Towards the synthesis of prevezol C: Total enantioselective synthesis of (-)-2-epi-prevezol C

Article abstract of DOI:10.1016/j.tetlet.2010.07.019

(-)-2-epi-Prevezol C was readily accessed from the chirons (-)- and (+)-limonene oxide in a total of nine steps and in 24% yield. This efficient enantioselective synthesis of this complex product utilises a highly stereoconvergent, substrate-controlled allylic alkylation strategy to assemble rapidly the unprecedented diterpene core.

Full text of DOI:10.1016/j.tetlet.2010.07.019

Tetrahedron Letters 51 (2010) 4808–4811
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Towards the synthesis of prevezol C: total enantioselective synthesis of
(ꢀ)-2-epi-prevezol C
Michael Blair, Craig M. Forsyth, Kellie L. Tuck
*
School of Chemistry, Monash University, Clayton, Vic. 3800, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
(ꢀ)-2-epi-Prevezol C was readily accessed from the chirons (ꢀ)- and (+)-limonene oxide in a total of nine
steps and in 24% yield. This efficient enantioselective synthesis of this complex product utilises a highly
stereoconvergent, substrate-controlled allylic alkylation strategy to assemble rapidly the unprecedented
diterpene core.
Received 19 May 2010
Revised 17 June 2010
Accepted 2 July 2010
Available online 8 July 2010
Ó 2010 Elsevier Ltd. All rights reserved.
A family of novel cytotoxic-brominated diterpenes, prevezols
B–E, was isolated from the organic extracts of the red alga
Laurencia obtusa (Fig. 1).1 Elucidation of their structures and
relative stereochemistry, by comprehensive spectral data analyses,
revealed a family of syn-bromohydrins with carbon skeletons that
were unprecedented in the literature. The structure of prevezol A
had been previously established, and the discovery of metabolites
C–E allowed revision of the earlier proposed structure of prevezol
B.2
Whilst modern spectroscopic techniques have advanced in
recent years, regio and stereoselective synthesis still remains a
requirement for the confirmation of structure and determination
of absolute stereochemistry. This has been highlighted recently
in an excellent review by Nicolaou and Snyder regarding the
misassignment of natural products3 and in an article by Burton
and co-workers, which reassessed the structures of elatenyne
and chloroenyne natural products extracted from Laurencia
majuscula.4 As the first step in the synthesis of these complex
natural products we investigated strategies for the synthesis of
prevezol C (Fig. 1), to explore routes for the formation of this
unique carbon skeleton and to confirm the reported structure of
the Western hemisphere (ring B).
we synthesised the easier trans-bromohydrin of prevezol C [(ꢀ)-
epi-prevezol C 1], epimeric at the 2-position (Fig. 2).
Br
Br
Br
OH
OH
OH
A
B
OH
OH
OH
O
HO
prevezol C
prevezol B
prevezol A
Br
Br
HO
H
HO
OH
OH
HO
HO
H
H
H
H
prevezol E
prevezol D
Figure 1. The structures of prevezols A–E.
The relative stereochemistry of the Eastern hemisphere (ring A)
of prevezol C was assigned due to spectral agreement with the val-
ues reported for prevezol A,1,2 which was identified by comparison
of the 13C NMR data of obtusadiol and rogioldiol A (Fig. 2).2,5 The
relative stereochemistry of the Western hemisphere (ring B) of
prevezol C was assigned based on NOESY correlations. Thus, before
synthesis of the challenging syn-bromohydrin moiety commenced
we needed to develop methodology to connect the Eastern and
Western hemispheres of prevezol C and to develop strategies for
constructing and confirming the required stereochemistry in the
Western hemisphere. Thus, as a prelude to the harder problem,
OH
Br
OH
A
A
Br
Br
Br
OH
rogioldiol A
obtusadiol
Br
OH
2
A
B
OH
OH
epi-prevezol C 1
* Corresponding author. Tel.: +61 3 99054510; fax: +61 3 99054597.
E-mail address: kellie.tuck@monash.edu (K.L. Tuck).
Figure 2. The structures of obtusadiol, rogioldiol A and epi-prevezol C (1).
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.07.019

M. Blair et al. / Tetrahedron Letters 51 (2010) 4808–4811
4809
Due to the structural homology shared by the two hemispheres
of epi-prevezol C (1) we envisaged a stereoconvergent synthetic
approach utilising limonene oxide (Scheme 1). The key disconnec-
KHMDS, DMPU,
allyl bromide, THF,
-78 °C to rt
92%
tion in the retrosynthesis is
a diastereoselective carbon–
O
O
OTBS
OTBS
carbon bond-forming alkylation reaction of the ketone 2 and the
allylic halide 3. It was anticipated that the alkylated product 4
would form diastereoselectively as a result of the isoprenyl substi-
tuent preferentially residing in the equatorial position, thereby
controlling the stereochemical course of the reaction. However if
this is not the case, epimerisation of the syn-isomer to the thermo-
dynamically stable anti-isomer 4 should be facile. Recent efforts in
our laboratories have focused on accessing the trans-diaxial and
trans-diequatorial diols of limonene oxide, which are derived from
a commercially available diastereomeric mixture of cis and trans
isomers, in high yields and diastereoselectivities.6 We hoped to
employ the trans-diaxial diol 5 in the total synthesis of (ꢀ)-epi-
prevezol C (1). It was also anticipated that the required allylic
halide 3 could be readily synthesised from (ꢀ)-trans-limonene
oxide via an allylic halogenation reaction.
The required tert-butyldimethylsilyloxy-ketone 2 was synthes-
ised according to our published procedure.7 Standard allylic alkyl-
ation conditions of the lithium enolate of the silyloxy ketone 2,
formed at ꢀ78 °C with LDA,8 with allyl bromide gave poor yields.
Employing a modified version of the conditions reported by
Palomo et al.,9 which used an excess of KHMDS in the presence
of 20% v/v DMPU (Scheme 2), gratifyingly afforded the allylated
species 6 in excellent conversion and as a single diastereomer, as
determined by analysis of its 1H NMR spectrum.
2
6
single diastereomer
Scheme 2. Synthesis of the model alkylated product 6.
control was obtained due to the 1,3-diaxial interaction encoun-
tered by the incoming electrophile with the axial TBS group, and
thereby alkylating in a twist-boat conformation,10 and not due to
base-catalysed epimerisation of an axially alkylated intermediate
(Fig. 3). Furthermore, quenching the reaction at ꢀ40 °C with
saturated ammonium chloride afforded the equatorial product,
exclusively.
Given the positive outcome of the model alkylation studies, we
next focused our attention on the synthesis of the key electrophile,
10-halo-trans-limonene oxide 3. Preparation of the allylic halide 3,
by halogenation of trans-limonene oxide was not trivial; tradi-
tional Wohl–Ziegler conditions (NBS, CCl4, AIBN or dibenzoylper-
oxide),11 and derivatives thereof (NBS, AcOH)12 gave complex
mixtures. However, optimisation of Moreno-Dorado’s conditions,13
utilising
a stoichiometric quantity of iodometrically titrated
sodium hypochlorite,14 in the presence of either CeCl3 (1.1 equiv)
or InCl3 (1.1 equiv),15 in a binary solvent system (1:1 CH2Cl2/
H2O) afforded the allylic chloride 3a in high yield (Scheme 3).
Minimal amounts of the bis-chloro by-product were observed in
the 1H NMR spectrum of the crude reaction mixture. Recently, Jas-
The stereochemistry at the a-carbon of the allylated product 6
was elucidated to be the thermodynamic trans product via analysis
of the COSY and NOESY spectra. The coupling constants (J = 11.8,
9.5 and 2.5 Hz) of the proton adjacent to the carbonyl group were
also consistent with the proton existing in the axial position. The
reaction of the allylated product 6 with 5% NaOMe (5 days) showed
no change in its 1H NMR spectrum, which is consistent with an
equatorial allyl chain. It is postulated that excellent diastereofacial
trzebska et al. demonstrated that the p-allyl complex derived from
an allylic chloride16 could be used as an alternative to the more
routinely employed allylic acetate and carbonate electrophiles.17
Unfortunately, application of these conditions to our allylic chlo-
ride 3a proved unsuccessful. The conditions for the model alkyl-
ation, KHMDS/DMPU (2.2 equiv/20% v/v), also did not result in
the alkylated product 4a. Pleasingly, conversion of the allylic chlo-
ride 3a into the corresponding more reactive iodo species 3b
yielded an extremely reactive electrophile which participated eas-
ily in an allylic alkylation reaction with the enolate derived from
ketone 2, even in the absence of DMPU (Scheme 3).18 Like the
model studies, the diterpene oxide 4a was also observed to be a
single diastereomer by 1H NMR spectroscopic analysis. The treat-
ment of the diterpene oxide 4a with TBAF (1.5 equiv) in refluxing
THF readily cleaved the tert-butyldimethylsilyl group to give 4b
diastereoselective
α-allylation
Br
O
OH
O
OH
OH
OH
4
2-epi-prevezol C 1
half-chair approach
I
allylic
halogenation
O
X
OTBS
O
syn-isomer
not observed
O
OH
OTBS
OH
H
K
2
5
2
3
diastereo-
selective
hydrolysis
OTBS
OTBS
H
O
K
ring
O
H
inversion
2
O
H
6
O
I
anti-isomer
formed exclusively
twist-boat approach
(+)-cis/trans-
limonene oxide
(-)-trans-
limonene oxide
Figure 3. The diastereomorphic transition states for substrate control to form the
desired allylated ketone 6.
Scheme 1. Retrosynthesis of epi-prevezol C 1.

4810
M. Blair et al. / Tetrahedron Letters 51 (2010) 4808–4811
O
O
O
O
NaOCl, CeCl3,
CH2Cl2, H2O, 0 °C
R
(CH3COO)3BHNa,
AcOH, THF, -40 °C
91%
O
OH
OH
3a
; R=Cl
OH
NaI, Me2CO
72%
91%
4b
8
3b; R=I
O
LiBr, AcOH,
THF -5 °C
64%
Br
KHMDS, -78 °C,
30 min; then 3b,
-78 °C to rt
O
OTBS
O
OH
93%
OR
2
4a; R=OTBS
4b
TBAF,THF, Δ
98%
; R=H
OH
OH
OH
OH
(-)-2-epi-prevezol C (1)
5% H2SO4(aq), THF
0 °C to rt
Scheme 4. Synthesis of epi-prevezol C (1).
29%
O
OH
7
Scheme 3. Synthesis of the key intermediate oxide 4b.
with a single diastereomer observed by 1H NMR spectroscopic
analysis (Scheme 3). As a final confirmation of the absolute stereo-
chemistry of the diterpene core, 4b was reacted under acidic con-
ditions to give the corresponding triol 7. An X-ray crystal structure
of the triol 7 confirmed that the alkylation reaction had occurred to
give the depicted stereochemistry (Fig. 4).19,20
Formation of the required trans-diol 8, the penultimate com-
pound, occurred exclusively via a chelation-controlled stereoselec-
tive ketone reduction employing sodium triacetoxyborohydride in
acetic acid/acetonitrile at ꢀ40 °C (Scheme 4).21,22 A regiospecific
acid-catalysed nucleophilic oxirane opening was achieved with
LiBr/AcOH in cold THF,23 to afford the 2-bromo-epimer of prevezol
C (1) in 64% yield.24
Comparison of the 13C NMR spectroscopy resonances of
2-epi-prevezol C (1) with prevezol C showed excellent correlation,
with the expected exceptions of the resonances of the epimeric
carbon, C2, and its neighbours (Supplementary data). In the Wes-
tern hemisphere the resonances of the stereogenic centres C9,
C10, C13 and C14 closely corresponded with those reported for
prevezol C (<0.6 ppm difference, Figure 5 and Supplementary
data). This leads us to believe that the relative stereochemistry of
the Western hemisphere has been correctly assigned in the natural
product.
Figure 5. Comparison of the 13C NMR signals of the Western hemisphere of
prevezol C and 2-epi-prevezol C (1). Horizontal and vertical axes show carbon
number and
Dd values. The numbering of the rings is consistent with that used by
Iliopoulou et al.1
synthesis of a prevezol analogue and the synthetic construction
of their unique carbocyclic framework.
Acknowledgements
In conclusion we have synthesised (ꢀ)-2-epi-prevezol C (1) in
nine steps and 24% overall yield from (ꢀ)-trans-limonene oxide
and (+)-cis/trans-limonene. This work represents the first total
The authors acknowledge the support of the School of
Chemistry, Monash University and Monash University.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2010.07.019.
References and notes
1. Iliopoulou, D.; Mihopoulos, N.; Vagias, C.; Papazafiri, P.; Roussis, V. J. Org. Chem.
2003, 68, 7667–7674.
2. Mihopoulos, N.; Vagias, C.; Mikros, E.; Scoullosb, M.; Roussis, V. Tetrahedron
Lett. 2001, 42, 3749–3752.
3. Nicolaou, K. C.; Snyder, S. A. Angew. Chem., Int. Ed. 2005, 44, 1012–1044.
4. Sheldrake, H. M.; Jamieson, C.; Pascu, S. I.; Burton, J. W. Org. Biomol. Chem. 2009,
7, 238–252 and references within.
5. (a) Howard, B. M.; Fenical, W. Tetrahedron Lett. 1978, 19, 2453–2456; (b)
Guella, G.; Marchetti, F.; Pietra, F. Helv. Chim. Acta 1997, 80, 684–694.
Figure 4. The X-ray crystal structure of triol 7.

M. Blair et al. / Tetrahedron Letters 51 (2010) 4808–4811
4811
6. Blair, M.; Andrews, P. C.; Fraser, B. H.; Forsyth, C. M.; Junk, P. C.; Massi, M.; Tuck,
K. L. Synthesis 2007, 1523–1527.
7. Blair, M.; Tuck, K. L. Tetrahedron: Asymmetry 2009, 20, 2149–2153.
8. Srikrishna, A.; Dethe, D. H. Org. Lett. 2004, 6, 165–168.
9. Palomo, C.; Oiarbide, M.; Mielgo, A.; Gonzalez, A.; Garcia, J. M.; Landa, C.;
Lecumberri, A.; Linden, A. Org. Lett. 2001, 3, 3249–3252.
10. House, H. O.; Trost, B. M. J. Org. Chem. 1965, 30, 2502–2512.
11. Constantino, M. G.; de Oliveira, K. T.; Polo, E. C.; da Silva, G. V. J.; Brocksom, T. J.
J. Org. Chem. 2006, 71, 9880–9883.
4.66 (s, 1H), 4.73 (s, 1H), 4.79 (t, J = 1.5 Hz, 1H). 13C NMR (125 MHz) d ꢀ2.3, 1.8,
18.4, 23.3, 24.0, 24.8, 26.1, 26.8, 30.5, 30.6, 31.2, 40.2, 42.1, 46.7, 55.1, 57.7,
59.6, 78.8, 107.4, 112.8, 146.3, 152.3, 212.3. IR (film): 3076 (w), 2932 (br s),
2858 (m), 1720 (s), 1644 (s), 1472 (m) 1448 (m), 1376 (m), 1254 (s), 1210 (w),
1170 (m), 1132 (m), 1095 (m), 1050 (m), 1005 (m), 956 (m), 861 (m), 835 (s).
HRMS (ESI) calcd for C26H44NaO3Si (M+Na)+ 455.2957; found 455.2957.
19. The X-ray crystal data of 7ꢂH2O (CCDC-776506) can be obtained free of charge
from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
datarequest/cif.
12. Srikrishna, A.; Hemamalini, P. J. Org. Chem. 1990, 55, 4883–4887.
13. Moreno-Dorado, F. J.; Guerra, F. M.; Manzano, F. L.; Aladro, F. J.; Jorge, Z. D.;
Massanet, G. M. Tetrahedron Lett. 2003, 44, 6691–6693.
14. Mendham, J.; Denney, R. C.; Barnes, J. D.; Thomas, M. J. K. Vogel’s
Quantitative Chemical Analysis, 6th ed.; Prentice Hall: England, 2000.
Chapter 10, p 437.
20. All attempts to convert the secondary alcohol into the corresponding bromide
with inversion of configuration were unsuccessful.
21. Evans, D. A.; Clark, J. S.; Metternich, R.; Novack, V. J.; Sheppard, G. S. J. Am.
Chem. Soc. 1990, 112, 866–868.
22. With NaBH4/MeOH (ꢀ5 °C; 5 min)
a 5:1 (trans/cis) mixture of diols was
obtained (determined by 1H NMR spectroscopic analysis).
15. Pisoni, D. S.; Gamba, D.; Fonseca, C. V.; da Costa, J. S.; Petzhold, C. L.; de Oliveira,
E. R.; Ceschi, M. A. J. Braz. Chem. Soc. 2006, 17, 321–327.
16. Negishi, E.; Matsushita, H.; Chatterjee, S.; John, R. A. J. Org. Chem. 1982, 47, 3188–3190.
17. Jastrzebska, I.; Scaglione, J. B.; DeKoster, G. T.; Rath, N. P.; Covey, D. F. J. Org.
Chem. 2007, 72, 4837–4843.
23. Bajwa, J. S.; Anderson, R. C. Tetrahedron Lett. 1991, 32, 3021–3024.
24. (ꢀ)-2-epi-Prevezol C (1): LiBr (72 mg, 829
introduced through a rubber septum, via the aid of a syringe, to a solution of
diterpene diol 8 (49 mg, 152.9 mol) in THF (1.5 mL) and AcOH (28.4 L,
458.7 mol) at 0 °C under an inert atmosphere. The reaction mixture was
lmol) in anhydrous THF (2 mL) was
l
l
l
18. (2S,5R,6S)-2-(tert-Butyldimethylsilyloxy)-2-methyl-6-(2-((1S,3S,6R)-6-methyl-7-
allowed to warm at an ambient temperature, and the reaction was deemed
complete after 2 h via TLC analysis (1:1 EtOAc/hexanes). The reaction mixture
was quenched with saturated NaHCO3(aq) (5 mL), partitioned with Et2O
(15 mL), washed with saturated brine (20 mL) and dried (MgSO4).
oxabicyclo[4.1.0]heptan-3-yl)allyl)-5-(prop-1-en-2-yl)
cyclohexanone
(4a):
KHMDS (0.5 M solution in toluene; 14.10 mL, 7.05 mmol) was added slowly
to a stirred solution of the ketone 2 (1.66 g, 5.88 mmol) in THF (35 mL) at
ꢀ78 °C, under an inert atmosphere. After 30 min, a solution of the allylic iodide
3b (1.82 g, 6.54 mmol) in THF (35 mL) was added slowly. The mixture was
warmed to an ambient temperature, stirred for 12 h and then the reaction was
quenched with cold saturated NH4Cl(aq) (75 mL). The mixture was extracted
with Et2O (100 mL); the organic fraction was washed with brine (100 mL),
dried over (MgSO4) and concentrated in vacuo. Flash column chromatography
(20:1 hexanes/EtOAc) afforded the title compound as a colourless oil (2.37 g,
Concentration under reduced pressure afforded the title compound as
a
colourless oil (39 mg, 64%). ½a D20
ꢁ
ꢀ18.0 (c 1.0, CHCl3). 1H NMR (400 MHz) d 1.41
(s, 3H), 1.45–1.49 (m, 2H), 1.56–1.69 (m, 6H) 1.69 (s, 3H), 1.72–1.79 (m, 3H),
1.89 (m, 8H) 2.16 (d, J = 14.5 Hz, 1H), 2.39–2.47 (m, 1H), 3.36 (s, 1H), 4.14–4.16
(m, 1H), 4.74–4.78 (m, 2H), 4.85 (s, 1H), 4.87 (s, 1H). 13C NMR (125 MHz) d
18.9, 26.0, 27.1, 28.4, 29.8, 33.1, 33.3, 35.3, 35.6, 36.2, 36.9, 44.3, 60.2, 71.79,
71.84, 73.7, 110.4, 111.9, 148.1, 151.0. IR (film): 3424 (br s), 3072 (w), 2928 (s),
2857 (m), 1702 (w), 1640 (m), 1451 (m), 1376 (m), 1260 (m), 1180 (m), 1099
(m), 1030 (m), 924 (w), 891 (m), 798 (m), 759 (m). HRMS (ESI) calcd for
93%). ½a 2D0
ꢁ
ꢀ8.9 (c 1.0, CHCl3). 1H NMR (500 MHz) d 0.03 (s, 3H), 0.147 (s, 3H),
0.94 (s, 9H), 1.30 (s, 3H), 1.31 (s, 3H), 1.42–1.48 (m, 3H) 1.61–1.67 (m, 2H), 1.71
(s, 3H) 1.77–1.87 (m, 3H), 1.99–2.08 (m, 4H), 2.15–2.21 (m, 1H), 2.28 (dd,
J = 15.9, 10.8 Hz, 1H), 2.97 (d, J = 5.5 Hz, 1H), 3.39–3.43 (m, 1H), 4.57 (s, 1H),
C
20H32BrO3 (MꢀH)ꢀ 399.1535; found 399.1536.