Efficient and stereoselective synthesis of precursors for epoxypolyene cyclizations via allylic substitutions

Article abstract of DOI:10.1055/s-2005-871933

A catalytic and highly stereoselective approach to precursors for epoxypolyene cyclizations based on copper-catalyzed allylic substitutions is described. The method allows introduction of various aromatic, benzyl, and alkyl groups without affecting epoxides. The compounds obtained are valuable starting materials for the synthesis of natural products and biologically active substances. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2005-871933

1954
LETTER
Efficient and Stereoselective Synthesis of Precursors for Epoxypolyene
Cyclizations via Allylic Substitutions
Efficient
E
poxypo
n
lyene
S
ynthe
d
sis reas Gansäuer,* José Justicia, Antonio Rosales, Björn Rinker
Kekulé-Institut für Organische Chemie und Biochemie, Universität Bonn, Gerhard-Domagk-Str. 1, 53121 Bonn, Germany
Fax +49(228)734760; E-mail: andreas.gansaeuer@uni-bonn.de
Received 19 May 2005
R2CuLi
Abstract: A catalytic and highly stereoselective approach to pre-
cursors for epoxypolyene cyclizations based on copper-catalyzed
allylic substitutions is described. The method allows introduction of
various aromatic, benzyl, and alkyl groups without affecting
epoxides. The compounds obtained are valuable starting materials
for the synthesis of natural products and biologically active sub-
stances.
X
R
or RMgX,
O
O
cat. Li2CuCl4
1: X = Br
2: X = OAc
R = alkyl, benzyl, aryl
Scheme 2 Synthesis of precursors for epoxypolyene cyclizations
utilizing organolithium and Grignard reagents
Key words: catalysis, cuprates, epoxides, Grignard reagent, epoxy-
polyene cyclization
benzyl and functionalized alkyl groups to epoxypolyenes.
We reasoned that this could be achieved by allylic sub-
Epoxypolyene cyclizations have emerged as a very pow- stitution reactions with cuprates derived from either
erful tool for the synthesis of various kinds of natural Grignard or organolithium reagents as shown in
products with important biological activities, e.g. poly- Scheme 2.
cyclic terpenoids and steroids.1
One serious point of concern is immediately apparent,
However, the specific usefulness of polycyclization however. It is well documented that epoxides can react
strategies is frequently and noticeably limited by an in- with cuprates5 prepared from Grignard reagents,6 or
efficient access to the starting materials. This is especially organolithium compounds7 through ring-opening via a
so when specific functional groups are necessary for con- nucleophilic substitution. Thus, the critical question of
trolling the selectivity of the termination of the cationic chemoselectivity of allylic substitution versus epoxide
sequences.
opening has to be resolved. To the best of our knowledge
a solution to this problem has not yet been described in the
literature.
In radical chemistry a more efficient strategy has recently
been realized by Cuerva et al.2 based on titanocene-
catalyzed3 domino cyclizations. The necessary substrates To retard epoxide opening we started our investigation
were prepared from epoxypolyene carbonates by a Stille with epoxygeranyl bromide (1) containing the highly re-
coupling with aryl stannanes as shown in Scheme 1.4
active allyl bromide group. The results of the introductory
studies with 1 and other electrophiles are summarized in
Table 1.
Pd(dba)2 (10 mol%)
O
O
When 1 was reacted with Bu2CuLi, the desired compound
3 was obtained in good yield and with high diastereoselec-
tivity (dr 96:4). This encouraging finding was surpassed
when the readily available and distinctly more stable and
less reactive epoxygeranyl acetate (2) was employed as
substrate. Epoxyolefin 3 was obtained in 79% yield with-
out any product arising from epoxide opening essentially
as a single isomer (dr >98:2). Thus, even the relatively
inert allylic acetates are more reactive towards cuprates
than epoxides.
Bu3SnPh-2-OBn,
48 h
O
O
OEt
OBn
70%
Scheme 1 Approach towards aryl epoxypolyprenes based on Stille
couplings
The toxicity of the stannanes, that are usually prepared
from Grignard or organolithium reagents, and more im-
portantly their limitations in transmetallation abilities,
only aryl groups were transferred, restricts the generality
of the method somewhat.
We also demonstrated that the Li2CuCl4-catalyzed (4
mol%)9 substitution reaction of BuMgBr with 2 and 4
gave the desired products in reasonable to high yield and
very high diastereoselectivity in all cases. This result is
even more surprising as Grignard reagents open epoxides
even more swiftly in the presence of copper salts. While
the experiments were initially carried out at a 1 mmol
scale, performing the reaction of 2 with 10 mmol of
substrate also gave a satisfactory result (72%). With
In connection with a project aimed at the development of
efficient novel radical tandem sequences we required
exactly such a broader method also enabling transfer of
SYNLETT 2005, No. 12, pp 1954–1956
Advanced online publication: 07.07.2005
DOI: 10.1055/s-2005-871933; Art ID: G12905ST
© Georg Thieme Verlag Stuttgart · New York
0
1
.0
8
.2
0
0
5

LETTER
Efficient Epoxypolyene Synthesis
1955
Table 2 Li2CuCl4-Catalyzed (4 mol%) Transfer of Alkyl and
Benzyl Grignard Reagent Groups to 2
Table 1 Copper-Mediated or -Catalyzed Transfer of n-Butyl
Groups to Epoxypolyene Electrophilesa
Substrate/methodb
Product
Yield (%)
75
Grignard reagent
Product
Yield (%)
70
1/A
BrMg
OBn
OBn
n-Bu
O
O
3
3
8
2/A
2/B
79
72
79
74
OPMB
OBn
BrMg
BrMg
OPMB
OBn
3
O
9
73
68
54
O
O
( )2
n-Bu
OAc
( )2
4/B
5
O
90
10
n-Bu
OAc
MeO
Ph-4-OMe
Ph-3-Me
O
O
2
2
MgCl
6/B
a For details see ref.8
7
O
11
b Method A: Bu2CuLi/THF. Method B: BuMgBr/THF/Li2CuCl4
(4 mol%).
MgCl
O
epoxyfarnesyl acetate (6) an excellent yield of the desired
product (90%) was obtained. It should be noted that for
the sake of convenience 6 was usually employed as a 2:1
mixture of diastereoisomers, which was prepared from
12
Table 3 Copper-Mediated or Li2CuCl4-Catalyzed (4 mol%)
commercially available farnesol that consists of the same Transfer of Aryl Organometallic Reagent Groups to 2
2:1 mixture of isomers. However, the more expensive
isomerically pure 6 can also be used without any isomer-
ization.
Nucleophile
Product
Yield (%)a
83/A
MgBr
From a practical point of view Li2CuCl4 catalysis is clear-
ly superior for two reasons. First, no excess of the organo-
metallic reagent has to be used and second Grignard
reagents10 are usually easier to prepare than the corres-
ponding organolithium compounds.
Ph
O
13
69/A
68/A
76/A
68/B
OMe
MgBr
Ph-2-OMe
Ph-4-OMe
2-Np
With these reaction conditions in hand we synthesized a
number of epoxyolefins containing different and func-
tionalized alkyl and benzyl substituents from 2 as summa-
rized in Table 2.
O
14
MeO
MgBr
The reactions proceeded smoothly in all cases. Only dur-
ing the synthesis of 8 we observed a minor amount of the
allylic isomer of 8 (dr 93:7) that could be easily removed
by column chromatography, however. The ether substitut-
ed products 8, 9, and 10 are important intermediates for
the synthesis of spirodihydrofuranolactams from Stachy-
botrys chartarum,11 e.g. stachybotrylactam. In the case of
the transfer of benzyl groups the use of Grignard reagents
also proved to be more efficient than the use of the
organolithium reagents. Compounds 11 and 12 constitute
valuable starting materials for the synthesis of natural
products from the tasmanite sediment12a and ferruginol,12b
respectively.
O
15
MgBr
O
16
Li
2-Furyl
O
O
17
a Method A: RMgBr/THF/Li2CuCl4 (4 mol%). Method B: R2CuLi/
THF.
Synlett 2005, No. 12, 1954–1956 © Thieme Stuttgart · New York

1956
A. Gansäuer et al.
LETTER
(6) Whitesell, J. K.; Lawrence, R. M.; Chen, H. H. J. Org.
Finally, as many meroterpenes arising from epoxypolyene
cyclizations contain aryl groups, we also checked the
transfer of various aryl groups to 2. It turned out that the
reaction of PhMgBr catalyzed by Li2CuCl4 (4 mol%) gave
the desired 13 in distinctly higher yield (83%, dr >97:3)
than the reaction with Ph2CuLi that resulted in a complex
mixture of products. 2-NaphthylMgBr gave a similar re-
sult in the preparation of 16. Various functionalized aryl
Grignard reagents, that were prepared by the excellent
methods recently introduced by Knochel,10 also gave sat-
isfactory yields of the desired epoxpolyenes. These results
are summarized in Table 3. In all cases investigated the
reaction was completely diastereoselective.
Chem. 1986, 51, 4779.
(7) (a) Johnson, C. R.; Herr, R. W.; Wieland, D. M. J. Org.
Chem. 1973, 38, 4263. (b) Sirat, H. M.; Thomas, E. J.;
Wallis, J. D. J. Chem. Soc., Perkin Trans. 1 1982, 2885.
(8) Typical Experimental Procedures.
Preparation of Compound 3.
To a solution of 2 (2.120 g, 10 mmol) in dry THF (90 mL) a
0.1 M solution of Li2CuCl4 (5.4 mL, 0.5 mmol) was added
dropwise after cooling to 0 °C. Then a solution of BuMgCl
(7 mL, 2 M, 14 mmol) was added dropwise over 20 min. The
mixture was stirred at 0 °C for 3 h and then at r.t. overnight.
Then sat. aq NH4Cl was added and the mixture extracted
with EtOAc, dried over MgSO4 and the solvent was removed
in vacuo. The residue was purified by SiO2 chromatography
(cyclohexane–EtOAc, 97:3) to yield 3 (1520 mg, 72%):
colorless oil. 1H NMR (400 MHz, CDCl3): d = 5.17 (t of app.
sext, J = 7.0, 1.1 Hz, 1 H), 2.69 (t, J = 6.3 Hz, 1H), 2.04–
2.19 (m, 2 H), 1.97 (q, J = 7.0 Hz, 2 H), 1.54–1.70 (m, 2 H),
1.60 (s, 3 H), 1.25–1.36 (m, 6 H), 1.29 (s, 3H), 1.25 (s, 3 H),
0.88 (t, J = 7.1 Hz, 3 H). 13C NMR (100 MHz, CDCl3,
DEPT): d = 133.9, 125.6, 64.3, 58.4, 36.5, 31.7, 29.6, 28.0,
27.6, 25.0, 22.7, 18.9, 16.1, 14.2. IR (film): 2925, 1460,
1375, 1120 cm–1. HRMS (EI): m/z calcd for C14H26O:
210.1984; found: 210.1986.
In the case of the introduction of the furyl group the orga-
nolithium reagent obtained via deprotonation of furan by
n-BuLi was employed. The isolated yield of 17 was in the
same range as those obtained with the aryl Grignards.
Compound 17 constitutes a key intermediate for the prep-
aration of pumiloxide.13
In summary, we have devised a short and efficient method
for the preparation of various precursors for epoxypoly-
ene cyclizations. Our compounds are important interme-
diates for the synthesis of a number of natural products
and their analogues. Our method features the use of readi-
ly available Grignard and organolithium reagents and cir-
cumvents the preparation, use, and removal of toxic
stannanes necessary for the synthesis of the same products
via Stille reactions.
Preparation of Compound 14.
To a vigorously stirred suspension of Mg (29 mg, 1.2 mmol)
in THF (2 mL) was added 2-bromoanisol (225 mg, 1.2
mmol). The mixture was stirred at reflux until dissolution of
the metal and added dropwise to a solution of 2 (85 mg, 0.4
mmol) and Li2CuCl4 (0.20 mL of a 0.1 M solution in THF,
0.02 mmol) in THF (2 mL) at 0 °C. Stirring was continued
for 4 h at r.t. After addition of a sat. solution of NH4Cl the
mixture was extracted with EtOAc, dried over MgSO4, and
the solvent was removed in vacuo. Purification by SiO2
chromatography (cyclohexane–EtOAc, 95:5) yielded 14 (72
mg, 69%): colorless oil. 1H NMR (400 MHz, CDCl3): d =
7.17 (td, J = 7.7, 1.5 Hz, 1 H), 7.12 (dd, J = 7.4, 1.5 Hz, 1 H),
6.87 (td, J = 7.4, 1.1 Hz, 1 H), 6.77 (br d, 1 H), 5.36 (tq,
J = 7.3, 1.2 Hz, 1 H), 3.82 (s, 3 H), 3.34 (d, J = 7.3 Hz, 2 H),
2.70 (t, J = 6.2 Hz, 1 H), 2.31–2.05 (m, 2 H), 1.73 (s, 3 H),
1.71–1.57 (m, 2 H), 1.26 (s, 3 H), 1.24 (s, 3 H). 13C NMR
(100 MHz, CDCl3, DEPT): d = 157.5, 135.1, 130.0, 129.5,
127.1, 123.4, 120.6, 110.4, 64.4, 58.5, 55.5, 36.5, 28.5, 27.6,
25.0, 18.9, 16.2. IR (film): 2970, 2930, 1720, 1600, 1490,
1460, 1380, 1290 cm–1. HRMS (EI): m/z calcd for C17H24O2:
260.1776; found: 260.1782.
Acknowledgment
J. J. is grateful to the University of Granada for a postdoctoral grant.
A. R. is generously supported by the Fundación Ramón Areces. A.
G. is indebted to the Fonds der Chemischen Industrie (Dozenten-
stipendium, Sachbeihilfen) for continuing financial assistance.
References
(1) (a) Abe, I.; Rohmer, M.; Prestwich, G. D. Chem. Rev. 1993,
93, 2189. (b) Wendt, K. U.; Schulz, G. E.; Corey, E. J.; Liu,
D. R. Angew. Chem. Int. Ed. 2000, 39, 2812.
(2) Justicia, J.; Rosales, A.; Buñuel, E.; Oller-López, J. L.;
Valdivia, M.; Haïdour, A.; Oltra, J. E.; Barrero, A. F.;
Cárdenas, D. J.; Cuerva, J. M. Chem.–Eur. J. 2004, 10, 1778.
(3) (a) Gansäuer, A.; Bluhm, H. Chem. Rev. 2000, 100, 2771.
(b) Gansäuer, A.; Lauterbach, T.; Narayan, S. Angew. Chem.
Int. Ed. 2003, 42, 5556.
(4) Justicia, J.; Oltra, J. E.; Cuerva, J. M. J. Org. Chem. 2004,
69, 5803.
(5) Review: Krause, N.; Gerold, A. Angew. Chem., Int. Ed.
Engl. 1997, 36, 186.
(9) Tamura, M.; Kochi, J. Synthesis 1971, 303.
(10) Knochel, P.; Dohle, W.; Gommermann, N.; Kneisel, F. F.;
Kopp, F.; Sapountzis, I.; Vu, V. A. Angew. Chem. Int. Ed.
2003, 42, 4302.
(11) Jarvis, B. B. Phytochemistry 2003, 64, 53.
(12) (a) Azevedo, D. A.; Aquino Neto, F. R.; Simoneit, B. R. T.
Org. Mass. Spectrom. 1990, 25, 475. (b) Brandt, C. W.;
Neubauer, L. G. J. Chem. Soc. 1939, 103.
(13) Raldugin, V. A.; Demenovka, L. I.; Pentegova, V. A. Khim.
Prir. Soedin. 1978, 3, 345.
Synlett 2005, No. 12, 1954–1956 © Thieme Stuttgart · New York