedures for compounds 13, 15, 16, 18,
19, 21, 26, 28ꢀ31, 33ꢀ35, and 1. This material is available
free of charge via the Internet at http://pubs.acs.org.
(20) Hua, D. H.; Sinai-Zingde, G.; Venkataraman, S. J. Am. Chem.
Soc. 1985, 107, 4088–4090.
(21) Nozaki, K.; Oshima, K.; Utimoto, K. Tetrahedron Lett. 1988,
29, 6125–6126.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 17, 2012
4669