Progress toward the total synthesis of saudin: Development of a tandem Stille-oxa-electrocyclization reaction

Article abstract of DOI:10.1021/ol050705b

(Chemical Equation Presented) A diastereoselective tandem Stille-oxa-electrocyclization reaction provides access to the core of the diterpenoid natural product saudin. Additionally, this new reaction sequence was extended to the convergent preparation of

Full text of DOI:10.1021/ol050705b

ORGANIC
LETTERS
2005
Vol. 7, No. 12
2413-2416
Progress toward the Total Synthesis of
Saudin: Development of a Tandem
Stille-Oxa-Electrocyclization Reaction
Uttam K. Tambar, Taichi Kano, and Brian M. Stoltz*
The Arnold and Mabel Beckman Laboratories of Chemical Synthesis, DiVision of
Chemistry and Chemical Engineering, California Institute of Technology,
Pasadena, California 91125
stoltz@caltech.edu
Received April 1, 2005
ABSTRACT
A diastereoselective tandem Stille-oxa-electrocyclization reaction provides access to the core of the diterpenoid natural product saudin.
Additionally, this new reaction sequence was extended to the convergent preparation of related polycyclic pyran systems.
Diabetes mellitus, a group of diseases characterized by
hyperglycemia, affects nearly 18.2 million people (6.3% of
the population) and is the sixth leading cause of death in
the United States (over 200,000 deaths per year).1 The disease
is controllable in most patients using a regimen of diet,
insulin injections, and oral hypoglycemic agents.2 In 1985,
Mossa and Cassady disclosed the structure and biological
activity of the novel caged diterpenoid saudin (1).3 The
chemical structure of 1 was proved unambiguously by single-
crystal X-ray analysis to be that depicted in Figure 1.
In the 20 years since the isolation of saudin, a substantial
effort has been undertaken to complete its total synthesis.
Recently, this effort resulted in the elegant syntheses of (()-
saudin by Winkler4 and (-)-saudin by Boeckman.5 Our
choice of saudin as a target molecule was based on its potent
hypoglycemogenic bioactivity and unique structure. Ad-
ditionally, we viewed this highly oxygenated, caged natural
product as an ideal template for the discovery and develop-
ment of new chemical reactions.
Our retrosynthetic analysis of saudin is outlined in Scheme
1. The polycyclic structure of saudin (1) exhibits an
impressive array of functionality and stereochemistry that
includes eight oxygenated carbons, seven stereocenters (two
(1) Centers for Disease Control and Prevention. National Diabetes Fact
Sheet: General information and national estimates on Diabetes in the United
States, 2003, rev ed.; Atlanta, GA: U.S. Department of Health and Human
Services, Centers for Disease Control and Prevention, 2004.
(2) Davis, S. N.; Granner, D. K. Insulin, Oral Hypoglycemic Agents,
and the Pharmacology of the Endocrine Panceas. In Goodman & Gilman’s
The Pharmacological Basis of Therapeutics; Hardman, J. G., Limbard, L.
E., Eds.; McGraw-Hill: New York, 1996; pp 1487-1517.
(3) Mossa, J. S.; Cassady, J. M.; Antoun, M. D.; Byrn, S. R.; McKenzie,
A. T.; Kozlowski, J. F.; Main, P. J. Org. Chem. 1985, 50, 916-918.
(4) Winkler, J. D.; Doherty, E. M. J. Am. Chem. Soc. 1999, 121, 7425-
7426.
Figure 1. Structure of saudin (1).
Importantly, saudin was found to induce hypoglycemia in
mice, and therefore could be an appealing lead structure for
the development of new agents to treat diabetes.
(5) Boeckman, R. K.; Ferreira, M. R. R.; Mitchell, L. H.; Shao, P. J.
Am. Chem. Soc. 2002, 124, 190-191.
10.1021/ol050705b CCC: $30.25
© 2005 American Chemical Society
Published on Web 05/13/2005

Initial model studies with variants of enone 5 revealed the
inadequacy of several transition metal-mediated reactions,
including Sonagashira, Heck, and Suzuki couplings. None-
theless, we were able to couple vinyl stannane 5c with the
cis-vinyl iodide 810 under modified Stille conditions,11 which
yielded dienone 9 (Scheme 3). This result established that
Scheme 1
Scheme 3
of which are quaternary centers), two lactone rings, and a
3-substituted furan. Initial retrosynthetic disconnection of the
C(1) and C(7) acetals exposes carboxylic acid 2 (Scheme
1), which, upon cleavage of the C(4)-C(5) linkage and
removal of the C(16)-methyl in a retro three component
coupling, arises from lactone 3. Opening the pyran ring of 3
in a retro-oxa-electrocyclization provides dienone 4, a
substrate that is suited for disconnection across the C(16)-
C(5) linkage via a number of possible transition metal-
mediated coupling reactions (e.g., Stille, Suzuki, Sonogashira,
Heck) between enone 5 and furan 6.
We initiated our study of the synthesis of saudin by
preparing variants of enone 5 and furan 6, with the hope
that we would unite the two compounds through a transition
metal-catalyzed reaction. The preparation of enone 5a
proceeded via the Robinson annulation of tetronic acid 76
and methyl vinyl ketone (Scheme 2).7 This enone was then
the bicyclic enone core structure was stable at least under
Stille conditions. The oxa-electrocyclization of enone 9 was
attempted under several conditions without success (heat,
UV light, Lewis acids).
Although the electrocyclization of model substrate 9 was
unsuccessful, we decided to apply the Stille coupling strategy
to fully elaborated substrates en route to saudin. The other
component for the Stille reaction (i.e., vinyl iodide 6a) was
synthesized from furaldehyde 10 in a straightforward manner
(Scheme 4). Treatment of this aldehyde with ethynyl
Scheme 4
Scheme 2
Grignard produced propargyl alcohol 11. Although oxidation
of this alcohol failed under several conditions (Swern
oxidation, Ley oxidation, and chromium-based oxidations),
Dess-Martin periodinane12 cleanly provided the desired
ynone, which was then converted to vinyl iodide 6a by
treatment with LiI and AcOH in MeCN.13
A series of conditions were examined for the Stille
coupling of vinyl stannane 5c and vinyl iodide 6a, and no
product was observed with several common Pd sources,
additives, and solvents. We then employed the conditions
used in the model system to generate dienone 9 with the
anticipation that the desired Stille product 4 would be
cleanly converted to bromoenone 5b by exposure to Br2 and
Et3N.8 The resulting product was easily transformed under
Stille conditions to vinyl stannane 5c,9 which was a viable
intermediate for transition metal-mediated coupling.
(10) For the synthesis of 8, see: Piers, E.; Wong, T.; Coish, P. D.; Rogers,
C. Can. J. Chem. 1994, 72, 1816-1819.
(11) Bellina, F.; Carpita, A.; De Santis, M.; Rossi, R. Tetrahedron 1994,
50, 12029-12046. (b) For a general review on the Stille reaction, see:
Farina, V.; Krishnamurthy, V.; Scott, W. J. Org. React. 1997, 50, 1-652.
(12) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277-
7287.
(6) For the synthesis of 7, see: Knight, D. W.; Pattenden, G. J. Chem.
Soc., Perkin Trans. 1 1975, 635-640.
(7) Prisbylla, M. P.; Takabe, K.; White, J. D. J. Am. Chem. Soc. 1979,
101, 762-763.
(8) Johnson, C. R.; Kozak, J. J. Org. Chem. 1994, 59, 2910-2912.
(9) Azizian, H.; Eaborn, C.; Pidcock, A. J. Organomet. Chem. 1981,
215, 49-58.
(13) Ma, S.; Lu, X.; Li, Z. J. Org. Chem. 1992, 57, 709-713.
2414
Org. Lett., Vol. 7, No. 12, 2005

produced. To our pleasant surprise, however, the combination
of catalytic Pd(PPh3)4, CuI, and DMF with the exclusion of
light facilitated the coupling of 5c and 6 to yield the furan
appended tricycle 3sthe result of a tandem Stille-oxa-
electrocylization reaction (Scheme 5). Interestingly, the
Scheme 7
Scheme 5
with n-butyllithium followed by 3-furaldehyde yielded the
coupled alcohol, which was oxidized with Jones’ reagent to
produce ynone 18. Conversion of 18 to vinyl iodide 14 was
affected by treatment of the ynone with LiI and AcOH in
MeCN to generate the desired cis-vinyl iodide in good yield
and as a single isomer. The synthesis of vinyl iodide 15 is
also depicted in Scheme 8. Aldehyde 1915 was converted to
an alkynyl anion by the Corey-Fuchs procedure.16 Subsequent
quenching of the anion with Weinreb amide 2017 yielded
ynone 21, which was readily converted to vinyl iodide 15
as a single olefin isomer with LiI and AcOH. Vinyl iodide
16 was rapidly synthesized from 1-butyn-3-ol (22), which
was first coupled to 3-furaldehyde (Scheme 8). Oxidation
of the resulting propargylic alcohol furnished ynone 23,
which was treated with LiI and AcOH in MeCN to yield
the desired vinyl iodide 16.
presence of CuI and the absence of light were both essential
for the success of this transformation.14
Initially, our strategy for the synthesis of saudin called
for a diastereoselective conjugate addition of a carbon
nucleophile into enone 3 to access a C(5) substituted product
(12, Scheme 6). Alternatively, we hypothesized that an
Scheme 6
With these three new vinyl iodides in hand, the key Stille-
oxa-electrocyclization reactions were attempted. Under iden-
tical conditions, smooth coupling occurred between stannane
Scheme 8
appropriately substituted enone 13 could undergo a diaste-
reoselective conjugate reduction to furnish a similar inter-
mediate. In addition to providing a more convergent route
to saudin, this new strategy would also give us a better sense
of the generality of our tandem Stille-oxa-electrocyclization
methodology.
Initial exploration of the new strategy began with poly-
cycles 13a-c, which required the generation of three new
vinyl iodides (i.e., 14, 15, 16, Scheme 7).
The synthesis of vinyl iodide 14 commenced with the silyl
protection of alcohol 17 (Scheme 8). Subsequent treatment
(14) In the presence of light, trans-dienone i was formed, presumably
as a result of cis-trans isomerization of vinyl iodide 6a followed by Stille
coupling:
Org. Lett., Vol. 7, No. 12, 2005
2415

5c and vinyl iodides 14, 15, and 16 to form the desired
polycycles 13a, 13b, and 13c in 92%, 78%, and 88% yield,
respectively (Scheme 9). Products 13a and 13c were formed
as single diastereomers, whereas 13b was produced as a 1:1
mixture of diastereomers at C(4).
Scheme 9
Figure 2. X-ray crystal structure of pyran intermediate 13c.
ates en route to the natural product.18 Current efforts are
focused on expanding the substrate scope of this tandem
reaction sequence, as well as advancing the aforementioned
pyran intermediates (3, 13a, 13b, and 13c) to saudin.
Acknowledgment. We are grateful to the California
Institute of Technology, NSF-PECASE, and NDSEG (gradu-
ate fellowship to U.K.T.) for generous financial support and
John F. Zepernick for experimental assistance. We thank Mr.
Larry M. Henling and Dr. Michael W. Day for X-ray
crystallographic expertise.
The bond connectivity and relative stereochemistry in 13c
were unambiguously confirmed by single-crystal X-ray
diffraction of the diketone (Figure 2).
In summary, we have developed a tandem Stille-oxa-
electrocyclization reaction that delivers the polycyclic pyran
core of the diterpenoid saudin in a convergent and rapid
fashion. We have also demonstrated the versatility of this
methodology toward the preparation of related polycyclic
pyran systems that may serve as useful synthetic intermedi-
Supporting Information Available: Experimental details
and characterization data for all new compounds including
X-ray data for 13c. This material is available free of charge
via the Internet at http://pubs.acs.org.
OL050705B
(18) Toward this end, we have encountered some level of difficulty
implementing the three component coupling chemistry outlined in Scheme
1. In particular C-C bond formation at the sterically congested C(16) center
has been difficult. We are currently developing catalytic asymmetric methods
to synthesize hindered quaternary carbon centers, and plan to apply these
strategies to override any inherent steric and stereochemical bias of the
pyran systems prepared in this study. For an example, see: Behenna, D.
C.; Stoltz, B. M. J. Am. Chem. Soc. 2004, 126, 15044-15045.
(15) For the synthesis of 19, see: Kiyooka, S.; Shahid, K. A.; Goto, F.;
Okazaki, M.; Shuto, Y. J. Org. Chem. 2003, 68, 7967-7978.
(16) Panek, J. S.; Hu, T. J. Org. Chem. 1997, 62, 4912-4913.
(17) For the synthesis of 20, see: Kinoshita, T.; Ichinari, D.; Sinya, J.
J. Heterocycl. Chem. 1996, 33, 1313-1317.
2416
Org. Lett., Vol. 7, No. 12, 2005