Studies Directed toward the Total Synthesis of Azaspiracid: Stereoselective Construction of C1-C12, C13-C19, and C21-C25 Fragments

Article abstract of DOI:10.1021/ol006674w

equation presented The efficient entry to the C₁-C₁₂, C₁₃-C₁₉, and C₂₁-C₂₅ fragments of azaspiracid is outlined. The C₁-C₁₂ portion is constructed using a key asymmetric allenyl borane addition to the corresponding α,β-unsaturated aldehyde. The synthesis of the C₁₃-C₁₉ portion utilizes an Evans asymmetric alkylation followed by Sharpless asymmetric dihydroxylation. In addition, a novel solution to the mismatched effects of a neighboring chiral oxazolidinone during a Sharpless dihydroxylation is detailed.

Full text of DOI:10.1021/ol006674w

ORGANIC
LETTERS
2000
Vol. 2, No. 24
3913-3916
Studies Directed toward the Total
Synthesis of Azaspiracid:
Stereoselective Construction of C −C ,
1
12
C −C , and C −C Fragments
13
19
21
25
Rich G. Carter* and David J. Weldon
Department of Chemistry and Biochemistry, UniVersity of Mississippi,
UniVersity, Mississippi 38677
rgcarter@olemiss.edu
Received September 29, 2000
ABSTRACT
The efficient entry to the C −C , C −C , and C −C25 fragments of azaspiracid is outlined. The C −C12 portion is constructed using a key
1
12
13
19
21
1
asymmetric allenyl borane addition to the corresponding r,â-unsaturated aldehyde. The synthesis of the C −C19 portion utilizes an Evans
13
asymmetric alkylation followed by Sharpless asymmetric dihydroxylation. In addition, a novel solution to the mismatched effects of a neighboring
chiral oxazolidinone during a Sharpless dihydroxylation is detailed.
Azaspiracid (1) was recently isolated in Killary Harbor,
Ireland, from the mussel Mytilus edulis.1 Two structurally
similar derivatives, azaspiracid-2 (2) and azaspiracid-3 (3),
were later discovered in the Arramore Island region of
Donegal, Ireland.2 Azaspiracids 1-3 have been shown to
possess considerable toxicity in vitro with lethal doses in
mice of 0.2, 0.11, and 0.14 mg/kg, respectively. The relative
stereochemistries of 1-3 were determined by extensive NMR
studies, but their absolute stereochemistry is, as yet, unde-
termined (Figure 1).
in the water samples. The low amounts isolated, as well as
apparent seasonal occurrence, have hindered further inves-
tigation.
In addition, the origin of the azaspiracids remains a
mystery. These compounds are believed to be dinoflagellate
in nature, due to their highly oxygenated polyether structure;
however, none of the known phytoplanktons were observed
(1) (a) Satake, M.; Ofuji, K.; Naoki, H.; James, K. J.; Furey, A.;
McMahon, T.; Silke, J.; Y.; Yasumoto, T. J. Am. Chem. Soc. 1998, 120,
9967. (b) MacMahon, T.; Silke, J. Harmful Algae News 1996, 14, 2.
(2) Ofuji, K.; Satake, M.; McMahon, T.; Silke, J.; James, K. J.; Naoki,
H.; Oshima, Y.; Yasumoto, T. Nat. Toxins 1999, 7, 99.
Figure 1. Azaspiracid and structural derivatives.
10.1021/ol006674w CCC: $19.00 © 2000 American Chemical Society
Published on Web 11/07/2000

The structural architecture of spirocycle 1 possesses a
challenging array of features: 47 carbons, 20 stereocenters,
and 9 rings, including a unique azaspiro linkage fused to a
[3.3.1] bicyclic nonane ring system. The formidable structural
features of this molecule, in combination with the biological
toxicity, make azaspiracid (1) an attractive synthetic target.3
Other spirocyclic compounds, such as pinnatoxin A and
okadaic acid, have also garnered considerable synthetic
interest.4
Strategy. Because of the structural complexity and size
of azaspiracid, it is imperative that any synthetic approach
toward 1 be highly convergent. For this reason, 1 was broken
down into five relatively equal fragments A-E (Scheme 1).
tandem reaction sequence.5 The necessary C25,26 alkene will
be constructed using a Still-Gennari coupling6 between
aldehyde ABC and phosphonoester DE. The C20,21 bond will
be formed from a dithiane addition of C to the aldehyde
AB, setting the C20 stereochemistry. Finally, the C12,13 bond
should be available via a sulfone ester coupling between
fragments A and B followed by spirocyclization.
Fragment A. The critical C6 ether linkage, embedded in
the dihydropyran ring, represents the major challenge in the
construction of fragment A. This stereocenter was envisioned
to arise from a boron-mediated, asymmetric allenyl addition7
to the R,â-unsaturated aldehyde. Successful examples of
asymmetric propargyl or allenyl additions to unsaturated
aldehydes are relatively rare,8 in part to the decreased
reactivity imparted by the alkene.8a
The construction of fragment A was accomplished in nine
steps from commercially available 1,4-butanediol (4) (Scheme
2). Monoprotection of 4 as its TBDPS ether followed by
Scheme 1. Retrosynthetic Plan for Azaspiracid (1)
Scheme 2. Synthesis of the A Fragmenta
a (i) TBDPSCl, DMAP, Et3N, CH2Cl2; (ii) TPAP, NMO, mo-
lecular sieves, CH2Cl2; Ph3PdCHCO2Me, 78% (two steps); (iii)
DIBAL-H, CH2Cl2, -78 to 0 °C, 99%; (iv) TPAP, NMO, molecular
sieves, CH2Cl2, 89%; (v) triphenylpropargyl stannane 11, CH2Cl2,
-78 °C, 74%, 80% ee; (vi) TESOTf, 2,6-lutidine, CH2Cl2, -78
°C, 85%; (vii) n-BuLi, CeCl3 (1.3 equiv), THF; 12, 63%; (viii)
Lindlar’s catalyst, H2, quinoline, hexanes, 57%; (ix) Et3N‚HF,
MeOH, CH2Cl2, then PPTS, 64%.
TPAP oxidation and in situ Wittig olefination provided the
trans-ester 6 in 78% overall yield. DIBAL-H reduction of 6
with subsequent oxidation using TPAP proceeded without
incident in 89% yield. Incorporation of the propargyl
To construct the C24,25 anti-configuration, we envision an
oxidation of the aryl allylic selenide ABCDE with in situ
[2,3]-sigmatropic rearrangement. Our laboratory has recently
reported the first catalytic method for accomplishing this
Chemical Society: Washington, DC, 2000; ORGN-766. (c) Illig, A. M.;
Aiguade, J.; Forsyth, C. J. Abstracts of Papers, 219th National Meeting of
the American Chemical Society, San Francisco; American Chemical
Society: Washington, DC, 2000; CHED-373. (d) Forsyth, C. J.; Aiguade-
Bosch, J.; Dounay, A. B.; Hao, J. Abstracts of Papers, 218th National
Meeting of the American Chemical Society, New Orleans; American
Chemical Society: Washington, DC, 1999; ORGN-619.
(3) (a) Hao, J.; Aiguade, J.; Forsyth, C. Abstracts of Papers, 219th
National Meeting of the American Chemical Society, San Francisco;
American Chemical Society: Washington, DC, 2000; ORGN-767. (b)
Aiguade, J.; Hao, J.; Forsyth, C. J. Abstracts of Papers, 219th National
Meeting of the American Chemical Society, San Francisco; American
3914
Org. Lett., Vol. 2, No. 24, 2000

functionality using Corey’s bromoborane reagent 11 ef-
ficiently provided the desired alcohol 8 in 74% yield. The
enantiomeric excess (ee) of 80% as well as the absolute
configuration were determined via Mosher ester analysis.9
Formation of the dianion of 8 followed by addition of the
Weinreb amide 12 (prepared from the commercially available
acid chloride) resulted in significant amounts of elimination
of the â-phenyl sulfonyl group. For this reason, the 2° alcohol
function was masked using TESOTf to provide the silyl ether
9. Lithiation of the acetylene using n-BuLi, followed by
addition to the amide 12, in the presence of CeCl3, yielded
the desired acetylenic ketone 10 in 63% yield. Partial
reduction using Lindlar’s catalyst in the presence of quinoline
incorporated the necessary cis alkene in 57% yield. Finally,
after considerable experimentation, desilylation using Et3N‚
HF followed by in situ ketalization using PPTS provided
the desired fragment A in 64% yield.
1 g. Ultimately, the iodide 16 was effectively synthesized
using triphenylphosphine and iodide in the presence of a
slight excess of imidazole in 91% yield. Finally, the key
alkylation of the iodide 16 with the sodium enolate derived
from commercially available oxazolidinone 17 was success-
fully accomplished in 99% yield with excellent diastereo-
selectivity (ds > 20:1).
Our interest now shifted to the critical dihydroxylation
using the Sharpless asymmetric dihydroxylation methodology
(Table 1).11 The Sharpless mnemonic predicted that AD mix
Table 1. Selected Examples Sharpless Dihydroxylation
Selectivity Based on Carbonyl Substitutiona
Fragment B. Incorporation of the critical C14 stereocenter
represented the first major hurdle in the construction of
fragment B. Treatment of 13 with 2-methoxy-1,3-dioxolane
in the presence of zinc chloride gave the desired enal 14 in
61% yield (Scheme 3).10 Reduction of the aldehyde 14 using
Scheme 3. Construction of the C14 Stereocentera
a i) LiOBn, BnOH, THF, 0 °C, 90%, 85% recovered auxiliary. † The
d.s. was not determined due to impurities.
R should give the desired stereochemistry; however, it should
be noted that Brimble and co-workers have recently observed
a complete reversal in selectivity using the AD mixes on an
Evans alkylated product.12 Treatment of alkene 18 with AD
mix R*13 did induce dihydroxylation, with contaminant
cyclization, to provide the desired lactone 19 in a disap-
pointing 36% yield. The stereochemical outcome of the
dihydroxylation was secured via X-ray crystal analysis of
19. The yield of this transformation could be improved
dramatically by buffering the solution with three equivalents
of sodium bicarbonate; however, poor selectivity was
observed (2.5:1).14 One possible explanation for the poor
selectivity in the dihydroxylation could be a mismatched
interaction between the oxazolidinone and the chiral osmium
species. To investigate this hypothesis, the alkene 18 was
converted into its benzyl ester 20. Subsequent dihydroxyl-
ation, with in situ lactonization, indeed did lead to improved
selectivity (10:1) of 19 without a significant decrease in yield
(77%). The major diastereomer could be easily recrystallized
to greater than 20:1 selectivity in 68% isolated yield.
a (i) 2-methoxy-1,3-dioxolane, ZnCl2, CH2Cl2, 61%; (ii) DIBAL-
H, Et2O, -78 to 25 °C, 76%; (iii) Ph3P, I2, imidazole, MeCN, Et2O
(1:3), 91%; (iv) NaHMDS, THF, -78 °C; 16, -78 °C, 99%, >95%
ds.
DIBAL-H provided the allylic alcohol 15. The iodide 16 was
initially constructed via the mesylate (Ms2O, Et3N, 0 °C,
92%); however, conversion to halide using sodium iodide
in acetone proved to be capricious on amounts greater than
(4) For a recent review, see: Brimble, M. A.; Fare´s, F. A. Tetrahedron
1999, 55, 7661.
(5) Carter, R. G.; Bourland, T. C. J. Chem. Soc., Chem. Commun. 2000,
2031.
(6) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405.
(7) Corey, E. J.; Yu, C.-M.; Lee, D.-H. J. Am. Chem. Soc. 1990, 112,
878.
(8) (a) Keck, G. E.; Krishnamurthy, D.; Chen, X. Tetrahedron Lett. 1994,
35, 8323. (b) Marshall, J. A.; Grant, C. M. J. Org. Chem. 1999, 64, 8214.
(9) (a) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem.
Soc. 1991, 113, 4092. (b) Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc.
1973, 95, 512. (c) Sullivan, G. R.; Dale, J. A.; Mosher, H. S. J. Org. Chem.
1973, 38, 2143.
(11) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. ReV.
1994, 94, 2483.
(12) Allen, P. A.; Brimble, M. A.; Prabaharan, H. Synlett 1999, 295.
(13) AD mix R* ) [(DHQ)2PHAL (154 mg), K2OsO2‚2H2O (25.5 mg),
K2CO3 (2.90 g), K3Fe(CN)6 (7.00 g)] Commercially available AD mix R
proved to be slow and inefficient.
(14) The undesired diastereomer appears to be preferentially consumed
in the dihydroxylation in the absence of NaHCO3.
(15) Nicolaou, K. C.; Daines, R. A.; Chakraborty, T. K. J. Am. Chem.
Soc. 1987, 109, 2208-10.
(10) Jung, M. E.; Gardiner, J. M. J. Org. Chem. 1991, 56, 2614.
Org. Lett., Vol. 2, No. 24, 2000
3915

With a viable route to the lactone 19, the next challenge
was the construction of the bicyclic methoxy acetal B
(Scheme 4). This reorganization required the conversion of
Scheme 5. Synthesis of Fragment Ca
Scheme 4. Completion of the B Fragmenta
a (i) DIBAL-H, CH2Cl2, -60 °C, 95%; (ii) HS(CH2)3SH,
BF3‚Et2O, CH2Cl2, 72%; (iii) TESCl, Et3N, CH2Cl2, 84%.
steps from commercially available methyl (R)-(-)-3-hy-
droxy-2-methylpropionate), reduction to the lactol 24 was
accomplished using DIBAL-H in 95% yield. Subsequent
treatment of 24 with 1,3-propanedithiol in the presence of
BF3‚Et2O provided the thioacetal 25 in 72% yield. It should
be noted that less than 5% epimerization at C22 was observed
during this sequence. Finally, protection the 1° hydroxyl
function as its TES ether provided fragment C in 84% yield.
The stereoselective construction of the C1-C12, C13-C19,
and C21-C25 fragments of azaspiracid has been achieved in
nine, eight, and eight steps, respectively. The synthesis of
the C1-C12 portion utilizes the Corey allenyl borane meth-
odology as an effective method for asymmetric addition to
a functionalized, R,â-unsaturated aldehyde. The C13-C19
fragment is constructed using an Evans alkylation followed
by a Sharpless asymmetric dihydroxylation. A unique
solution for increased diastereoselectivity in a mismatched
system is reported. Our continued progress toward the total
synthesis of 1 will be reported in due course.
a (i) p-TsOH, MeOH, 64% (after three cycles); (ii) imidazole,
MeCN, 90 °C, 97%.
the γ-lactone into the desired ∆-lactone and formation of
the furan acetal via the C16 hydroxyl function. Initially,
treatment of acetal 19 under aqueous acidic conditions did
lead to removal of the C19 acetal; however, approximately
10% elimination of the â-hydroxyl function was also
observed. Fortunately, treatment of 19 under methanolic
acidic conditions led to conversion to desired furan acetal
21 in 64% yield after three cycles. The remaining mass
balance consisted of the anomeric methoxy furan 21a (12%)
and the desired bicycle B (7%) as well as the dimethoxy
acetal 22. These products were easily separable, and resub-
mission of 22 and 21a led to the same equilibrium mixture.
Finally, lactonization of 21 with imidazole in acetonitrile15
at reflux gave the desired fragment B in 97% yield. X-ray
crystal analysis of the bicycle B allowed for the conclusive
establishment of the C19 stereochemistry.
Acknowledgment. The authors thank the University of
Mississippi for their generous support of this work, Dr. Todd
D. Williams (University of Kansas) for mass spectral data,
and Dr. Walter E. Cleland, Jr. (University of Mississippi)
for X-ray crystal data.
Fragment C. The synthesis of the dithiane fragment C
was based on close analogy to literature precedent (Scheme
5).16 Starting from the known lactone 23 (available in five
(16) (a) Andrus, M. B.; Li, W.; Keyes, R. F. J. Org. Chem. 1997, 62,
5542. (b) Chen, S.-H.; Horvath, R. F.; Joglar, J.; Fisher, M. J.; Danishefsky,
S. J. J. Org. Chem. 1991, 56, 5834. (c) Mohr, P.; Waespe-Sarcevic, N.;
Tamm, C.; Gawronska, K.; Gawronski, J. K. HelV. Chim. Acta. 1983, 66,
2501. (d) Smith, A. B.; Maleczka, R. E.; Leazer, J. L.; Leahy, J. W.;
McCauley, J. A.; Condon, S. M. Tetrahedron Lett. 1994, 35, 4911. (e)
Kocienski, P. J.; Brown, R. C. D.; Pommier, A.; Procter, M.; Schmidt, B.
J. Chem. Soc., Perkin Trans. 1 1998, 9.
Supporting Information Available: Experimental pro-
cedures and spectral characterization for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL006674W
3916
Org. Lett., Vol. 2, No. 24, 2000