Total synthesis of (-)-spinosyn A: Examination of structural features that govern the stereoselectivity of the key transannular Diels-Alder reaction

Article abstract of DOI:10.1021/jo7024515

(Chemical Equation Presented) A study of elements of stereochemical control in transannular Diels-Alder reactions leading to the decahydro-as-indacene core of (-)-spinosyn A is described. Initial studies focused on macrocyclic pentaene 9, which includes C

Full text of DOI:10.1021/jo7024515

Total Synthesis of (-)-Spinosyn A: Examination of Structural
Features That Govern the Stereoselectivity of the Key Transannular
Diels-Alder Reaction
SusAnn M. Winbush,† Dustin J. Mergott,‡,§ and William R. Roush*,†
Department of Chemistry, Scripps Florida, Jupiter, Florida 33458, and Department of Chemistry,
UniVersity of Michigan, Ann Arbor, Michigan 48105
roush@scripps.edu
ReceiVed NoVember 15, 2007
A study of elements of stereochemical control in transannular Diels-Alder reactions leading to the
decahydro-as-indacene core of (-)-spinosyn A is described. Initial studies focused on macrocyclic pentaene
9, which includes C(6)-Br and C(8)-OTBS substituents. Excellent selectivity (>95:5) was observed in
the cycloaddition of 9 as a consequence of 1,3-allylic strain interactions involving the C(6) and C(8)
substituents in the disfavored TS-2. The major cycloadduct 22 was used in a formal synthesis of (-)-
spinosyn A. The TDA cyclizations of 12 (which lacks the C(8)-OTBS unit of 9), 13 (which lacks the
C(6)-Br substituent of 12), and 14 (which lacks the C(6)-Br and C(21)-Et substituents of 12) were also
studied. Macrocycles 12 and 13 served as precursors to (-)-spinosyn A and the (-)-spinosyn A aglycon
(34), respectively. It is striking that substrates 12-14 give very similar distributions of transannular
Diels-Alder cycloadducts, indicating that the C(6)-Br and C(21)-stereocenter do not play a significant
role in the diastereoselectivity of the TDA cycloaddition of spinosyn A precursor 12. It is likely that
some as yet unidentified conformational or structural features of macrocycles 12-14 contribute to the
levels of diastereoselectivity achieved, since these TDA reactions are more selective for the C(7)-C(9)
stereochemical relationship found in the natural product than are the IMDA reactions of trienes 4 and 7.
Introduction
Our strategy for the synthesis of 1, inspired by the proposed
biogenesis5,6 summarized in Scheme 1, involves the transannular
Diels-Alder (TDA)8,9 cycloaddition of an intermediate of type
3 followed by a nucleophile-induced Michael-type7 ring closure
of macrocycle 2. Previous studies of the transannular variant
Spinosyn A (1), a polyketide natural product possessing potent
insecticidal activity, is the major component of a biosynthetic
mixture generated by the soil microbe Saccharopolyspora
spinosa.1a-c The natural product mixture is currently marketed
by Dow AgroSciences as an agricultural insecticide against a
variety of insects.1c-e Among many attractive features of this
highly active class of insecticides is their low toxicity to
beneficial insects and rapid degradation in the environment.1f
Total syntheses of spinosyn A have been reported by the
laboratories of Evans2 and Paquette,3 in addition to our own.4
(1) (a) Kirst, H. A.; Michel, K. H.; Martin, J. W.; Creemer, L. C.; Chio,
E. H.; Yao, R. C.; Nakatsukasa, W. M.; Boeck, L.; Occolowitz, J. L.;
Paschal, J. W.; Deeter, J. B.; Jones, N. D.; Thompson, G. D. Tetrahedron
Lett. 1991, 31, 4839. (b) Kirst, H. A.; Michel, K. H.; Mynderase, J. S.;
Chio, E. H.; Yao, R. C.; Nakatsukasa, W. M.; Boeck, L. D.; Occolowitz,
J. L.; Paschal, J. W.; Deeter, J. B.; Thompson, G. D. In Fermentation-
DeriVed Tetracyclic Macrolides 1992; pp 214-225. (c) Creemer, L. C.;
Kirst, H. A.; Paschal, J. W. J. Antibiot. 1998, 51, 795. (d) Sparks, T. C.;
Crouse, G. D.; Durst, G. Pest. Manage. Sci. 2001, 57, 896. (e) Dagani, R.
Chem. Eng. News 1999, 77 (July 5), 30. (f) Tomkins, A. R.; Holland, P.
T.; Thomson, C.; Wilson, D. J.; Malcolm, C. P. Proc. 52nd N. A Plant
Protection Conf. 1999, 94-97.
† Scripps Florida.
‡ University of Michigan.
§ Present address: Lilly Research Laboratories, Indianapolis, Indiana 46285.
10.1021/jo7024515 CCC: $40.75 © 2008 American Chemical Society
Published on Web 01/24/2008
1818
J. Org. Chem. 2008, 73, 1818-1829

Total Synthesis of (-)-Spinosyn A
SCHEME 1. Proposed Biogenesis of (-)-Spinosyn A
SCHEME 2. IMDA Reaction of 4 (work of Evans2)
SCHEME 3. IMDA Cycloaddition of 715
of the Diels-Alder reaction in our laboratory have demonstrated
several cases of enhanced stereoselectivity compared to analo-
gous intramolecular cycloadditions.10,11 In these cases, the
manner in which the diene and dienophile approach each other
is restricted by conformational constraints which contributes to
the level of diastereofacial selectivity observed.8 At the outset
of the (-)-spinosyn A synthesis, we hoped that the TDA reaction
of 3, or its synthetic equivalents, would similarly display
enhanced diastereoselectivity compared to traditional intramo-
lecular Diels-Alder substrates.
Initial design and selection of the substrate(s) for our synthetic
studies were influenced by the work of Evans2 (Scheme 2). Data
reported for the Me2AlCl-promoted intramolecular Diels-Alder
(IMDA)12 reaction of 4 indicated that a 6:1 mixture of 5 and 6
was obtained, favoring 5 with the incorrect diastereomeric
relationship between the C(9)-alkoxy group relative to the C(7)-
C(11) ring fusion, opposite to what is needed for the indacene
core of spinosyn A.
While our premise was that conformational effects would
contribute to a synthetically useful transannular cycloaddition
of intermediates patterned after 3, it was not obvious that these
effects would counteract the stereochemical influence of the
C(9)-substituent observed in the IMDA reaction of 4. We chose,
therefore, to employ a steric directing group strategy,13,14
originally developed in our laboratory to effect stereochemical
control in the intramolecular Diels-Alder reaction involving
placement of a temporary substituent at C(6) of the diene so as
to introduce nonbonded interactions in transition states we
wished to disfavor.14 On the basis of a series of preliminary
studies, it was apparent that incorporation of a bromine atom
at the C(6) position of tetraene 7 partially offset the intrinsic
stereochemical directing influence offered by the resident C(9)-
alkoxy function (Scheme 3). The stereoselectivity of the thermal
IMDA cycloaddition of 7 was only 53:37:5:5 (ca. 1.5:1 dr with
respect to the two most abundant diastereomers), whereas
selectivities were only slightly better under Lewis acid promoted
conditions (62:30:5:3 dr).15 Therefore, we decided to introduce
a second removable substituent at the C(8) position as illustrated
in macrocycle 9 (Scheme 4). We anticipated this additional
directing unit would introduce an allylic 1,3-strain16 interaction
(2) Evans, D. A.; Black, W. C. J. Am. Chem. Soc. 1993, 115, 4497.
(3) Paquette, L. A.; Gao, Z.; Ni, Z.; Smith, G. F. J. Am. Chem. Soc.
1998, 120, 2543.
(4) Mergott, D. J.; Frank. S. A.; Roush, W. R. Proc. Natl. Acad. Sci.
U.S.A. 2004, 11955.
(5) (a) Kirst, H. A.; Michel, K. H.; Mynderse, J. S.; Chio, E. H.; Yao,
R. C.; Nakatsukasa, W. M.; Boeck, L. D.; Occolowitz, J. L.; Paschal, J.
W.; Deeter, J. B.; Thompson, G. D. In Synthesis and Chemistry of
Agrochemicals III; Baker, D. R., Fenyes, J. G., Steffens, J. J., Eds.; ACS
Symp. Ser, No. 504; American Chemical Society: Washington, DC, 1992.
(b) For a recent study on the biosynthesis of the spinosyns: Kim, H. J.;
Pongdee, R.; Wu, Q.; Hong, L.; Liu, H.-w. J. Am. Chem. Soc. 2007, 129,
14582.
(6) Review of Diels-Alder reactions in biosynthesis: (a) Stockings, E.
M.; Williams, R. M. Angew. Chem., Int. Ed. 2003, 42, 3078. (b) Oikawa,
H.; Tokiwano, T. Nat. Prod. Rep. 2004, 21, 321.
(7) Little, R. D.; Masjedizadeh, M. R.; Wllquist, O.; McLoughlin, J. I.
Org. React. 1995, 47, 315.
(8) Review of transannular Diels-Alder reactions in total synthesis:
Marsault, E.; Toro´, A.; Nowak, P.; Deslongchamps, P. Tetrahedron 2001,
57, 4243.
(9) Review of Diels-Alder reactions in natural product synthesis: (a)
Nicolaou, K. C.; Synder, S. A.; Montagnon, T.; Vassilikogiannakis, G.
Angew. Chem., Int. Ed. 2002, 41, 1668. (b) Winkler, J. D. Chem. ReV. 1996,
96, 167.
(10) Roush, W. R.; Koyama, K.; Curtin, M. L.; Moriarty, K. J. J. Am.
Chem. Soc. 1996, 118, 7502.
(12) Roush, W. R. Intramolecular Diels-Alder reactions. Comp. Org.
Synth. 1991, 5, 513.
(13) Boeckman, R. K., Jr.; Barta, T. E. J. Org. Chem. 1985, 50, 3423.
(14) (a) Roush, W. R.; Kageyama, M. Tetrahedron Lett. 1985, 26, 4327.
(b) Roush, W. R.; Kageyama, M.; Riva, R.; Brown, B. B.; Warmus, J. S.;
Moriarty, K. J. J. Org. Chem. 1991, 56, 1192.
(11) Tortosa, M.; Yahekis, N. A.; Roush, W. R. J. Am. Chem. Soc.
Submitted for publication.
(15) Mergott, D. J.; Frank, S. A.; Roush, W. R. Org. Lett. 2002, 4, 3157.
J. Org. Chem, Vol. 73, No. 5, 2008 1819

Winbush et al.
SCHEME 4. Transition States for TDA Cycloaddition of 9
FIGURE 1. Macrocycles studied in this work.
SCHEME 5. Strategy for Synthesis of (-)-Spinosyn A (1)
via 9
with the C(6)-Br atom, thus further destabilizing transition state
2 (TS-2) relative to transition state 1 (TS-1).15
During the course of this work, we synthesized and studied
the transannular Diels-Alder reactions of macrocycles 9 and
12 (the latter is a key intermediate in our published total
synthesis of 1).4 Macrocycles 13 and 14 were also constructed
in an effort to elucidate the factors that control the diastereo-
selectivity of this pivotal transannular Diels-Alder reaction
(Figure 1). The results of these studies are described herein.
Results and Discussion
Our initial approach to the synthesis of (-)-spinosyn A (1)
targeted macrocycle 9, with two directing groups, as the key
cyclization substrate.17 We envisaged that macrocycle 9 could
be accessed through the Horner-Wadsworth-Emmons (HWE)18
coupling of 15 and 16 (Scheme 5).
Assembly of macrocycle 9 commenced with the coupling of
aldehyde 15 (which was prepared by Swern oxidation of the
known allylic alcohol precursorssee the Supporting Information
for details)15,19 and known â-ketophosphonate 16 (Scheme 6).4
Treatment of phosphonate 16 with activated barium hydroxide20
followed by addition of aldehyde 15 afforded polyene 17 in
94% isolated yield. Treatment of triene 17 with aqueous acetic
acid effected deprotection of the TES ether.21,22 Acylation of
the resulting alcohol with diethyl phosphonoacetic acid (18)
provided phosphonate 19 in 81% yield. Suzuki23,24 coupling of
19 with vinyl boronic acid 2025 then afforded the corresponding
allylic alcohol in 63% yield. Subsequent oxidation of the primary
alcohol by using the Parikh-Doering26 protocol provided
aldehyde 21, which was used directly in the next step without
purification.
Treatment of 21 with lithium chloride27 and diisopropylethy-
lamine effected a tandem macrocyclization28-30 and transannular
Diels-Alder reaction, in which the undetected macrocycle 9
directly converts to cycloadduct 22 in 78% yield with >95:5
diastereoselectivity (Scheme 7). The stereochemistry of the
adduct 22 was assigned based on 2D NOESY and coupling
constant analysis. Key NOE cross-peaks observed in the
NOESY spectrum include those between H(12) and both H(4)
and H(7), which indicates these three protons reside on the same
face of tricycle 22. The large J values observed between H(8),
H(7), and H(11) are consistent with a C(7)-C(11) trans-ring
fusion, as are the NOE interactions observed for H(8) and
H(11).31
(16) Hoffman, R. W. Chem. ReV. 1989, 89, 1841.
(17) Mergott, D. J., 2004, Ph.D. Thesis, University of Michigan, Ann
Arbor, MI.
(18) Wadsworth, W. S., Jr.; Emmons, W. D. J. Am. Chem. Soc. 1961,
83, 1733.
(19) Frank, S. A.; Roush, W. R. J. Org. Chem. 2002, 67, 4316.
(20) Paterson, I.; Yeung, K.-S.; Smaill, J. B. Synlett 1993, 774.
(21) Hart, T. W.; Metcalfe, D. A.; Scheinmann, F. J. Chem. Soc., Chem.
Commun. 1979, 156.
(22) Nelson, T. D.; Crouch, R. D. Synthesis 1996, 1031.
(23) Miyaura, N.; Suzuki, A. J. Chem. Soc., Chem. Commun. 1979, 866.
(24) Frank, S. A.; Chen, H.; Kunz, R. K.; Schnaderbeck, M. J.; Roush,
W. R. Org. Lett. 2000, 2, 2691.
(25) Roush, W. R.; Champoux, J. A.; Peterson, B. C. Tetrahedron Lett.
1996, 37, 8989.
(26) Parikh, J. R.; Doering, W. v. E. J. Am. Chem. Soc. 1967, 89, 5505.
(27) Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfeld, A. P.;
Masamune, S.; Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25, 2183.
(28) Burri, K. F.; Cardone, R. A.; Chen, W. Y.; Rosen, P. J. Am. Chem.
Soc. 1978, 100, 7069.
(29) Stork, G.; Nakamura, E. J. Org. Chem. 1979, 44, 4010.
(30) Nicolaou, K. C.; Seitz, S. P.; Pavia, M. R.; Petasis, N. A. J. Org.
Chem. 1979, 44, 4011.
1820 J. Org. Chem., Vol. 73, No. 5, 2008

Total Synthesis of (-)-Spinosyn A
SCHEME 6. Synthesis of Macrocyclization Precursor 21
SCHEME 8. Vinylogous Morita-Baylis-Hillman
Cyclization of 22
TABLE 1. Vinylogous Morita-Baylis-Hillman Reaction of 22
SCHEME 7. Transannular Diels-Alder Reaction of 9
entry
PMe3 (equiv)
concn (M)
ratio (23:24:25)
yield (%)
1a
2b
3c
4d
5e
1.5
1.5
0.6
8.0
8.0
0.05
0.05
0.05
0.005
0.005
79:15:6
58:38:4
51:45:4
91:4:5
91
95
96
90
80
81:11:8
a Conducted on 0.001 mmol scale. b Conducted on 0.024 mmol scale.
c Conducted on 0.052 mmol scale. d Conducted on 0.002 mmol scale.
e Conducted on 0.16 mmol scale.
tetracycle 23 as the major component of a 79:15:6 product
mixture, along with olefin migration product 24 (whose olefin
geometry was not rigorously assigned) and C(3)-diastereomer
25 (Scheme 8). Subsequent experiments with 22, conducted on
a larger scale, however, afforded the desired tetracycle 23 in a
58:38:4 ratio along with 24 and 25 (entry 2, Table 1).
Based on the assumption that the olefin migration is initiated
by deprotonation of the allylic C(4)-H of 22, attempts were made
to minimize formation of 24 by modification of the cyclization
conditions. While it seemed unlikely that trimethylphosphine
(pKa ) 9)35 could be responsible for promoting the olefin
migration via direct deprotonation of the C(4)-H, it was found
that the amount of 24 increased when Me3P (0.6 equiv) was
added portion-wise to a solution of 22 in tert-amyl alcohol (entry
3, Table 1).
Ketone and ester enolates are intermediates along the viny-
logous Morita-Baylis-Hillman reaction pathway (Scheme 9).
In the presence of an alcoholic solvent, such as tert-amyl
alcohol, it is probable that some alkoxide ion is also generated
by protonation of the enolates (see 27 to 28).36 This alkoxide
should be capable of initiating olefin migration via deprotonation
1
(31) Pretsch, E.; Bu¨hlmann, P.; Affolter, C. H-NMR Spectroscopy. In
Structure Determination of Organic Compounds, 3rd ed.; Springer: New
York, 2000.
(32) Basavaiah, D.; Rao, P. D.; Hyma, R. S. Tetrahedron 1996, 52, 8001.
(33) Wang, L. C.; Luis, A. L.; Agaplou, K.; Jang, H. Y.; Krische, M. J.
J. Am. Chem. Soc. 2002, 124, 2402.
(34) Frank, S. A.; Mergott, D. J.; Roush, W. R. J. Am. Chem. Soc. 2002,
124, 2404.
(35) Henderson, Wm. A.; Streuli, C. A. J. Am. Chem. Soc. 1960, 82,
5791.
We developed the vinylogous Morita-Baylis-Hillman
(MBH)32-34 reaction specifically for use in constructing the
C(3)-C(14) bond of tetracycle 23. Treatment of transannular
cycloadduct 22 with Me3P in tert-amyl alcohol provided
J. Org. Chem, Vol. 73, No. 5, 2008 1821

Winbush et al.
SCHEME 9. Reaction Pathway for the Vinylogous MBH Reaction of 22
SCHEME 10. Synthesis of the Spinosyn A Pseudoglycon
(31): Formal Synthesis of (-)-Spinosyn A
of the C(4)-H of 22. As this deprotonation is an intermolecular
process, it was reasoned that carrying out the reaction under
high dilution conditions would suppress the C(2)-C(3) olefin
migration. Indeed, when 22 was treated with Me3P (8 equiv
added in 3 portions) in tert-amyl alcohol at 0.005 M (a 10-fold
dilution over previous experiments) the desired MBH product
23 was obtained in 90% yield as a 91:4:5 mixture with
diastereomer 25 and olefin migration product 24 (entry 4, Table
1). When this experiment was performed on a larger scale (entry
5) the ratio dropped somewhat (81:11:8); however, 23 was still
obtained in 80% yield. It is conceivable that the large excess
of Me3P used in the experiments leads to very high conversion
of 22 to 26, thereby minimizing the amount of 22 available for
the alkoxide-induced deconjugation reaction.
With the steric directing groups having served their purpose
in inducing excellent selectivity in the transannular cycloaddition
of 9, it was necessary to excise both units for elaboration of 23
to the natural product. Deprotection of the TBS ether was
accomplished in 92% yield by treatment of tetracycle 23 with
Et3N‚3HF.37 Acylation of the resultant alcohol with thiocarbo-
nyldiimidazole gave thioester 29 in quantitative yield.38 Expo-
sure of 29 to (TMS)3SiH39,40 and AIBN at 80 °C effected clean
reductive removal of the thiocarbonyloxy and bromine substit-
uents, providing tetracycle 30 in 68% yield. Finally, removal
of the PMB group from 30 with DDQ afforded the spinosyn A
pseudoaglycon 31 quantitatively (Scheme 10). This material was
identical in all respects to natural samples generously provided
by both Paquette and Kirst. Synthetic 31 also matched material
synthesized in our laboratory from 33, which was subsequently
converted to (-)-spinosyn A (Scheme 11).4
Diels-Alder reaction of macrocycle 12 provided a 73:12:9:6
mixture of four cycloadducts, from which tricycle 33 was
isolated as the major diastereomer. Given the selectivity of the
conversion of 12 to 33 is lower than that for the conversion of
9 to 22, it is apparent that the C(8)-OTBS unit of 9 plays a
beneficial role in enhancing the stereoselectivity of the TDA
reaction of 9 (>95:5 dr). Nevertheless, the selectivity of the 12
to 33 cycloaddition (ca. 6:1 dr) is synthetically useful (as
evidenced by our use of 33 in our total synthesis of (-)-spinosyn
A),4 and the synthesis of 12 is several steps shorter than the
synthesis of 9. The conversion of 12 to 33 is also more selective
than the ca. 1.5-2:1 selectivity achieved in the IMDA reaction
of 7 (Scheme 3).4,15
It is instructive to compare the stereoselectivity of the
transannular Diels-Alder reaction of 9 (Scheme 7) with that
of our previously published spinosyn A precursor 12 (Scheme
11).4 The tandem HWE cyclization of 32 and transannular
(36) Stewart, I. C.; Bergman, R. G.; Toste, F. D. J. Am. Chem. Soc.
2003, 125, 8696.
(37) McClinton, M. A. Aldrichim. Acta 1995, 28, 31.
(38) Hartwig, W. Tetrahedron 1983, 39, 2609.
(39) Ballestri, M.; Chatgilialoglu, C.; Clark, K. B.; Griller, D.; Giese,
B.; Kopping, B. J. Org. Chem. 1991, 56, 678.
(40) Chargilialoglu, C. Acc. Chem. Res. 1992, 25, 188.
The fact that the transannular cycloaddition of 12 (Scheme
11) is more selective than the IMDA reaction of 7 (Scheme 3)
suggests that some additional factors contribute to stereochem-
1822 J. Org. Chem., Vol. 73, No. 5, 2008

Total Synthesis of (-)-Spinosyn A
SCHEME 11. Transannular Diels-Alder Reaction of 124
SCHEME 13. Synthesis of Aldehyde 35
SCHEME 14. Synthesis of Macrocyclization Precursor 43
SCHEME 12. Strategy for Synthesis of (-)-Spinosyn A
Aglycon 34 via 13
in the macrocycle. Based on the expectation that the C(6)-Br
group was playing a beneficial role in TS-4, we surmised that
the conformational preferences induced by C(21) served to
reinforce the (assumed) favorable effects of the C(6)-Br in TS-4
leading to enhanced selectivity compared to the model intramo-
lecular Diels-Alder substrate 7 (Scheme 3).4,15
Based on this analysis, we elected to study the transannular
cycloaddition of a substrate lacking any stereochemical directing
groups. Accordingly, we targeted the synthesis of (-)-spinosyn
A aglycon 34 by way of macrocyclic pentaene 13. Our strategy
to prepare macrocycle 13 closely mimicked that used to access
9 and 12, with appropriate modifications of precursor 35
(Scheme 12).
ical control of the transannular event. We speculated the C(21)-
stereocenter, adjacent to the macrolactone carbonyl, might be
a conformational control element capable of enhancing the
diastereoselectivity of the TDA reaction (Scheme 11).4 In the
energetically preferred conformation of the macrolactone link-
age, C(21)-H eclipses the C(1)-carbonyl group, allowing the
ethyl chain to extend into a sterically unencumbered region.16
The C(21)-stereocenter thus dictates the face of the diene that
the dienophile may approach, by serving as a “turning point”
The synthesis of vinyl iodide 35 began with enantiomerically
enriched alcohol 36,41 which is readily available in three steps
J. Org. Chem, Vol. 73, No. 5, 2008 1823

Winbush et al.
SCHEME 15. Transannular Diels-Alder Reaction of 13
with 94% ee from commercially available propane 1,3-diol.
Protection of 36 as a p-methoxybenzyl (PMB) ether followed
by deprotection of the TBDPS ether gave alcohol 37.21,22 Olefin
dihydroxylation of 37, followed by oxidative cleavage of the
resultant diol afforded hemiacetal 38.42 Hemiacetal 38 was
treated with Ph3PdCHCO2Me3 to give 39. Oxidation of alcohol
39 followed by the application of the catalytic Takai-Utimoto
olefination43 led to vinyl iodide 40 with an optimized E:Z ratio
of 29:1. Reduction of the methyl ester to the primary alcohol
and subsequent oxidation of the allylic alcohol to the enal
completed the synthesis of fragment 35 (Scheme 13).
obtained; the third minor product could not be isolated in
sufficient quantity or purity for complete structural characteriza-
tion.44 The diastereoselectivity for the conversion of 13 to 44
was 70:18:12, as determined by HPLC analysis of reaction
mixtures (Scheme 15).45
Macrocycle 13 was not detected at any stage of the reaction,
as it spontaneously underwent cycloaddition under conditions
of the intramolecular HWE reaction. However, small amounts
of material believed to be the macrocyclic (Z)-enoate 46 (J2,3
) 11. 6 Hz) were isolated and observed to undergo cycload-
dition, either at ambient temperature or more rapidly upon mild
heating (Scheme 16). Characterization of 46 was complicated
by its tendency to cyclize to a mixture of adducts 47 during all
manipulations at ambient temperature (including attempted
chromatographic purification). Attempts to purify and assign
stereochemistry to the cycloadducts produced from 46 were not
performed owing to the limited quantities of 46 that were
available.
Synthesis of TDA cyclization precursor 43 proceeded via the
adjoining of fragments 35 and 16 by treatment with Ba(OH)2,20
followed by a series of standard functional group manipulations
depicted in Scheme 14.
Treatment of 43 with LiCl and iPr2NEt in CH3CN27 at
ambient temperature over 3 days led to a mixture of cycload-
ducts (58% yield, combined) from which the major cycloadduct
44 was obtained in 34% yield. Two minor diastereomers of
which adduct 45 was the second most abundant were also
(44) The HWE-TDA cyclization of 43 was also performed with NaH in
THF at 23 °C over 5-10 h (51% yield). However, while product yields
were comparable, the NaH-promoted HWE reaction mixture contained more
byproducts than did the LiCl-iPr2NEt reaction mixtures.
(41) Smith, A. B., III; Safonov, I. G.; Corbett, R. M. J. Am. Chem. Soc.
2002, 124, 11102.
(42) VanRheenen, V.; Kelly, R. C.; Cha, D. Y. Tetrahedron Lett. 1976,
23, 1973.
(43) Takai, K.; Ichiguchi, T.; Hikasa, S. Synlett 1999, 8, 1268.
(45) Diastereomer ratio determined by analytical reverse-phase HPLC
(acetonitrile:water:trifuloroacetic acid) with C-18 OD column UV detection
at λ ) 215, 230, 254, 280, and 420 nm.
1824 J. Org. Chem., Vol. 73, No. 5, 2008

Total Synthesis of (-)-Spinosyn A
SCHEME 16. TDA Cycloaddition of Isomeric Macrocycles
13 and 46
SCHEME 18. Strategy for Synthesis of 48 via Macrocycle
14
SCHEME 17. Synthesis of (-)-Spinosyn A Aglycon 34
SCHEME 19. Synthesis of â-Ketophosphonate 49
H(11), and H(12) with both H(7) and H(4), as indicated in the
three-dimensional representation of 34 in Scheme 17.
Remarkably, the diastereoselectivity of the transannular
cycloaddition of 13 to 44 is quite similar to that observed for
the conversion of 12 to 33, suggesting the C(6)-bromine
directing group is not a significant element of stereocontrol in
the transannular Diels-Alder reaction of 12. The carbocyclic
framework of macrocycle 13 alone must be responsible for the
favorable stereochemical outcome of this cycloaddition.
The final series of experiments in this study were designed
to probe our hypothesis that the C(21)-stereocenter might be a
stereochemical control element for these transannular Diels-
Alder reactions. Specifically, we decided to study the TDA
cyclization of macrocycle 14, lacking the C(21)-stereocenter (see
Scheme 18).
The synthesis of â-ketophosphonate 49 (Scheme 19) pro-
ceeded by way of the known intermediate 50,4 which was
protected as a TES ether before conversion of the Weinreb
amide unit to the â-ketophosphonate functionality of 49.47
The Horner-Wadsworth-Emmons coupling of 35 and 49
led to the isolation of polyene 51 in 92% yield.20 Intermediate
51 was then converted to macrocyclization substrate 53 through
the series of transformations shown in Scheme 20.
Partial stereochemical assignment of 44 was based on the
1H NOE interactions summarized in Scheme 15. In these
stereochemical assignments, the C(9) stereocenter served as an
important reference point, as its configuration was determined
with confidence as it was installed through the highly enanti-
oselective asymmetric Brown allylation reaction.46 However,
it was difficult to observe the expected H(9)-H(11) cross-peak,
since H(11) is buried in the hydrocarbon region of the 1H NMR
spectrum of 44. We opted to convert 44 to the spinosyn aglycon
34 through a vinylogous Morita-Baylis-Hillman cyclization,32-34
followed by deprotection of the PMB ethers (Scheme 17).
Spectroscopic data obtained for synthetic aglycon 34 com-
pared favorably with the literature data for the natural (-)-
spinosyn A aglycon (1H NMR, 13C NMR, HRMS).1 Further,
through a series of 2D NOESY and 1D NOE experiments, NOE
cross-peaks were observed between H(9) and H(11), H(3) and
Following the protocol for the macrocyclization reactions of
21, 32, and 43, intermediate 53 was initially subjected to the
(46) Brown, H. C.; Jadhav, P. K. J. Am. Chem. Soc. 1983, 105, 2092.
(47) Theisen, P. D.; Heathcock, C. H. J. Org. Chem. 1988, 53, 2374.
J. Org. Chem, Vol. 73, No. 5, 2008 1825

Winbush et al.
SCHEME 20. Synthesis of Precursor 53
lethylamine did not result in macrocyclization; only products
of hydrolysis of the C(21) O-acyl group were observed.
Gratifyingly, treatment of 53 with NaH in THF at 0.001 M
concentration led to the isolation of adduct 54.48,49 The
diastereoselectivity for this one-pot cyclization was 71:18:16:5
(as determined by HPLC analysis),45 of which 54 and 55
were identified as the two most abundant cycloadducts (Scheme
21).
The stereochemical arrangement about the newly generated
tricyclic core of the two major adducts 54 and 55 was established
1
through H NOE studies (see Scheme 21). The crucial H(9)-
H(11) cross-peak for the most abundant diastereomer 54,
together with J7,11 ) 11.6 Hz, confirmed that 54 has the
appropriate trans-fused C(7)-C(11) configuration, as in the case
of (-)-spinosyn A (1). Stereoisomer 55 displayed an NOE cross-
peak between H(9) and H(7) together with J7,11 ) 12.0 Hz,
indicating that 55 has the relative stereochemistry shown,
opposite to the natural product (1).
The stereochemical assignment of the most abundant dias-
tereomer 54 was further supported by extensive 1H NOE studies
performed on tetracycle 56, which was generated as shown in
Scheme 22 by the Me3P-mediated vinylogous MBH cyclization.
2D NOESY data obtained for 56 suggest diastereomer 54 has
the same stereochemistry about the indacene core as the natural
product (1).
Masamune27 conditions for HWE olefination; however, treat-
ment of 53 with an excess of lithium chloride and diisopropy-
SCHEME 21. Cycloaddition of 14 and Assignment of Stereochemistry of Two Major Adducts 54 and 55
1826 J. Org. Chem., Vol. 73, No. 5, 2008

Total Synthesis of (-)-Spinosyn A
SCHEME 22. Synthesis of Tetracycle 56 To Confirm
Stereochemical Assignment of 54
striking that the stereoselectivities of TDA cycloadditions of
12, 13 (which lacks the C(6)-Br unit of 12), and 14 (which lacks
the C(6)-Br and C(21)-Et substituents of 12) are very similar
(Table 2). Given that the cycloadditions of 12-14 are more
selective for the C(7)-C(9) stereochemical relationship found
in the spinosyns than the IMDA reactions of 4 (Scheme 2)2 or
7 (Scheme 3),4,15 we conclude that some as yet unidentified
conformational preference of the macrocycles in the competing
transition states TS-5 and TS-6 (Scheme 15) and TS-7 and TS-8
(Scheme 21) contributes to the enhanced diastereoselectivity
of the transannular cyclizations. In contrast, the TDA cycload-
dition of 9, which possesses an extra C(8)-OTBS substituent
compared to 12, undergoes a highly diastereoselective transan-
nular cyclization (>95:5 dr). The latter result is attributed to
1,3-allylic strain interactions involving C(6)-Br and C(8)-OTBS
in the disfavored TS-2.
Experimental Section
(1R,2S,3aS,4S,5E,8R,9S,13S,16E,18R,20aS)-20-Bromo-1-(tert-
butyldimethylsiloxy)-13-ethyl-9-(4-methoxybenzyloxy)-8-methyl-
2-[(6-deoxy-2,3,4-trimethyl-O-methyl-R-L-mannopyranosyl)oxy]-
2,3,3a,4,18,20a-hexahydro-1H-indeno[5,4-e]oxacyclopentadeca-
5,16-diene-7,15-dione (22). THF and H2O were degassed separately
via the freeze-pump-thaw method (3 cycles, vent to argon). Vinyl
boronic acid 20 (190 mg, 1.9 mmol) was transferred in MeOH to
a flask containing 19 (370 mg, 0.34 mmol) and this mixture was
concentrated under reduced pressure, placed under high vacuum,
and vented to argon. The evacuate/vent cycle was repeated three
times and then degassed THF (9 mL) was added followed by
degassed H2O (3 mL). Pd(PPh3)4 (275 mg, 0.24 mmol) was added
and the resulting yellow suspension was stirred for 5 min at which
point Tl2CO3 (320 mg, 0.68 mmol) was added.6 The resulting
yellow, heterogeneous reaction was protected from light with
aluminum foil and stirred under argon for 50 min. At this point,
additional Pd(PPh3)4 (120 mg, 0.10 mmol) and a small amount of
Tl2CO3 were added. Stirring was continued for an additional 1 h at
which point phosphonate 19 was consumed as judged by TLC
analysis. The pale yellow reaction mixture was diluted with Et2O
followed by 1 M NaHSO4. The resulting bright yellow biphasic
mixture was stirred vigorously for 10 min and then filtered through
Celite, rinsing with Et2O and EtOAc. The layers were separated
and the aqueous layer was extracted with EtOAc. The combined
organics were washed with brine, dried over Na2SO4, filtered, and
concentrated yielding a yellow solid. Purification of this material
by column chromatography (40 × 140 mm silica, 1:1 hexanes:
acetone) afforded the allylic alcohol SI-3 (231 mg, 63%) as a yellow
oil: 1H NMR (500 MHz, CDCl3) δ 7.25-7.21 (m, 2 H), 7.15 (dd,
J ) 15.1, 9.8 Hz, 1 H), 6.88-6.84 (m, 2 H), 6.32-6.15 (m, 5 H),
5.91 (d, J ) 8.3 Hz, 1 H), 5.03 (d, J ) 1.2 Hz, 1 H), 4.83-4.77
(m, 1 H), 4.64 (dd, J ) 8.3, 4.2 Hz, 1 H), 4.47 (d, A of AB system,
J ) 11.0 Hz, 1 H), 4.41 (d, B of AB system, J ) 11.0 Hz, 1 H),
4.31-4.27 (m, 2 H), 4.18-4.11 (m, 4 H), 3.79 (s, 3 H), 3.79-
3.76 (m, 1 H), 3.65-3.60 (m, 1 H), 3.57 (dd, J ) 9.5, 6.4 Hz, 1
H), 3.55-3.53 (m, 1 H), 3.51 (s, 3 H), 3.48 (s, 3 H), 3.43 (s, 3 H),
3.41 (dd, J ) 9.8, 3.2 Hz, 1 H), 3.09 (t, J ) 9.3 Hz, 1 H), 2.95-
2.89 (m, 1 H), 2.92 (dd, J ) 21.5, 3.2 Hz, 2 H), 2.47-2.36 (m, 2
H), 2.30-2.20 (m, 1 H), 1.58-1.41 (m, 8 H), 1.32 (t, J ) 7.1 Hz,
6 H), 1.20 (d, J ) 6.4 Hz, 3 H), 1.15 (d, J ) 6.8 Hz, 3 H), 0.88
(m, 3 H), 0.88 (s, 9 H), 0.07 (s, 3 H), 0.02 (s, 3 H); 13C NMR (125
MHz, CDCl3) δ 202.5, 165.5 (d, J ) 6.6 Hz), 159.1, 142.4, 141.3,
132.0, 131.1, 130.4, 129.4, 128.5, 128.4, 127.5, 125.4, 113.7, 97.9,
81.9, 81.1, 80.9, 80.0, 77.4, 74.5, 72.1, 68.2, 62.5 (d, J ) 1.8 Hz),
60.7, 58.9, 57.6, 55.2, 48.7, 35.2, 34.9, 33.8, 33.3, 32.6, 26.7, 25.7,
TABLE 2. Summary of TDA Diastereoselectivities
conversion
diastereoselectivity
9 to 22
12 to 33
13 to 44
14 to 54
>95:5
73:12:9:6
70:18:12
71:18:6:5
It is striking that the diastereoselectivity of the TDA cycload-
ditions of 12, 13, and 14 are very similar (Table 2). Clearly,
neither the C(21)-Et substituent nor the C(6)-Br steric directing
group contribute substantially to the diastereoselectivity of the
TDA cyclization of 12. That the TDA events of all three
substrates are more selective than the IMDA reaction of 7
(Scheme 3),4,15 and are considerably more selective for the
correct C(7)-C(9) stereochemistry than is the IMDA reaction
of 4 (Scheme 2),2 remains suggestive that some conformational
feature of the 22-membered ring of 9, 12, and 13 is responsible
for the enhanced, and more favorable, diastereoselectivity of
these TDA reactions. However, it has not been possible to
identify the factor (or factors) that is responsible for this
enhanced selectivity owing to the conformational flexibility (i.e.,
multiple conformations) of the developing 15-membered ring
in the competing TDA transition states (TS-5 and TS-6 in
Scheme 15, and TS-7 and TS-8 in Scheme 21).
Conclusion
We have studied potential elements of stereochemical control
in transannular Diels-Alder (TDA) reactions leading to the
perhydroindacene ring system of (-)-spinosyn A (1). It is
21.4, 17.9, 17.6, 16.3 (d, J ) 6.1 Hz), 12.8, 9.4, -4.3, -5.0; [R]26.8
D
(48) Stocksdale, M. G.; Ramurthy, S.; Miller, M. J. J. Org. Chem. 1998,
63, 1221.
(49) Menche, D.; Hassfeld, J.; Li, J.; Rudolph, S. J. Am. Chem. Soc.
2007, 129, 6100.
-41.8 (c 9.5, CHCl3); IR (thin film) 3402, 2932, 2858, 1732, 1683,
1656, 1636, 1612, 1593, 1514, 1463, 1441, 1389, 1368, 1251, 1198,
1174, 1118, 1055, 966, 914, 838, 779, 756, 723, 696, 666 cm-1
;
J. Org. Chem, Vol. 73, No. 5, 2008 1827

Winbush et al.
HRMS (ES) calcd for C51H84BrO15PSi [M + H]+ m/z 1077.4398,
found 1077.4399.
layer was extracted three times with EtOAc. The organic layers
were combined and washed with brine, dried over anhydrous
MgSO4, filtered, and concentrated. Purification of the crude mixture
(a 70:18:12 mixture of cycloadducts; HPLC analysis) by silica gel
column chromatography (4:1 hexanes:ethyl acetate) allowed for the
isolation of 44 (5.6 mg, 35%, colorless oil) as the most abundant
diastereomer, along with an inseparable mixture of minor diaster-
eomers (overall 9.4 mg, 58% total product). The mixture of minor
diastereomers was then subjected to normal phase HPLC (4:1
hexanes:ethyl acetate) allowing for partial separation to provide
diastereomer 45 as the second most abundant reaction product. The
ratio of diastereomers was determined by reverse-phase analytical
HPLC (C-18 OD column 65:35:0.01 acetonitrile:water:trifluoro-
acetic acid) at simultaneous UV detection of λ ) 215, 230, 254,
280, and 450 nm.
The above allylic alcohol (231 mg, 0.21 mmol) was azeotropi-
cally dried by coevaporation from benzene and then dissolved in
CH2Cl2 (10.5 mL). i-Pr2NEt (0.26 mL, 1.5 mmol) was added
followed by DMSO (0.13 mL, 1.8 mmol). The reaction mixture
was cooled to 0 °C and SO3‚pyr (140 mg, 0.88 mmol) was added.
The resulting reaction mixture was stirred at 0 °C for 25 min at
which point TLC analysis indicated the disappearance of SI-3. The
reaction mixture was diluted with saturated NaHCO3 (∼7 mL), the
ice bath was removed, and the biphasic mixture was stirred
vigorously for 5 min. This mixture was diluted with EtOAc, the
layers were separated, and the aqueous layer was extracted with
EtOAc. The combined organics were washed with brine, dried over
Na2SO4, filtered, and concentrated yielding a yellow oil. Residual
pyridine was removed by coevaporation from benzene and the
resulting aldehyde 21 was used directly in the next reaction.
The crude aldehyde 21 from the preceding experiment (theoreti-
cally, 0.21 mmol) was dissolved in CH3CN (210 mL) under argon.
Dry LiCl (340 mg, 8.0 mmol) was added and the resulting
heterogeneous mixture was stirred slowly for 5 min to allow the
stir bar to grind up the LiCl. i-Pr2NEt (1.25 mL, 7.2 mmol) was
added and the mixture was slowly stirred until the LiCl was very
finely ground. The stirring velocity was then increased and the
reaction mixture was stirred for 12 h at which point ESMS analysis
indicated that the starting aldehyde 21 had been consumed. The
reaction mixture was poured into a biphasic mixture of EtOAc (300
mL) and 1 M NaHSO4 (100 mL). The reaction flask was rinsed
with EtOAc (100 mL) and this rinse was added to the biphasic
mixture. The mixture was agitated, the layers were separated, and
the aqueous layer was extracted with EtOAc. The combined
organics were washed with brine, dried over Na2SO4, filtered, and
concentrated yielding the crude product (g95:5 mixture of cy-
cloadducts). Purification of this material by column chromatography
(35 × 70 mm silica, 3:2 hexanes:EtOAc) afforded 22 (150 mg,
78%) as a yellow foam. An analytical sample of 22 was prepared
by preparative HPLC (3:2 hexanes:EtOAc): 1H NMR (500 MHz,
CDCl3) δ 7.27-7.23 (m, 2 H), 6.90-6.86 (m, 2 H), 6.62 (dd, J )
16.6, 4.9 Hz, 1 H), 6.45 (dd, J ) 15.6, 9.3 Hz, 1 H), 6.05 (dd, J )
16.6, 1.7 Hz, 1 H), 5.83 (dd, J ) 4.4, 2.7 Hz, 1 H), 5.71 (dd, J )
15.4, 0.5 Hz, 1 H), 4.91 (d, J ) 1.7 Hz, 1 H), 4.90-4.85 (m, 1 H),
4.48 (d, A of AB system, J ) 11.0 Hz, 1 H), 4.42 (d, B of AB
system, J ) 10.7 Hz, 1 H), 3.98 (app dt, J ) 6.4, 3.9 Hz, 1 H),
3.91 (dd, J ) 9.8, 5.9 Hz, 1 H), 3.81 (s, 3 H), 3.65 (dd, J ) 3.2,
1.7 Hz, 1 H), 3.55 (s, 3 H), 3.50-3.46 (m, 1 H), 3.49 (s, 3 H),
3.48 (s, 3 H), 3.46 (dd, J ) 9.3, 3.2 Hz, 1 H), 3.34-3.29 (m, 1 H),
3.13 (t, J ) 9.5 Hz, 1 H), 3.04 (app quint., J ) 7.1 Hz, 1 H),
2.79-2.73 (m, 1 H), 2.67-2.61 (m, 1 H), 2.52 (app quint., J )
7.3 Hz, 1 H), 1.76 (app dq, J ) 11.2, 7.6 Hz, 1 H), 1.64-1.24 (m,
10 H), 1.25 (d, J ) 6.1 Hz, 3 H), 1.17 (d, J ) 6.8 Hz, 1 H), 0.97
Method B: To a solution of dienal 43 (12.2 mg, 0.0153 mmol,
1.0 equiv) in THF (15 mL) was added NaH (1.8 mg, 0.075 mmol,
5.0 equiv) 95% in oil dispersion. The reaction was stirred 5 h at
ambient temperature, at which point aqueous saturated NH4Cl (1
mL) solution was added to quench the reaction. The mixture was
diluted with Et2O and H2O. The aqueous layer was extracted with
Et2O three times. The organic layers were combined and washed
with brine, dried over MgSO4, filtered, and concentrated. Purifica-
tion of the mixture (see method A) allowed for the isolation of 44
(0.0028 mg, 35%, colorless oil) as the most abundant diastereomer,
along with an inseparable mixture of minor diastereomers (overall
0.005 mg, 51% total product). Data for 44: [R]23.7 -62.1 (c 1.0,
D
CHCl3); 1H NMR (400 MHz, CDCl3) δ 7.28-7.22 (m, 4H), 6.89-
6.85 (m, 4H), 6.73 (dd, J ) 16.6, 4.2 Hz, 1H), 6.51 (dd, J ) 15.8,
9.4 Hz, 1H), 6.12 (dd, J ) 16.6, 1.8 Hz, 1H), 5.96 (d, J ) 9.6 Hz,
1H), 5.71 (d, J ) 15.6 Hz, 1H), 5.38 (dt, J ) 9.6, 3.4 Hz, 1H),
4.91-4.83 (m, 1H), 4.49 (d, J ) 10.8 Hz, 1H), 4.41 (d, J ) 10.8
Hz, 1H), 4.39 (br s, 2H), 4.16-4.11 (m, 1H), 3.80 (s, 6H), 3.55-
3.50 (m, 1H), 3.35-3.27 (m, 1H), 3.09 (dq, J ) 6.8, 1.6 Hz, 1H),
2.85-2.78 (m, 1H), 2.60 (ddd, J ) 6.4, 6.4, 6.4 Hz, 1H), 2.38-
2.27 (m, 1H), 2.04 (dd, J ) 13.0, 6.6 Hz,1H), 1.64-1.35 (m, 11H),
1.19 (d, J ) 6.8 Hz, 3H), 0.84 (t, J ) 7.4 Hz, 3H); 13C NMR (100
MHz, CDCl3) δ 203.5, 165.6, 159.2, 147.4, 131.1, 130.8, 130.5,
129.5, 129.3, 127.8, 122.9, 113.8, 81.4, 78.1, 76.4, 71.9, 70.5, 55.3,
46.3, 44.3, 42.8, 42.3, 41.3, 37.4, 36.7, 33.6, 32.2, 27.6, 21.1, 15.5,
9.9; IR (thin film, NaCl) 2918, 2850, 1710, 1660, 1586, 1513, 1462,
1355, 1301, 1248, 1207, 1173, 1109, 1060, 1034, 982, 821, 752
cm-1; HRMS (ESI) m/z 665.3476 [calcd M + Na+ C40H50O7Na
665.3449].
Data for 45: (2R,3aR,5aR,6E,10S,14S,15R,17E,18aR,18bS)-
10-Ethyl-2,14-bis(4-methoxybenzyloxy)-15-methyl-3,3a,10,11,-
12,13,14,15,18a,18b-decahydro-1H-indeno[5,4-e][1]-
oxacyclopentadecine-8,16(2H,5aH)-dione (45). [R]24.3 -9.0 (c
D
1
0.2, CHCl3); H NMR (400 MHz, CDCl3) δ 7.29-7.21 (m, 4H),
6.91-6.84 (m, 4H), 6.75 (dd, J ) 15.4, 10.2 Hz, 1H), 6.64 (dd, J
) 15.2, 11.2 Hz, 1H), 6.25 (d, J ) 14.8 Hz, 1H), 6.19 (dt, J ) 9.0,
3.0 Hz, 1H), 5.80 (dt, J ) 9.2, 2.8 Hz, 1H), 5.58 (d, J ) 15.6 Hz,
1H), 4.92-4.83 (m, 1H), 4.59 (d, A of AB system, J ) 11.2 Hz,
1H), 4.45 (d, B of AB system, J ) 11.2 Hz, 1H), 4.20-4.06 (m,
1H), 3.81 (s, 3H), 3.80 (s, 3H), 3.32-3.28 (m, 1H), 2.87 (qd, J )
6.8, 6.2 Hz, 1H), 2.56 (app t, J ) 10.0 Hz, 1H), 2.48 (dt, J ) 13.2,
7.2 Hz, 1H), 2.16 (app q, J ) 10.0 Hz, 1H), 2.00-1.95 (m, 1H),
1.81-1.11 (m, 12H), 1.08 (d, J ) 6.8 Hz, 3H), 0.86 (t, J ) 7.2
Hz, 3H); 13C NMR (100 MHz, CDCl3) δ 200.8, 166.2, 159.1, 147.3,
143.7, 135.2, 131.6, 130.5, 129.4, 129.2, 129.1, 128.3, 128.1, 124.3,
113.8, 82.1, 78.9, 71.2, 70.5, 55.3, 49.1, 44.6, 43.8, 43.6, 41.1, 36.9,
34.5, 32.5 (app d), 27.8, 23.0, 13.71, 9.72; IR (thin film, NaCl)
2920, 2851, 1714, 1657, 1586, 1513, 1463, 1377, 1302, 1248, 1172,
1035, 888, 821, 760, 720 cm-1; HRMS (ESI) m/z 665.3403 [calcd
M + Na+ C40H50O7Na 665.3449].
(s, 9 H), 0.88 (t, J ) 7.3 Hz, 3 H), 0.19 (s, 3 H), 0.10 (s, 3 H); 13
C
NMR (125 MHz, CDCl3) δ 202.6, 165.2, 159.2, 145.0, 144.7, 131.3,
130.4, 130.3, 129.5, 123.8, 123.6, 113.8, 99.0, 82.2, 81.3, 81.1,
78.0, 76.6, 76.1, 71.9, 68.6, 61.1, 59.3, 57.8, 55.3, 51.5, 46.8, 45.1,
43.8, 37.5, 35.2, 33.4, 32.2, 27.5, 26.2, 21.1, 17.9, 17.7, 15.2, 9.9,
-3.3, -4.5; [R]26.8 -150.0 (c 0.2, CHCl3); IR (thin film) 2930,
D
2857, 1713, 1664, 1614, 1514, 1462, 1361, 1250, 1123, 1143, 1119,
1104, 1057, 1035, 863, 839, 774, 680 cm-1; HRMS (ES) calcd for
C47H71BrO11Si [M + Na]+ m/z 941.3847, found 941.3847.
(2R,3aS,5aS,6E,10S,14S,15R,17E,18aS,18bR)-10-Ethyl-2,14-
bis(4-methoxybenzyloxy)-15-methyl-3,3a,10,11,12,13,14,15,18a,-
18b-decahydro-1H-indeno[5,4-e][1]oxacyclopentadecine-8,16-
(2H,5aH)-dione (44). Method A: To a solution of dienal 43 (20
mg, 0.25 mmol, 1.0 equiv) in CH3CN (25 mL) was added LiCl
(39 mg, 0.93 mmol, 34 equiv; stored in a glovebox and flame dried
under vacuum prior to use). Distilled diisopropylethylamine (0.15
mL, 0.85 mmol, 34 equiv) was added to the mixture, and the
reaction was stirred for 3 days. The reaction was quenched with 1
N KHSO4 (1 mL), then diluted with EtOAc and H2O. The aqueous
(2R,3aS,5aS,6E,14S,15R,17E,18aS,18bR)-2,14-Bis(4-methoxy-
benzylozy)-15-methyl-3,3a,10,11,12,13,14,15,18a,18b-decahydro-
1H-indeno[5,4-e][1]oxacyclopentadecine-8,16(2H,5aH)-dione (54).
To a solution of dienal 53 (50 mg, 0.066 mmol, 1.0 equiv) in THF
1828 J. Org. Chem., Vol. 73, No. 5, 2008

Total Synthesis of (-)-Spinosyn A
(13 mL) was added NaH (7 mg, 0.292 mmol, 4.4 equiv) 95% in
oil dispersion. The reaction was stirred for 23 h at ambient
temperature, at which point aqueous saturated NH4Cl (1 mL)
solution was added. The mixture was diluted with Et2O and H2O.
The aqueous layer was extracted three times with Et2O. The organic
layers were combined and washed with brine, dried over MgSO4,
filtered, and concentrated. Purification of the crude mixture (a
71 : 18 6 5 mixture of cycloadducts by HPLC analysis) by normal
phase HPLC (3:1 hexanes:ethyl acetate) led to the isolation of 54
(16.2 mg, 40% overall mixture, colorless oil) as the most abundant
diastereomer, along with a mixture of partially separable minor
Data for 55: (2R,3aR,5aR,6E,14S,15R,17E,18aR,18bS)-2,14-
Bis(4-methoxybenzyloxy)-15-methyl-3,3a,10,11,12,13,14,15,18a,-
18b-decahydro-1H-indeno[5,4-e][1]oxacyclopentadecine-8,16-
(2H,5aH)-dione (55). [R]24.4D + 45.2 (c 0.46, CHCl3); 1H NMR δ
7.29-7.21 (m, 4H), 6.90-6.85 (m, 4H), 6.76 (ddd, J ) 15.6, 7.2,
2.8 Hz, 1H), 6.14 (d, J ) 16.4 Hz, 1H), 6.02 (d, J ) 10.0, 3.4 Hz,
1H), 5.68 (dd, J ) 15.6, 0.8 Hz, 1H), 4.52-4.35 (m, 4H), 4.25-
4.19 (m, 1H), 4.15-4.05 (m, 2H), 3.81 (s, 3H), 3.80 (s, 3H), 3.59-
3.54 (m, 1H), 3.31-3.25 (m, 1H), 3.07 (dq, J ) 7.2, 1.2 Hz, 1H),
2.71 (app dt, J ) 10.8, 7.2 Hz, 1H), 2.41-2.34 (m, 1H), 1.96-
1.28 (m, 7H), 1.17 (d, J ) 6.8 Hz, 3H), 1.13 (dd, J ) 6.8, 3.2 Hz,
2H); 13C NMR δ 203.0, 165.5, 159.2, 148.6, 148.2, 131.4, 130.9,
130.7, 130.4, 129.4, 129.3, 129.1, 126.6, 123.3, 113.8 (app d), 80.7,
78.3, 72.0, 70.7, 64.6, 55.3, 47.6, 45.4, 43.4, 42.3, 40.0, 37.9, 36.2,
32.1, 28.1, 22.4, 15.0; IR (thin film, NaCl) 2934, 1715, 1613, 1513,
1455, 1302, 1248, 1173, 1035, 821, 731 cm-1; HRMS (ESI) m/z
637.3093 [calcd M + Na+ C38H46O7Na 637.3135].
diastereomers of which 55 proved to be the second most abundant.
1
Data for 54: [R]25.2 -91.3 (c 1.0, CHCl3); H NMR (400 MHz,
D
CDCl3) δ 7.29-7.22 (m, 4H), 6.90-6.85 (m, 4H), 6.81-6.74 (m,
1H), 6.07 (dd, J ) 16.8, 1.6 Hz, 1H), 6.02 (app d, J ) 10.0 Hz,
1H), 5.66 (dd, J ) 15.6, 1.2 Hz, 1H), 5.45 (dt, J ) 10.0, 3.4 Hz,
1H), 4.52 (d, J ) 3.2 Hz, 2H), 4.38 (s, 2H), 4.26-4.20 (m, 1H),
4.13-4.05 (m, 2H), 3.81 (s, 3H), 3.80 (s, 3H), 3.52-3.45 (m, 1H),
3.32-3.25 (m, 1H), 3.14 (dq, J ) 16.8, 1.2 Hz, 1H), 2.88-2.82
(m, 1H), 2.52-2.46 (m, 1H), 2.34-2.25 (m, 1H), 2.03 (dd, J )
13.0, 6.6 Hz, 1H), 1.70-1.34 (m, 8H), 1.29 (dd, J ) 12.6, 4.6 Hz,
1H), 1.22 (d, J ) 6.8 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ
203.1, 165.7, 159.3, 159.2, 149.1, 147.3, 131.7, 130.9, 130.5 (app
d), 129.4 (app d), 126.5, 123.2, 113.9 (app d), 82.0, 78.2, 72.7,
70.6, 65.1, 55.3 (app d), 47.3, 45.1, 42.5, 41.8, 40.6, 37.0, 36.7,
33.8, 28.0, 23.8, 15.1; IR (thin film, NaCl) 2929, 1714, 1658, 1613,
1586, 1513, 1456, 1353, 1302, 1247, 1172, 1034, 986, 912, 822,
754, 666 cm-1; HRMS (ESI) m/z 637.3132 [calcd M + Na+
C38H46O7Na 637.3136].
Acknowledgment. Financial support provided by the Na-
tional Institutes of Health (GM 26782) is gratefully acknowl-
edged. We thank Dr. Ted Kamenecka for assistance with HPLC
analysis of the TDA cycloadducts.
Supporting Information Available: Complete experimental
1
details and H and 13C NMR spectra. This material is available
free of charge via the Internet at http://pubs.acs.org.
JO7024515
J. Org. Chem, Vol. 73, No. 5, 2008 1829