Gram scale synthesis of the C(18)-C(34) fragment of amphidinolide C

Article abstract of DOI:10.1021/ol1030074

The synthesis of the C(18)-C(34) fragment of amphidinolide C has been achieved via two routes, culminating in both the shortest (11 steps) and highest yielding (26% overall yield) approaches to this segment. The highly convergent approach will facilitate

Full text of DOI:10.1021/ol1030074

ORGANIC
LETTERS
2011
Vol. 13, No. 4
572–575
Gram Scale Synthesis of the C(18)-C(34)
Fragment of Amphidinolide C
Nicholas A. Morra and Brian L. Pagenkopf*
Department of Chemistry, The University of Western Ontario, London, Ontario N6A
5B7, Canada
bpagenko@uwo.ca
Received November 8, 2010
ABSTRACT
The synthesis of the C(18)-C(34) fragment of amphidinolide C has been achieved via two routes, culminating in both the shortest (11 steps) and
highest yielding (26% overall yield) approaches to this segment. The highly convergent approach will facilitate the synthesis of analogues,
including the C(18)-C(29) fragment of amphidinolide F. Synthetic highlights include the selective methylation of a diyne, and the highly efficient
use of a second generation cobalt catalyst in the Mukaiyama oxidative cyclization to form the trans-THF ring.
Amphidinolide C belongs to a large family of macrolides
and linear polyketides isolated from the symbiotic marine
dinoflagellate Amphidinium sp by Kobayashi and co-
workers.1 Since their discovery, the amphidinolides have
attracted considerable synthetic attention due to their
structural diversity and range of biological activities. Am-
phidinolide C in particular exhibits potent cytotoxicity
against murine lymphoma and human epidermoid carci-
noma cells (5.8 and 4.6 ng/mL, respectively).2 To date,
amphidinolide C has not been synthesized in a laboratory,
although the syntheses of several fragments have been
reported.3 Interestingly, amphidinolide F, which is iden-
tical to amphidinolide C from C(1) through to C(28),
displays significantly lower activity (1.5 and 3.2 μg/mL,
respectively), which suggests that the side chain from C(25)
onward plays a crucial role in the bioactivity.2 As such, an
efficient gram scale synthesis of the C(18)-C(34) portion
of amphidinolide C (1) and related analogues is of great
importance.
To facilitate the synthesis of analogues, the initial retro-
synthetic disconnection of the C(18)-C(34) segment 1 was
envisioned to be a diastereoselective alkynylation of THF-
aldehyde 2 by alkyne 3 (Scheme 1). Aldehyde 2 can be
efficiently accessed via Mukaiyama aerobic oxidative cycli-
zation of an appropriately protected pentenol 4. Thekeystep
for formation of alkyne 3 was envisioned to be a selective
methylation of either diyne 5 or propargyl alcohol 6.
The synthesis began by the opening of known epoxide 8
with allyl Grignard to provide cyclization precursor 4 in
quantitative yield (Scheme 2).4 Our second generation
water-soluble cobalt catalyst Co(nmp)2 (7) provided
THF-alcohol 10 in 97% yield, which was a dramatic
(1) Kubota, T.; Tsuda, M.; Kobayashi, J. Org. Lett. 2001, 3, 1363–
1366.
(2) Kubota, T.; Kobayashi, J. J. Nat. Prod. 2007, 70, 451–460.
(3) (a) C(18)-C(34): Roy, S.; Spilling, C. D. Org. Lett. 2010, 12,
5326–5329. (b) C(1)-C(5): Paudyal, M. P.; Rath, N. P.; Spilling, C. D.
ꢀ
ꢀ
Org. Lett. 2010, 12, 2954–2957. (c) C(1)-C(9): Ferrie, L.; Figadere, B.
Org. Lett. 2010, 12, 4976–4979. (d) C(7)-C(20): Mahapatra, S.; Carter,
R. G. Org. Biomol. Chem. 2009, 7, 4582–4585. (e) C(18)-C(29):
Armstrong, A.; Pyrkotis, C. Tetrahedron Lett. 2009, 50, 3325–3328. (f)
C(1)-C(9): Bates, R. H.; Shotwell, J. B.; Roush, W. R. Org. Lett. 2008,
9, 4343–4346. (g) C(19)-C(34): Mohapatra, D. K.; Rahaman, H.;
Chorghade, M. S.; Gurjar, M. K. Synlett 2007, 567–570. (h) C(1)-C(10)
and C(17)-C(29): Kubota, T.; Tsuda, M.; Kobayashi, J. Tetrahedron
2003, 59, 1613–1625. (i) C(1)-C(9): Ishiyama, H.; Ishibashi, M.;
Kobayashi, J. Chem. Pharm. Bull. 1996, 44, 1819–1822. (j) C(11)-C(29):
Shotwell, J. B.; Roush, W. R. Org. Lett. 2004, 12, 3865–3868.
r
10.1021/ol1030074
Published on Web 01/21/2011
2011 American Chemical Society

Scheme 1. Retrosynthetic Analysis
Scheme 3. First Generation Synthesis of 3
sensitive vinyl stannane that was exchanged for an iodine
in situ to provide 14 in an excellent 97% yield.9 To the best
of our knowledge, this is the first reported example of a
selective stannylation/iodination sequenceona 5-hydroxy-
1,3-diyne, and this procedure provides an attractive alter-
native to accessing these types of highly unsaturated
systems.10 A surprisingly challenging TBS protection was
accomplished by using TBSOTf and triethylamine, which
was followed by a Stille cross coupling with tetramethyltin
to afford 15 in a modest 64% yield. The terminal alkyne
was revealed by cleavage of the triethylsilane moiety, using
K2CO3 in MeOH. This sequence provided access to the
desired fully functionalized alkyne 3 in 6 steps and 34%
yield from commercially available hexanal. The Trost
protocol to form 5 worked exceptionally well on small
scale, but difficulties with scalability and the prohibi-
tively high cost of dimethylzinc urged us to pursue a
route that was not reliant on asymmetric alkynylation
chemistry.
Our second generation route also started with 2-methyl-
enehexanal (13), which was elaborated through a three-
step procedure consisting of a racemic acetylide addition,
oxidation to the ketone, and subsequent CBS reduction
(Scheme 4). Alcohol 6 was obtained in 90% ee, even
while using a high catalyst loading of the CBS reagent
(10 mol %). This level of selectivity is relatively low
when compared to many other CBS reductions,11 but is
consistent with other reported asymmetric reductions of
propargyl ketones.12 After some protecting group manip-
ulations, alkyne 16 was converted to the propargylic ester
17, thereby setting the stage for installation of the required
methyl group by a copper(I)-catalyzed Michael addition to
give 18. The desired terminal alkyne 3 was obtained by
Scheme 2. Synthesis of THF-Aldehyde 2
improvement over the previous yields obtained with tradi-
tional oxidative cyclization catalysts.5 Lastly, Swern oxi-
dation provided THF-aldehyde 2 in an efficient three-step
procedure on multigram scale.
Our first generation synthesis of alkyne 3 began with an
asymmetric alkynylation of 2-methylenehexanal (13),
which was easily obtained in 94% yield from inexpensive
hexanal (Scheme 3). The Trost protocol (12, Me2Zn, 10
mol % 11) proved the most effective for this transforma-
tion, providing propargyl alcohol 5 in a respectable
90% ee.6-8 Treatment of 5 with tributyl tin hydride and
a substoichiometric amount of triethylborane resulted in a
hydroxyl-directed radical stannylation reaction, giving the
(4) Epoxide 8 can be accessed on multi-gram scale by using Jacobsen’s
hydrolytic kinetic resolution procedure: Tokunaga, M.; Larrow, J. F.;
Kakiuchi, F.; Jacobsen, E. N. Science 1997, 277, 936–938. See the
Supporting Information for details.
(5) (a) Palmer, C.; Morra, N. A.; Stevens, A. C.; Bajtos, B.; Machin,
B. P.; Pagenkopf, B. L. Org. Lett. 2009, 11, 5614–5617. (b) Wang, J.;
Morra, N. A.; Zhao, H.; Gorman, J. S. T.; Lynch, V.; McDonald, R.;
Reichwein, J. F.; Pagenkopf, B. L. Can. J. Chem. 2009, 47, 328–334.
(6) Trost, B. M.; Weiss, A. H.; Wangelin, A. K. J. Am. Chem. Soc.
2006, 128, 8–9.
(7) At the time of this work such asymmetric additions with diynes
had not been reported, but it has since been described: Trost, B. M.;
Chan, V. S.; Yamamoto, D. J. Am. Chem. Soc. 2010, 132, 5186–5192.
(8) Other methods explored in our laboratory to make 5 included
asymmetric reduction (CBS, alpine borane, Negishi) and asymmetric
alkynylation (Carreira).
(10) (a) Macbecin, I.; Belardi, J. K.; Micalizio, G. C. Angew. Chem.,
Int. Ed. 2008, 47, 4005–4008. (b) Archazolid, B.; Roethle, P. A.;
Chen, I. T.; Trauner, D. J. Am. Chem. Soc. 2007, 129, 8960–8961. (c)
Papulacandin, D.; Denmark, S. E.; Regens, C. S.; Kobayashi, T. J. Am.
Chem. Soc. 2007, 129, 2774–2776. (d) Iejimalide, B.; Fuerstner, A.;
Nevado, C.; Tremblay, M.; Chevrier, C.; Teply, F.; Aiessa, C.; Waser,
M. Angew. Chem., Int. Ed. 2006, 45, 5837–5842.
(9) Dimopoulos, P.; Athlan, A.; Manaviazar, S.; George, J.; Walters,
M.; Lazarides, L.; Aliev, A.; Hale, K. J. Org. Lett. 2005, 7, 5369–5372.
(11) Helal, C. J.; Corey, E. J. Angew. Chem., Int. Ed. 1998, 37, 1986–
2012.
Org. Lett., Vol. 13, No. 4, 2011
573

Table 1. Coupling of 2 and 3
Scheme 4. Second Generation Synthesis of Alkyne 3
entry
conditions
toluene, -78 °C
DME, -78 °C
THF, -78 °C
yield (%)a
anti:syn
1
2
3
4
5
6
7
8
9
95
83
1.5:1
2.5:1
3:1
87
ether, -78 °C
98
4:1
ether, 3 equiv of LiCl, -78 °C
toluene, Et2AlCl, -78 °C
ether, Ti(OiPr)3Cl, -78 °C
MTBE, -78 °C
86
5:1
68
1.5:1
2:1
72
92b
93b
8:1
MTBE, -90 °C
20:1
a 0.1 mmol scale. b 2.0 mmol scale.
conversion of ester 18 to the corresponding aldehyde,
followed by a Corey-Fuchs homologation of 19. This
second generation, multigram scale synthesis delivered
alkyne 3 in 46% yield over 11 steps from hexanal.
Scheme 5. Completion of the C(18)-C(34) Segment 1 of
Amphidinolide C
With a cost-effective and scalable route to alkyne 3 and
aldehyde 2, efforts were made to couple the two fragments
stereoselectively. Initially, it was envisioned that an asym-
metric method could be used to enhance the diasteroselec-
tivity of the addition, given our previous success with this
strategy.13 Unfortunately, after initial attempts proved
unsuccessful with both the Trost6 and Carreira14 alkynyla-
tion methods, we turned to traditional substrate controlled
diastereoselective additions. In this regard, a variety of
solvents, additives, and counterions were explored. In each
case, the desired syn diastereomer was never observed as
the major product. Also, attempts to oxidize the secondary
alcohol to the ketone and perform an asymmetric reduc-
tion resulted in poor dr values.15
Initial reactions in toluene, dimethoxyethane, and THF
(Table 1, entries 1-3) provided at best a 3:1 selectivity for
the anti diastereomer 20. Performing the reaction in diethyl
ether provided a modest increase in dr (Table 1, entry 4),
while adding G3 equiv of dry LiCl increased selectivity to
5:1 (Table 1, entry 5). Transmetalation of the acetylide to
the aluminum or titanium derivative has been shown
to increase dr in alkynylation reactions of this type;13
however, a drop in selectivity and yield was observed
(Table 1, entries 6 and 7). After several other conditions
were screened we were relieved to find that treatment of the
lithium acetylide of 3 with the THF-aldehyde 2 in dry
methyl tert-butyl ether (MTBE) resulted in a promising 8:1
dr. Ultimately, it was discovered that cooling the reaction
to-90°Cprior to aldehydeadditionresultedinanincrease
in selectivity to a 20:1 dr for 20 (Table 1, entries 8 and 9),
which proved reproducible on gram scale.
Inversion of alcohol 20 via Mitsunobu reaction (DIAD,
4-nitrobenzoic acid, PPh3) furnished the desired syn con-
figuration in 90% yield (Scheme 5). Finally, treatment of
21 with Red-Al provided the C(18)-C(34) segment of
amphidinolide C (1) on gramscaleasa singlediastereomer.
In summary, we have reported both the shortest (11
steps, 22% overall yield) and highest yielding (16 steps,
26% overall yield) stereoselective syntheses of the
C(18)-C(34) segment of amphidinolide C (1). The diver-
gent nature of the synthesis should facilitate the prepara-
tion of analogues, including amphidinolide F.
(12) (a) CBS: Murakami, Y.; Yoshida, M.; Shishido, K. Tetrahedron
Lett. 2009, 50, 1279–1281. (b) Alpine borane: Midland, M. M.;
Tramontano, A.; Kazubski, A.; Graham, R. S.; Tsai, D. J. S.; Cardin,
D. B. Tetrahedron 1984, 40, 1371–80. (c) Lipase reduction: Matsuda, M.;
Yamazaki, T.; Fuhshuku, K.; Sugai, T. Tetrahedron 2007, 63, 8752–
8760.
(13) (a) Zhao, H.; Gorman, J. S. T.; Pagenkopf, B. L. Org. Lett. 2006,
8, 4379–4382. (b) Wang, J.; Pagenkopf, B. L. Org. Lett. 2007, 9, 3703–
3706. (c) Krause, N.; Seebach, D. Chem. Ber. 1987, 120, 1845–1851.
(14) Anand, N. K.; Carreira, E. M. J. Am. Chem. Soc. 2001, 123, 687–
9688.
(15) The highest dr achieved via diastereoselective reduction was 2:1.
See the Supporting Information.
574
Org. Lett., Vol. 13, No. 4, 2011

Acknowledgment. We thank the University of Western
Ontario and the National Sciences and Engineering Research
Council of Canada (NSERC) for financial assistance. N.M.
(CGS-D3) thanks NSERC for a graduate fellowship. We
thank Dr. Mike Chong for use of his polarimeter.
Supporting Information Available. Experimental pro-
cedures and characterization of all new compounds
and copies of NMR spectra. This material is avail-
able free of charge via the Internet at http://pubs.acs.
org.
Org. Lett., Vol. 13, No. 4, 2011
575