Combined C-H activation/cope rearrangement as a strategic reaction in organic synthesis: Total synthesis of (-)-colombiasin A and (-)-elisapterosin B

Article abstract of DOI:10.1021/ja056877l

The total synthesis of (-)-colombiasin A (2) and (-)-elisapterosin B (3) has been achieved. The key step is a C-H functionalization process, the combined C-H activation/Cope rearrangement, between methyl (E)-2-diazo-3-pentenoate and 1-methyl-1,2-dihydrona

Full text of DOI:10.1021/ja056877l

Published on Web 01/31/2006
Combined C-H Activation/Cope Rearrangement as a Strategic
Reaction in Organic Synthesis: Total Synthesis of
(-)-Colombiasin A and (-)-Elisapterosin B
Huw M. L. Davies,* Xing Dai, and Matthew S. Long
Contribution from the Department of Chemistry, UniVersity at Buffalo, The State UniVersity of
New York, Buffalo, New York 14260-3000
Received October 7, 2005; E-mail: hdavies@acsu.buffalo.edu
Abstract: The total synthesis of (-)-colombiasin A (2) and (-)-elisapterosin B (3) has been achieved. The
key step is a C-H functionalization process, the combined C-H activation/Cope rearrangement, between
methyl (E)-2-diazo-3-pentenoate and 1-methyl-1,2-dihydronaphthalenes. When the reaction is catalyzed
by dirhodium tetrakis((R)-(N-dodecylbenzenesulfonyl)prolinate), Rh2(R-DOSP)4, an enantiomer differentiation
step occurs where one enantiomer of the dihydronaphthalene undergoes the combined C-H activation/
Cope rearrangement while the other undergoes cyclopropanation. This sequence controls the three key
stereocenters in the natural products such that the remainder of the synthesis is feasible using standard
chemistry.
Natural product synthesis continues to be a fertile area and
proving ground for the development of new synthetic methods.
On occasion, certain classes of natural products with a rich
combination of promising biological activity and intriguing
structural architecture become highly attractive synthetic targets.
A class of compounds that is generating much current interest
is a super-family of diterpenes that were isolated from gorgonian
corals.1-8 The diverse family of diterpenes, comprising from
bicyclic to hexacyclic systems, is completely derived biosyn-
thetically from (+)-elisabethatriene (1).2 Examples of these
natural products are (-)-colombiasin A (2),3,4 (-)-elisapterosin
B (3),5 and (+)-erogorgiaene (4).6 Many members of this super-
family display substantial biological activity as anti-inflamma-
tory, anticancer, antitubercular, and/or general antibacterial
agents.9 Due to the common biosynthetic ancestry of these
natural products,2 all have three distinctive stereocenters. From
a synthetic perspective, these three stereocenters have repre-
sented considerable challenges1,3-8 because there are no con-
venient neighboring functional groups available to assist in their
stereocontrol. This paper will describe a C-H functionalization
strategy that has the potential to be a universal solution of the
stereochemical issues associated with the synthesis of the three
stereocenters in these natural products.
(1) For a general review on the isolation, synthesis, and biosynthesis of these
natural products, see: Heckrodt, T. J.; Mulzer, J. Top. Curr. Chem. 2005,
244, 1-42.
(2) Coleman, A. C.; Kerr, R. G. Tetrahedron 2000, 56, 9569-9574.
(3) (-)-Colombiasin A, (a) Isolation: Rodriguez, A. D.; Ramirez, C. Org. Lett.
2000, 2, 507-510. Total synthesis: (b) Nicolaou, K. C.; Vassilikogiannakis,
G.; Magerlein, W.; Kranich, R. Angew. Chem., Int. Ed. 2001, 40, 2482-
2486. (c) Nicolaou, K. C.; Vassilikogiannakis, G.; Magerlein, W.; Kranich,
R. Chem. Eur. J. 2001, 7, 5359-5371. (d) Kim, A. I.; Rychnovsky, S. D.
Angew. Chem., Int. Ed. 2003, 42, 1267-1270. (e) Harrowven, D. C.;
Pascoe, D. D.; Demurtas, D.; Bourne, H. O. Angew. Chem., Int. Ed. 2005,
44, 1221-1222. (f) Boezio, A. A.; Jarvo, E. R.; Lawrence, B. M.; Jacobsen,
E. N. Angew. Chem., Int. Ed. 2005, 44, 6046-6050.
(4) For approaches toward (-)-colombiasin A: (a) Harrowven, D. C.; Tyte,
M. J. Tetrahedron Lett. 2001, 42, 8709-8711. (b) Chaplin, J. H.; Edwards,
A. J.; Flynn, B. L. Org. Biomol. Chem. 2003, 1, 1842-1844.
(5) (-)-Elisapterosin B, (a) Isolation: Rodriguez, A. D.; Ramirez, C.;
Rodriguez, I. I.; Barnes, C. L. J. Org. Chem. 2000, 65, 1390-1398. Total
synthesis: (b) Waizumi, N.; Stankovic, A. R.; Rawal, V. H. J. Am. Chem.
Soc. 2003, 125, 13022-13023. References 3d, 3e, and 3f.
(6) (+)-Erogorgiaene, (a) Isolation: Rodriguez, A. D.; Ramirez, C. J. Nat.
Prod. 2001, 64, 100-102. Total synthesis: (b) Cesati, R. R.; Armas, J.
D.; Hoveyda, A. H. J. Am. Chem. Soc. 2004, 126, 96-101. (c) Davies, H.
M. L.; Walji, A. M. Angew. Chem., Int. Ed. 2005, 44, 1733-1735. Formal
synthesis: (d) Harmata, M.; Hong, X. Tetrahedron Lett. 2005, 46, 3847-
3849.
(7) Elisabethin A, total synthesis: (a) Heckrodt, T. J.; Mulzer, J. J. Am. Chem.
Soc. 2003, 125, 4680-4681. Review: (b) Zanoni, G.; Franzini, M. Angew.
Chem., Int. Ed. 2004, 43, 4837-4841.
(8) Pseudopteroxazole: (a) Johnson, T. W.; Corey, E. J. J. Am. Chem. Soc.
2001, 123, 4475-4479. (b) Davidson, J. P.; Corey, E. J. J. Am. Chem.
Soc. 2003, 125, 13486-13489. (c) Harmata, M.; Hong, X.; Barnes, C. L.
Org. Lett. 2004, 6, 2201-2203. (d) Harmata, M.; Hong, X. Org. Lett. 2005,
7, 3581-3583.
The problems associated with the stereochemical issues of
these natural products can be readily seen in the published
9
10.1021/ja056877l CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 2485-2490
2485

A R T I C L E S
Scheme 1
Davies et al.
In this paper we describe a very different approach to control
the three stereocenters in these natural products. The approach
is based on a “combined C-H activation/Cope rearrangement”
between the vinyldiazoacetate 15 and dihydronaphthalenes 14,
which generates the three stereocenters in one step.6c
Our group has been developing new strategic reactions for
organic synthesis, which are based on regioselective intermo-
lecular C-H functionalization processes.11 Our approach to
achieve the C-H functionalization is by means of intermolecular
C-H insertions of rhodium carbenoids. Two major variants of
this theme have been discovered. The first is the direct C-H
insertion which can be conducted in a highly enantioselective
manner using the dirhodium tetraprolinate complex Rh2(S-
DOSP)4 as catalyst (eq 1).11 With use of this reaction, equivalent
transformations have been achieved to several of the classic
reactions of organic synthesis, such as the Aldol reaction,12
Mannich reaction,13 Michael addition,14 and the Claisen rear-
rangement.15 The second is the “combined C-H activation/Cope
rearrangement”, a transformation that often occurs in >98% ee
and >98% de (eq 2).16 This reaction occurs during allylic C-H
functionalization by vinyldiazoacetates.
syntheses of (-)-colombiasin A (2) and (-)-elisapterosin B
(3).3-5 The end game in the syntheses of these compounds has
been elegantly achieved by means of cycloaddition approaches
(Scheme 1). Nicolaou et al. demonstrated that (-)-colombiasin
A could be readily prepared by an intramolecular [4 + 2]
cycloaddition from the diene 5,3b,3c while Kim and Rychnovsky
developed a rapid entry into (-)-elisapterosin B by a Lewis
acid catalyzed [5 + 2] cycloaddition from 5.3d Recently,
Jacobsen and co-workers have shown that (-)-colombiasin A
can be converted into (-)-elisapterosin B by a Lewis acid
catalyzed reaction, possibly occurring by a retro [4 + 2]
cycloaddition followed by a [5 + 2] cycloaddition.3f
Even though the end game solution for the synthesis of (-)-
colombiasin A and (-)-elisapterosin B is very elegant and
efficient, the stereoselective synthesis of the three distinctive
stereogenic centers has been much more challenging (Scheme
2). In the synthesis related to (-)-colombiasin A, three main
retrosynthetic strategies have been developed. The first approach
employed a Tsuji allylation from 7 but this suffered from poor
regiocontrol, producing a 1:2.4 mixture of the 1,3- and 3,3-
rearrangement (8) products.3b,3c Furthermore, 8 is formed as the
wrong diastereomer, and several additional steps were required
to achieve the necessary epimerization. An alternative strategy
has been an intermolecular Diels-Alder reaction of benzo-
quinone 9 with a diene 10. Due to the lack of stereocontrol, the
exocyclic stereocenter in the diene needed to be stereospecifi-
cally introduced prior to the cycloaddition. In the initial process
reported by Kim and Rychnovsky, the diastereoselectivity in
the cycloaddition was low (1:1.7),3d but recently, Jacobsen and
co-workers have greatly improved this process by using chiral
Lewis acids to influence the diastereoselectivity of this
cycloaddition.3f Due to the stereochemical challenges of these
natural products, many groups have avoided the problem by
starting their syntheses with commercially available mono-
terpenes.8a,8b,10 This strategy has been recently used by Har-
rowven et al. starting from the monoterpene 12 which is
converted to 13 and includes a nice cascade to complete the
total synthesis of (-)-colombiasin A.3e This approach is effective
but it does have the drawback that a different annulation strategy
would have to be designed for each natural product synthesis.
The combined C-H activation/Cope rearrangement also has
the potential of being a surrogate for some of the classic
reactions of organic synthesis.17 This can be illustrated by
considering a hypothetical approach for the synthesis of (-)-
colombiasin A that would in principle be applicable to many
other members of these diterpenes (Scheme 3). A flexible
precursor to (-)-colombiasin A would be the ester 17. If 17 is
derived from the diene 18, a hypothetical approach to generate
18 with controlled stereochemistry would be a tandem Claisen
rearrangement/Cope rearrangement from 19. Both reactions
would be expected to proceed through a chair transition state
where the ester stereogenic center in 19 would dictate the
stereochemistry in the formation of the two new stereogenic
centers in 18. This scheme has to be considered hypothetical
because there would be no driving force for the Cope
(11) For recent reviews: (a) Davies, H. M. L.; Beckwith, R. E. J. Chem. ReV.
2003, 103, 2861-2903. (b) Davies, H. M. L.; Loe, Ø. Synthesis 2004, 16,
2595-2608.
(12) (a) Davies, H. M. L.; Beckwith, R. E. J.; Antoulinakis, E. G.; Jin, Q. J.
Org. Chem. 2003, 68, 6126-6132. (b) Davies, H. M. L.; Antoulinakis, E.
G. Org. Lett. 2000, 2, 4153-4156.
(13) (a) Davies, H. M. L.; Hansen, T.; Hopper, D. W.; Panaro, S. A. J. Am.
Chem. Soc. 1999, 121, 6509-6510. (b) Davies, H. M. L.; Venkataramani,
C. Org. Lett. 2001, 3, 1773-1775. (c) Davies, H. M. L.; Venkataramani,
C. Angew. Chem., Int. Ed. 2002, 41, 2197-2199. (d) Davies, H. M. L.;
Jin, Q. Org. Lett. 2004, 6, 1769-1772.
(14) Davies, H. M. L.; Ren, P. J. Am. Chem. Soc. 2001, 123, 2070-2071.
(15) (a) Davies, H. M. L.; Gregg, T. M. Tetrahedron Lett. 2002, 43, 4951-
4953. (b) Davies, H. M. L.; Walji, A. M.; Townsend, R. J. Tetrahedron
Lett. 2002, 43, 4981-4983. (c) Davies, H. M. L.; Ren, P.; Jin, Q. Org.
Lett. 2001, 3, 3587-3590.
(16) (a) Davies, H. M. L.; Jin, Q. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 5472-
5475. (b) Davies, H. M. L.; Jin, Q. J. Am. Chem. Soc. 2004, 126, 10862-
10863.
(17) Davies, H. M. L.; Beckwith, R. E. J. J. Org. Chem. 2004, 69, 9241-9247.
(9) Rodriguez, I. I.; Shi, Y.-P.; Garcia, O. J.; Rodriguez, A. D.; Mayer, A. M.
S.; Sanchez, J. A.; Ortega-Barria, E.; Gonzalez, J. J. Nat. Prod. 2004, 67,
1672-1680.
(10) Corey, E. J.; Lazerwith, S. E. J. Am. Chem. Soc. 1998, 120, 12777-12782.
9
2486 J. AM. CHEM. SOC. VOL. 128, NO. 7, 2006

(−)-Colombiasin A and (−)-Elisapterosin B Synthesis
A R T I C L E S
Scheme 2
Scheme 3
Scheme 4
dronaphthalene undergoes the C-H activation/Cope rearrange-
ment to form 21 while the other enantiomer undergoes a
cyclopropanation to form 22. Completion of the synthesis of
(+)-erogorgiaene (4) was readily achieved in four additional
steps from 21. As the C-H functionalization begins at a site
well away from the aromatic ring, we made the hypothesis that
this reaction would be little influenced by the aromatic ring
functionality. In this paper we demonstrate that a highly
oxygenated aromatic ring is equally compatible with this
carbenoid chemistry, leading to very direct access to (-)-
colombiasin A and (-)-elisapterosin B.
Three dihydronaphthalenes with different protecting groups,
methyl (14a), acetyl (14b), and tert-butyldimethylsilyl (14c)
were chosen as appropriate substrates for the crucial combined
C-H activation/Cope rearrangement. The synthesis of 14 started
from the p-quinone 9 following a [4 + 2] cycloaddition sequence
described by Nicolaou et al. (Scheme 5).3b,3c Reaction of the
quinone 9 with the diene 24 generated the cycloadduct which
on isomerization to the corresponding quinol could be trapped
under different conditions to afford the dimethyl derivative ether
25a, the diacetyl derivative 25b, and the disilyl derivative 25c.
Acidic hydrolysis of the resulting TBS-enol ether in 25 gave
rise to the ketone 26 in good yields. Initially, we investigated
utilizing a reduction/elimination strategy to form the CdC
double bond of 14 from the â-tetralone 26, but this gave a
rearrangement.16a The “combined C-H activation/Cope rear-
rangement” would be an equivalent of this hypothetical reaction,
as illustrated in the conversion of 20 to 18.
We have communicated a proof-of-concept study which
demonstrated that the combined C-H activation/Cope rear-
rangement can be effectively applied to the rapid construction
of (+)-erogorgiaene (Scheme 4).6c The C-H functionalization
with the vinyldiazoacetate 15 is especially impressive because
it is an enantiodivergent step. One enantiomer of the dihy-
9
J. AM. CHEM. SOC. VOL. 128, NO. 7, 2006 2487

A R T I C L E S
Scheme 5
Davies et al.
Scheme 6
Scheme 7
Scheme 8
mixture of double-bond isomers. The overall transformation
could be achieved through initially converting the â-tetralone
26 to the corresponding vinyl triflate 27 using Comins’ reagent,18
before carrying out a palladium-catalyzed reductive coupling.19
With quantities of the three dihydronaphthalenes 14 in hand,
the key rhodium carbenoid step was examined. From the
(18) Comins, D. L.; Dehghani, A. Tetrahedron Lett. 1992, 33 (42), 6299-6302.
(19) Scott, W. J.; Stille, J. K. J. Am. Chem. Soc. 1986, 108, 3033-3040.
9
2488 J. AM. CHEM. SOC. VOL. 128, NO. 7, 2006

(−)-Colombiasin A and (−)-Elisapterosin B Synthesis
A R T I C L E S
Scheme 9
conception of the project, it was proposed that the functionality
on the aromatic ring would not interfere with the combined C-H
activation step. This was definitely the case with the dimethoxy
derivative 14a as the Rh2(R-DOSP)4-catalyzed reaction of 14a
with the vinyldiazoacetate 15 gave a 1:1 mixture of the C-H
functionalization product 28 and the cyclopropane 29 as single
diastereomers (Scheme 6). Furthermore, the C-H functional-
ization product 28 was formed with the correct relative
stereochemistry for the natural products and in 92% ee.
catalyzed cyclopropanations.20b Further studies are in progress
to fully understand the nature of the acetoxy effect.
In contrast to the diacetoxy system, the disilyl derivative 14c
was an exceptional substrate for the combined C-H activation/
Cope rearrangement. The Rh2(R-DOSP)4-catalyzed reaction of
15 with 14c gave a 1:1 mixture of the C-H functionalization
product 32 and the cyclopropane 33 (Scheme 8). Since the two
products could not be separated at this stage, the mixture was
hydrogenated and then reduced to the alcohols 34 and 35. The
desired C-H functionalization product 35 was isolated in 34%
yield (68% in theory) as a single diastereomer in >95% ee over
three steps.
The enantiomer differentiation in these reactions can be
rationalized as illustrated in Scheme 9. Excellent predictive
models have been developed for both the rhodium prolinate-
catalyzed C-H activation/Cope rearrangement16b and the cy-
clopropanation.21 The chiral catalysts are considered to adopt a
D2 symmetric arrangement and can be viewed simply with a
blocking group in the front and another in the back. Applying
these models to the 4-methyl-1,2-dihydronaphthalenes as sub-
strates leads to an interesting prediction. The matched enanti-
omer for the C-H activation/Cope rearrangement is opposite
to the matched enantiomer for the cyclopropanation. Conse-
quently, a situation exists for enantiomer differentiation in which
one enantiomer preferentially undergoes the C-H activation/
Cope rearrangement to form 33, while the other undergoes
cyclopropanation to form 32. From a practical perspective, this
The C-H functionalization chemistry of the diacetyl deriva-
tive 14b gave a surprising result (Scheme 7). The Rh2(R-
DOSP)4-catalyzed reaction between 15 and 14b proceeded in
high yield (92%) but the 1:4.7 ratio of the C-H functionalization
product 30 to the cyclopropane 31 was much different from
the 1:1 ratio of the reaction with 14a. Even so, both 30 and 31
were produced with very high diastereoselectivity (>95% de)
but the enantioselectivity for the C-H functionalization product
30 was 89% ee, while the cyclopropane 31 was formed in only
30% ee. This result is not consistent with our previous
observations and is indicative that the acetoxy groups interfere
with the chiral discrimination by the prolinate catalyst on the
cyclopropanation, although the C-H functionalization selectivity
is not markedly changed. A possible explanation for this strange
effect is that the acetoxy group coordinates to the carbenoid
prior to the cyclopropanation event and this interferes with the
chiral influence of the catalyst. Ester coordination to a carbenoid
has been implicated in asymmetric cyclopropanations with chiral
ester auxilaries20a and the presence of methyl benzoate as an
additive greatly enhances the turnover numbers of rhodium-
(20) (a) Davies, H. M. L.; Huby, N. J. S.; Cantrell, W. R., Jr.; Olive, J. L. J.
Am. Chem. Soc. 1993, 115, 9468-9479. (b) Davies, H. M. L.; Venkat-
aramani, C. Org. Lett. 2003, 5, 1403-1406.
9
J. AM. CHEM. SOC. VOL. 128, NO. 7, 2006 2489

A R T I C L E S
Scheme 10
Davies et al.
is even better than kinetic resolution because the complex
dihydronaphthalene can be used as the limiting agent.
diene 40 to boron trifluoride etherate at -78 °C for 1 h resulted
in a [5 + 2] cycloaddition to give (-)-elisapterosin B in 51%
yield. Under these conditions (-)-colombiasin A methyl ether
41 was generated as a side product in 21% yield. The spectral
data for (-)-colombiasin A and (-)-elisapterosin B were in full
agreement with the literature data.3,5
In conclusion, the total synthesis of (-)-colombiasin A (2)
and (-)-elisapterosin B (3) has been achieved in 14 and 13
steps, respectively, from quinone 9. The key step is the
combined C-H activation/Cope rearrangement, which is an
excellent method for the construction of the three stereogenic
centers common to the various diterpenes isolated from
Pseudopterogorgia elisabethae.
Although both dihydronaphthalene 14a and 14c gave excel-
lent results in the C-H activation/Cope rearrangement, 14c was
used to complete the total synthesis since it would be easier to
unveil the quinone moiety by deprotection of TBS ethers rather
than methyl ethers. Thus, conversion of the alcohol 35 to the
key diene 40 was achieved using very standard steps (Scheme
10). PCC oxidation of 35 followed by a Grignard addition
generated the allylic alcohol 38. Conversion of 38 to the triflate
followed by elimination generated the diene 39, which was
readily desilylated and air-oxidized to the quinone 40.
Kim and Rychnovsky have previously shown that the diene
40 can be converted to (-)-colombiasin A by an intramolecular
Diels-Alder reaction, while treatment of 40 with boron trif-
luoride etherate generates (-)-elisapterosin B by means of a [5
+ 2] cycloaddition.3d Thus, when diene 40 was heated at 180
°C in toluene, (-)-colombiasin A methyl ether (41) was isolated
in 88% yield).3b,3c The total synthesis of (-)-colombiasin A (2)
was completed by demethylation of 41 with AlCl3.3d Exposing
Acknowledgment. The National Science Foundation (CHE-
0350536) is gratefully acknowledged.
Supporting Information Available: Full experimental pro-
cedures and characterization data for all new compounds. This
material is available free of charge via the Internet at http://
pubs.acs.org. See any current masthead page for ordering
information and Web access instructions.
(21) Nowlan, D. T., III; Gregg, T. M.; Davies, H. M. L.; Singleton, D. A. J.
Am. Chem. Soc. 2003, 125, 15902-15911.
JA056877L
9
2490 J. AM. CHEM. SOC. VOL. 128, NO. 7, 2006