A palladium-catalyzed carbo-oxygenation: The bielschowskysin case

Article abstract of DOI:10.1021/ol401285d

An asymmetric synthesis of an advanced tetracyclic intermediate toward the synthesis of bielschowskysin (1) is described. A biomimetic [2 + 2]-photocyclization was used to establish the cyclobutane core of bielschowskysin. Macrocyclization under Heck conditions led to an unprecedented carbo-oxygenation of a 1,1-disubstituted double bond.

Full text of DOI:10.1021/ol401285d

ORGANIC
LETTERS
2013
Vol. 15, No. 12
3098–3101
A Palladium-Catalyzed Carbo-
oxygenation: The Bielschowskysin Case
Martin Himmelbauer, Jean-Baptiste Farcet, Julien Gagnepain, and Johann Mulzer*
€
University of Vienna, Department of Organic Chemistry, Wahringer Str. 38,
1090 Vienna, Austria
johann.mulzer@univie.ac.at
Received May 10, 2013
ABSTRACT
An asymmetric synthesis of an advanced tetracyclic intermediate toward the synthesis of bielschowskysin (1) is described. A biomimetic [2 þ 2]-
photocyclization was used to establish the cyclobutane core of bielschowskysin. Macrocyclization under Heck conditions led to an unprecedented
carbo-oxygenation of a 1,1-disubstituted double bond.
Over the past years the Rodrıguez group has reported
the isolation of a stunning variety of terpenoids from the
Caribbean Sea plume Pseudopterogorgia kallos, among
which bielschowskysin (1)1 (Scheme 1) has attracted an
unusual amount of interest.
Scheme 1. Biosynthesis and Reported [2 þ 2]-Approaches
Partly this is due to its significant antiplasmodial activity
against the malaria causing protozoan parasite Plasmodium
falciparum and its cytotoxic activity against two human
cancer cell lines. Most significantly, its densely functiona-
lized polycyclic diterpenoid structure including an unprece-
dented tricyclo[9.3.0.02,10]-tetradecane ring system and 11
stereogenic centers has rendered bielschowskysin a highly
competitive target in synthetic chemistry.
So far, activities from numerous research groups, including
our own,2 have resulted in several advanced intermediates3
and test systems.4
According to the studies by Roethle and Trauner the
biosynthesis of different furanocembranoids could be re-
lated to bipinnatin J (2) and should therefore be accessible
from this natural product within a short number of
steps including oxidations, rearrangements, and cyclo-
additions.5,6 In particular, it is proposed that epoxidation
of the Δ7,8 double bond of bipinnatin J (2) followed by the
(1) Marrero, J.; Rodrıguez, A. D.; Baran, P.; Raptis, R. G.; Sanchez,
J. A.; Ortega-Barria, E.; Capson, T. L. Org. Lett. 2004, 6, 1661–1664.
(2) (a) Farcet, J.-B.; Himmelbauer, M.; Mulzer, J. Org. Lett. 2012, 14,
2195–2197. (b) Farcet, J.-B.; Himmelbauer, M.; Mulzer, J. Eur. J. Org.
addition of water and a consecutive formal [2 þ 2]-cycload-
dition could lead to bielschowskysin (Scheme 1).
To date this biomimetic [2 þ 2]-photocycloaddition
Chem. 2013accepted for publication.
(3) (a) Doroh, B.; Sulikowski, G. A. Org. Lett. 2006, 8, 903–906.
(b) Miao, R.; Gramani, S. G.; Lear, M. J. Tetrahedron Lett. 2009, 50,
1731–1733. (c) Jana, A.; Mondal, S.; Md. Hossain, F.; Ghosh, S.
Tetrahedron Lett. 2012, 53, 6830–6833.
(4) (a) Nicolaou, K. C.; Adsool, V. A.; Hale, C. R. H. Angew. Chem.,
Int. Ed. 2011, 50, 5149–5152. (b) Meyer, M. E.; Phillips, J. H.; Ferreira,
E. M.; Stoltz, B. M. Tetrahedron 2013, 1–9.
strategy has been pursued by four groups. However, the
(5) (a) Roethle, P. A.; Trauner, D. Org. Lett. 2006, 8, 345–347.
(b) Roethle, P. A.; Hernandez, P. T.; Trauner, D. Org. Lett. 2006, 8,
5901–5904. (c) Roethle, P. A.; Trauner, D. Nat. Prod. Rep. 2008, 8, 298–317.
(6) Li, Y.; Pattenden, G. Nat. Prod. Rep. 2011, 28, 1269–1310.
r
10.1021/ol401285d
Published on Web 05/31/2013
2013 American Chemical Society

intermediates advanced by Sulikowski (3),3a Lear (4),3b
Nicolaou (5),4a and Gosh (6)3c are deficient in functiona-
lization and 3ꢀ5 lack the crucial all-carbon quaternary
center at C-12 (Scheme 1).
Scheme 3. Preparation of the Allene Building Block
Scheme 2. Retrosynthetic Analysis
Our retrosynthetic plan (Scheme 2) is centered around
key intermediate 8, which was to be assembled from
components 9 to 11. An allylation with 2-bromo-3-tri-
methylsilyl propene (9) should lead to vinyl bromide 8 as
the substrate of a palladium mediated Heck macrocycliza-
tion. Hopefully, this would furnish cyclononadiene 7
which might be carried on to the final target by allylic
oxidation and formation of the dihydrofuran ring.
Scheme 4. Preparation of the Coupling Partner
The synthesis of allene 10(Scheme 3) started withknown
alkyne 12, easily available from (ꢀ)-malic acid.3a,b,7 Con-
version to epoxide 14 was followed by regioselective ring
opening with diethylmalonate. In situ lactonization and
Krapcho decarboxylation8 gave butyrolactone 15 in 56%
9
ꢀ
yield from 12. The SearlesꢀCrabbe protocol was used
for generating the allene. Finally, deprotonation of the
lactone, treatment with chlorotrimethylsilane, and addi-
tion of phenylselenyl chloride furnished building block 10
as a 1:1.5 mixture of diastereomers in 83% yield.
Coupling partner 11(Scheme 4) was prepared from known
R-D-ribofuranose 1710 via lactone 18 (diastereomerically
pure). On subjecting the protected diol 19 to Swern oxidation
conditions, the primary triethylsilyl protecting group was
selectively removed and aldehyde 11 was obtained in 97%
yield.
Scheme 5. Fragment Coupling and [2 þ 2]-Photocycloaddition
Both building blocks 10 and 11 are readily available
in gram quantities and easy to couple by aldol addition.
Thus, deprotonation of the seleno lactone 10 at low
temperature followed by addition of aldehyde 11 resulted
in a mixture of all four diastereomeric adducts which was
used without separation in the next step (Scheme 5).
Regioselective oxidative elimination of the phenylsele-
nide gave an inseparable 1:1 mixture of diastereoisomers
(7) Saito, S.; Hasegawa, T.; Inaba, N.; Nishida, R.; Fujii, T.; Nomizu,
S.; Muriwake, T. Chem. Lett. 1984, 1389–1392.
(8) Krapcho, P. A.; Weimaster, J. F.; Eldridge, J. M.; Jahngen,
E. G. E., Jr.; Lovey, A. J.; Stephens, W. P. J. Org. Chem. 1978, 43,
138–147.
(20 and 21) which was irradiated in degassed cyclohexane
in quartz tubes with commercially available UV-C-lamps
in a homemade UV-reactor for 4 h to provide the tetra-
cyclic photoadducts 22 and 23 in 67% combined yield.
(9) Searles, S.; Li, Y.; Nassim, B.; Lopes, M.-T. R.; Tran, P. T.;
ꢀ
Crabbe, P. J. Chem. Soc., Perkin Trans. 1 1984, 747.
Org. Lett., Vol. 15, No. 12, 2013
3099

Additionally regioisomers 24 and 25, originating from the
cyclization of the terminal allenic double bond, were
isolated in a 15% combined yield. Flash column chroma-
tography atthisstage providedus withpureisomers 22and
23 for the envisaged Heck macrocyclization.
Scheme 7. Introduction of the Bromoallyl Appendage
Scheme 6. Preparation of Single Crystals for X-ray Analysis
For the assignment of the newly created stereocenters,
olefin 22 was epoxidized with a freshly prepared solution
of dimethyldioxirane with a d.r. of 6.7:1 (Scheme 6).
Standard acetylation provided 26, suitable for single crys-
tal diffraction.
The synthesis was separately carried on with diastereo-
merically pure 22 and 23. To unify the silyl protect-
ing groups, global deprotection with acetic acid in THF
was followed by treatment with chlorotriethylsilane to give
TES-derivative 27 and 28 in excellent yield (Scheme 7).
Again, under the Swern oxidation conditions the primary
silyl group was removed selectively. Gratifyingly, Baeyerꢀ
Villiger oxidation of the aldehyde with meta-chloroper-
benzoic acid in dichloromethane at 0 °C was much faster
than the epoxidation of the exo-methylene group so that
formates 29 and 30 were generated in fair yield. Lewis acid
mediated allylation with silane 9 gave trans-isomers 31 and
32 as single diastereomers, presumably via an oxonium
intermediate which was attacked from the less hindered
ring face.11,12 Obviously, the formate is such a superior
leaving group that the second anomeric center at C-16 is
not touched.
exo-pathway is precluded. A general mechanistic picture
(Scheme 8) suggests that the usual oxidative addition
generates σ-alkenylpalladium(II) complex II which adds
to the exo-methylene double bond in a exo-trig fashion
forming neopentylpalladium intermediate III. As β-hy-
dride elimination at this stage is impossible, a 3-exo-trig
ring closure to cyclopropane IV should occur (Scheme 8,
path A). Elimination of palladium would then give the
desired diene V in a formal overall endo-trig cyclization.
Scheme 8. Mechanistic Rationalization
With an appropriate bromoallyl appendage in place
we tackled the Heck macrocyclization. Although a pre-
ference for exo-mode Heck cyclizations exists,13 endo-type
reactions have also been observed,14,15 generally when the
(10) (a) Rosenthal, A.; Nguyen, L. B. J. Org. Chem. 1969, 34, 1029–
1034. (b) Xie, M.; Berges, D. A.; Robins, M. J. Org. Chem. 1996, 61,
5178–5179 and references cited therein.
(11) (a) Martinez, H. O.; Reinke, H.; Michalik, D.; Vogel, C. Synth-
esis 2009, 11, 1834–1840. (b) McDevitt, J.; Lansbury, P. T., Jr. J. Am.
Chem. Soc. 1996, 118, 3818–3828.
(12) Schmitt, A.; Reißig, H.-U. Eur. J. Org. Chem. 2000, 3893–3901.
(13) Overman, L. E. Pure Appl. Chem. 1994, 66, 1423–1430. Dounay,
A. B.; Overman, L. E. Chem. Rev. 2003, 103, 2945–2963. Schulman,
J, M.; Friedman, A. A.; Panteleev, J.; Lautens, M. Chem. Commun. 2012,
48, 55–57.
(14) de Meijere, A.; Meyer, F. E. Angew. Chem., Int. Ed. 1994, 33,
2379–2411.
(15) Rigby, J. H.; Hughes, R. C.; Heeg, M. J. J. Am. Chem. Soc. 1995,
117, 7834–7835 and references cited therein.
A wide variety of Heck conditions were applied to
precursors 31 and 32 (Table 1, Supporting Information).
3100
Org. Lett., Vol. 15, No. 12, 2013

For 31, a standard procedure16 led to a complex product
mixture, in which traces of the desired diene 33 were
detected by mass analysis (Scheme 9). Trying to improve
this result, we used the findings by Rigby et al.,15 who have
reported that electronic effects and a relatively small metal
coordination sphere of the palladium tend to favor the
endo-pathway in Heck cyclizations. On this basis, we
subjected 31 to Jeffery conditions.17,18 To our surprise,
this reaction stereoselectively led to product 34 in 55%
yield. Thus, the exo-methylene group has been attacked
from the less hindered face of the cage-shaped precursor
to form a tricyclo[8.3.0.02,9]tridecane ring system instead
of the desired “natural” tricyclo[9.3.0.02,10]tetradecane
framework (Scheme 9). The stereochemistry and connectivity
of 34 were determined by 2D NMR analysis (see Supporting
Information).
elimination leading to 34 (path B). Thus, an eight-mem-
beredmacrocycle (34) anda newlyformedcarbonꢀoxygen
bond were generated in a single step from vinyl bromide 31
in acceptable overall yield.
To our knowledge this reaction which converts a
1,1-disubstituted olefin into an allylic neopentyl acetate
so far has not been described in the literature.
In conclusion, we have developed a stereocontrolled route
to an advanced macrocyclization precursor 31 within a
longest linear sequence of 15 steps from the literature known
alkyne 12 with an overall yield of 13%. A biomimetic [2 þ 2]-
photocyclization was used to install the all-carbon quatern-
ary center at C-12. In this step the western [3.2.0]-carbon
core of bielschowskysin with all-carbon atoms of the cyclo-
butane moiety is set up correctly. Moreover, the stereocen-
ters at C-1 and C-2 have been introduced with acetal building
block 11, which could be a suitable building block for
other syntheses. The Heck macrocyclization of 31 revealed
an unprecedented carbo-oxygenation reaction of a vinyl
bromide onto a 1,1-disubstituted double bond. This led to
the complex macrocycle 34 featuring a tricyclo[8.3.0.02,8-
]tridecane ring system and an allylic neopentyl acetate. Work
to generalize this methodology is in progress.
Scheme 9. Macrocyclization and Carbo-oxygenation
Acknowledgment. Financial support from the Univer-
sity of Vienna (doctoral programm, Initiativkolleg Func-
tional Molecules IK I041-N) and from the Austrian
Science Fund (FWF, Project No. P22180) is gratefully
€
acknowledged. We thank H. Kahlig, L. Brecker, and S.
Felsinger for NMR assistance and A. Roller and V. Arion
(University of Vienna) for X-ray analysis.
Supporting Information Available. Experimental pro-
1
So, obviously unlike the carbohalogenations reported
by Lautens19 and Tong,20 acetoxy-palladation of I to VI
is followed by syn-addition and reductive β-bromide
cedures and full characterization including copies of H
and 13C NMR spectra and crystal structure analysis of 26
(CIF). This material is available free of charge via the
Internet at http://pubs.acs.org.
˚
(16) Pd(OAc)2 (0.1 equiv), PPh3 (0.2 equiv), Ag2CO3 (3.0 equiv), 4 A
MS, toluene (0.01 M), 80 °C, 3 d. For a detailed procedure, see
Supporting Information.
(17) (a) Jeffery, T. J. Chem. Soc., Chem. Commun. 1984, 1287–1289.
(b) Jeffery, T. Synthesis 1987, 70–71. (c) Jeffery, T. Tetrahedron 1996, 52,
10113–10130.
(19) (a) Newman, S. G.; Lautens, M. J. Am. Chem. Soc. 2011, 133,
1778–1780. (b) Newman, S. G.; Howell, J. K.; Nicolaus, N.; Lautens, M.
J. Am. Chem. Soc. 2011, 133, 14916–14919.
(20) Liu, H.; Li, C.; Qiu, D.; Tong, X. J. Am. Chem. Soc. 2011, 133,
6187–6193.
(18) Pd(OAc)2 (0.1 equiv), NaOAc (5.0 equiv), Bu4NCl (2.0 equiv),
4 A MS, DMF (0.01 M), 85 °C, 1.5 h. For a detailed procedure, see
Supporting Information.
˚
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 12, 2013
3101