Controlling influences in bisspiroketal formation: Synthesis of the ABC ring system of azaspiracid

Article abstract of DOI:10.1016/j.tet.2003.03.003

Substitution effects on the stereochemical outcome of bisspiroketalization on the C₁-C₁₇ carbon backbone of azaspiracid is presented. A possible explanation is offered to explain the observed stereochemical outcome.

Full text of DOI:10.1016/j.tet.2003.03.003

Tetrahedron 59 (2003) 8963–8974
Controlling influences in bisspiroketal formation: synthesis of the
ABC ring system of azaspiracid
†
Rich G. Carter, T. Campbell Bourland, Xiao-Ti Zhou and Melissa A. Gronemeyer
*
Department of Chemistry, Oregon State University, Corvallis, OR 97331, USA
Received 20 February 2003; accepted 21 March 2003
Dedicated to Dr Lawrence V. Puckett on the occasion of his retirement
Abstract—Substitution effects on the stereochemical outcome of bisspiroketalization on the C1–C17 carbon backbone of azaspiracid is
presented. A possible explanation is offered to explain the observed stereochemical outcome.
q 2003 Elsevier Ltd. All rights reserved.
1. Introduction
Bisspiroketals have generated a considerable amount of
synthetic and biological attention in recent years.1 One
focus area in bisspiroketal chemistry has been under-
standing the controlling stereochemical factors in bis-
spirocyclization. In bisspiroketals which possess a pyran
ring system, the anomeric effect plays an important
governing role in the stereochemical outcome by guiding
the relevant oxygen into an axial position.2 An additional
Figure 1. Transoidal and cisoidal bisspiroketals.
controlling influence that has been proposed involves a
minimization of the two dipoles present in the external
carbon–oxygen bonds of the bisspiroketals. This dipole
minimization argument has been put forth to explain the
these reasons, a stabilizing interaction (usually via hydrogen
observed slight (usually less than 1 kcal/mol) thermo-
bonding or metal-chelation)1,5 of a neighboring functional
dynamic preference for the transoidal spirocycle, such as
1, over the corresponding cisoidal spirocycle, such as 2
group has also been invoked to explain the stabilization of a
specific spiroketal stereochemistry.
(Fig. 1).3,4 Despite these important advances in the under-
standing of bisspirocyclization, more complicated systems
Our laboratory became focused on this issue during our
synthetic efforts toward the transoidal bisspiroketal
azaspiracid (3) (Fig. 2).5–7 Based on the solution confor-
(e.g. possessing multiple additional stereocenters and/or
functional groups) often fall victim to explanations which
may not take into full account several of the controlling
mation proposed by Yasumoto and co-workers,8,9 the C13
influences of the reaction. For example, once an additional
stereogenic center is placed on one or more of the rings,
spiroketal linkage of 3 does not appear to be oriented in the
anomeric conformation as depicted in simplified example 1.
accessing a conformer that satisfies the anomeric effect for
Instead, Yasumoto proposed the alternate chair confor-
both the cisoidal and the transoidal bisspirocycle can
mation 4 in which the center oxygen is located in an
become prohibitively high in energy, due to the develop-
equatorial orientation with respect to the C ring. Further-
ment of unfavorable 1,3-diaxial interactions in the pyran
more, no neighboring hydrogen bond donors are suitably
rings. In such systems where the desired stereochemistry is
positioned on the A, B, C or D rings for stabilization and a
not thermodynamically favored due to one (or more) of
metal-chelation argument has not been proposed to explain
this stereochemical and conformational arrangement. It
Keywords: azaspiracid; bisspiroketal; bisspirocyclization; Julia coupling;
substituent effects; anomeric effect.
would appear from these observations that further insight
into the controlling influences of bisspiroketalization is
necessary. Particularly intriguing is the possible effects
of the nature of substitution at C16 and C17 on the
stereochemical outcome of bisspirocyclization. Herein, we
*
Corresponding author. Tel.: þ1-541-737-9486; fax: þ1-541-737-9496;
e-mail: rich.carter@oregonstate.edu
Present address: Baylor Dental School, 3302 Gaston Ave., Dallas, TX
75246, USA.
†
0040–4020/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2003.03.003

8964
R. G. Carter et al. / Tetrahedron 59 (2003) 8963–8974
significant effect of this toxin on the European shellfish
industry13 and its daunting structure garnered our attention.
In order to access a variety of substitution patterns on the C
ring, our synthetic approach had to be significantly flexible
(Scheme 1). The bisspirocyclization precursor 7 was
designed to possess the necessary ketone function at C13
and a methoxy ketal function at C10. This approach allowed
for the incorporation of the A ring into the sulfone fragment
8.6a A variety of corresponding electrophiles 9–11 needed
to be screened in order to ascertain the ideal coupling
partner. Once the coupling strategy had been optimized, it
could be applied to systems possessing different degrees of
substitution at C16 and C17.
3. Results
As shown in Scheme 2, the formation of the C12,13 linkage
was explored on an electrophilic fragment with the D ring
already incorporated (electrophiles 12–146c). The ideal
coupling partner would be the C13 lactone 12 as it possesses
the correct oxidation state at C13 and the C17 hydroxyl is
internally protected. While a wealth of examples of
sulfone–ester couplings have been disclosed,15 significantly
fewer examples of sulfone–lactone couplings have been
reported.16 Unfortunately, the lactone 12 proved disappoint-
ing, even under our optimum protocol [LDA (2.2 equiv.),
THF, 2788C, 20%]. A considerable amount of decompo-
sition of the lactone 12 was consistently observed under a
variety of conditions. The corresponding methyl ester 13
proved only slightly more encouraging with up to a 40%
yield [LDA (1.2 equiv.), 13 (0.5 equiv.), then LDA
(0.5 equiv.), 13 (0.6 equiv.)]. The low yield may be
attributed to deprotonation of the ester by the lithiated
sulfone species; a considerable amount of epimerization
was observed in recovered 13. We were gratified to find that
the corresponding aldehyde 14 proved to be an effective
coupling partner providing a quantitative yield of the
hydroxy sulfone adduct as a mixture of three of the four
possible diastereomers. Subsequent Ley oxidation17 and Na/
Hg amalgam reduction yielded the bisspirocyclization
precursor 16. It should be noted that only Ley’s TPAP
oxidant proved effective in this transformation; both Swern
Figure 2. Selected bisspiroketals natural products.
disclose the unprecedented effects of C ring substitution on
the bisspiroketalization.
2. Background and general strategy
Azaspiracid was first observed in the mid 1990s when
several individuals became severely ill from eating mussels
harvested from Killary Harbor, Ireland.10 Yasumoto and co-
workers determined the relative configuration of 3 using 2D
NMR techniques.8,9 The toxic effects of azaspiracid have
been shown to include serious injury to the digestive tracts,
liver, pancreas, thymus and spleen in mice.11 Subsequent to
Yasumoto’s original report, several derivatives of azaspir-
acid have been isolated from Western Ireland12 and there is
growing evidence of the spread of azaspiracid throughout
other regions of Europe.13 Also, recent reports appear to link
the presence of azaspiracid to an ubiquitous alga.14 The
Scheme 1. Retrosynthetic strategy.

R. G. Carter et al. / Tetrahedron 59 (2003) 8963–8974
8965
Another possible bisspirocyclization precursor lies in the
keto sulfone 15 (Scheme 4). We hoped that the C12 sulfone
might impart a directing effect on the spiroketalization at
C13. Treatment of the keto sulfone 15 under acidic
conditions (CSA, MeCN) again led to a single bisspiroketal
18 [epimeric (1:1 ratio) at C12]. The new spiroketal 18 was
correlated with the previously established cisoidal species
17 via reductive removal of the sulfone. In addition, both
C12 stereoisomers of the cisoidal sulfone-containing
bisspiroketal 18 were verified individually by COSY and
NOESY correlations. Interestingly, the use of an alternate
solvent (PhH) in the bisspirocyclization of 15 did result in
the production of a small amount (5–10%) of an additional
bisspiroketal which was later identified as the non-natural20
transoidal bisspiroketal 20 via COSY and ROESY corre-
lations. The yield of this new bisspiroketal 20 could be
significantly improved (up to 43%) via elimination of the C
ring pyran bridge followed by treatment under acidic
conditions (CSA, PhMe, 2788C to rt). The new spiroketal
20 was isolated as a single a-stereochemistry at C12. In
addition, the cisoidal spiroketal 18 was also produced in
49% yield as a 3:1 diastereomeric ratio at C12 (3:1 a–b).
While resubmission of the cisoidal ketal 18 to the same
reaction conditions (CSA, PhMe) did not result in the
formation of the transoidal spiroketal 20, acid treatment
(CSA, PhMe) of the transoidal ketal 20 did lead to slow
formation of the cisoidal product 18 (40% conversion to 18
Scheme 2. Julia coupling strategy.
oxidation18 and Dess–Martin periodinane19 led to complex
mixtures of products.
With the ketone 16 in hand, our attention turned to the
bisspirocyclization (Scheme 3). Treatment of 16 under a
variety of acidic and Lewis acidic conditions resulted in
formation of single bisspiroketal 17 which was shown to
possess the unwanted cisoidal stereochemistry. This result
was observed independently and concurrently6b by Forsyth
and co-workers on a similar D ring containing substrate.7a
Nicolaou later reported the same stereochemical outcome
on their related, D ring functionalized systems.5 This
cisoidal arrangement allows for placement of central furan
oxygen in the anomeric (or pseudo-anomeric) positions with
respect to both pyran rings.
1
by H NMR after 24 h).21 These results indicate that this
transformation is operating under kinetic control in which
the sulfone substitution is able to retard the rate of
equilibration versus the C12 unfunctionalized series. It is
Scheme 3. Thermodynamic bisspirocyclization of D ring substrate.
Scheme 4. Formation of the non-natural transoidal bisspiroketal.

8966
R. G. Carter et al. / Tetrahedron 59 (2003) 8963–8974
Scheme 5. Bisspirocyclization of D ring truncated substrate.
unlikely that the desired natural transoidal bisspiroketal will
be accessible on the fully functionalized ABCD ring system.
t-BuOH/PhMe (1/1), 16–20 h] led to formation of two
separate bisspiroketals: the desired natural transoidal
bisspiroketal 24 and unwanted cisoidal product 25 in
equal amounts. Resubmission of the cisoidal isomer 25 to
the identical reaction conditions led to the same equilibrium
ratio (1:1) of products. It should be pointed out that the 2D
NMR studies of spiroketal 24 appeared to indicate that the C
ring existed in the chair conformation shown above. This
conformation places both the C13 furan oxygen and the C14
methyl in equatorial positions as is observed in the solution
structure of azaspiracid (3).
Thwarted by the inherit preference of the fully substituted
system to yield the unwanted cisoidal isomers 17 and 18, we
were intrigued by the possibility of alternate methods to
control the stereochemical outcome. In particular, the
degree and nature of substitution on the C ring might play
an important role in bisspirocyclization. We surmised that
the one (or both) of the substituents at C16 and C17 might be
inhibiting the formation of the desired natural transoidal
bisspiroketal. This approach, while enticing, was not well
precedented.
Encouraged by the formation of the transoidal bisspiroketal
24, the synthesis of a bisspirocyclization precursor contain-
ing C17 substitution was undertaken (Scheme 6). From
inspection of the conformation of the transoidal bis-
spiroketal 24 (with both the C13 furan oxygen and the C14
methyl located in equatorial arrangements on the C ring),
substitution at C17 should be well-tolerated as the
To test this hypothesis, the bisspirocyclization precursor
236d possessing no substitution at C16 or C17 was
synthesized in an analogous fashion as described previously
(Scheme 5). We were pleased to observe that treatment of
the ketone 23 under acidic conditions [0.04 M CSA,
Scheme 6. Bisspirocyclization of C17 substrate.

R. G. Carter et al. / Tetrahedron 59 (2003) 8963–8974
8967
Scheme 7. Bisspirocyclization of the C16 substrate.
substituent would adopt an equatorial orientation. We chose
to explore a linear approach for the synthesis of this series.
Starting from the keto sulfone 22, bisspirocyclization using
CSA in MeCN followed by elimination of the pyran bridge
and oxidation yielded the aldehyde 26. This approach
allowed for the protection of the C13 carbonyl while freeing
the C17 position. Brown allylation of 26 yielded the desired
Resubmission of the cisoidal product 33 to the same
reaction conditions [0.04 M CSA, t-BuOH/PhMe (1/1), 16–
20 h] resulted in the thermodynamic 3:5 ratio of products
(32–33). Interestingly, treatment of ketone 31 at lower
temperatures (210 to 48C, 21 h) and lower molarity of CSA
(0.003 M) under otherwise identical reaction conditions (1:1
t-BuOH/PhMe) led to the cisoidal bisspirocycle 33 as the
predominate product by TLC. Gratifyingly, further warming
of the reaction to room temperature for an additional 48 h,
resulted in the previously observed (3:5 ratio of 32–33)
equilibrium mixture. It would appear from this experiment
that the cisoidal bisspirocycle 33 is the result of kinetic
control while the transoidal bisspiroketal 32 is available
under equilibrating, thermodynamic conditions.
1
C17 stereocenter as a single diastereoisomer by H NMR.
Removal of the sulfone and submission to the same acidic
conditions [0.04 M CSA, t-BuOH/PhMe (1/1), 14.5 h]
yielded solely the cisoidal bisspiroketal 28 in 76% yield.
With the success of the C16/C17 unsubstituted series 23 and
the disappointment of the C17 substituted substrate 27, the
remaining C16 series had to be synthesized (Scheme 7).
Using a convergent approach and the key Julia coupling, the
ketone 316e was constructed. Treatment of 31 under the
standard conditions [0.04 M CSA, t-BuOH/PhMe (1/1),
16–20 h] once again yielded two bisspiroketals, the desired
natural transoidal product 32 and the unwanted cisoidal
isomer 33, in a 3:5 ratio (32–33). The transoidal
bisspiroketal 32 appears to adopt an alternate C ring chair
conformation based on the NMR data versus both
azaspiracid (3) and the D ring truncated system 24. The
proposed conformation for the transoidal bisspiroketal 32
places the C13 furan oxygen in an anomeric position while
locating the C16 benzyloxy substituent in an equatorial
4. Discussion
The bisspiroketalization results clearly show that the degree
and nature of substitution at C16 and C17 has a dramatic
impact on the stereochemical outcome of the bisspiro-
cyclization. Based on the reported data, a working
hypothesis can be put forth: the formation of the desired
natural transoidal bisspiroketal 6 is blocked by a C17
substituent while C16 substitution appears to have little
impact on the stereochemical outcome of the transform-
ation.23 One possible explanation for the C17 blocking effect
may involve the potential transition states shown below in
Scheme 8 in which the approaching hydroxyl function
should enter in a pseudo-axial fashion during the C13
spiroketal formation.24 Given this requirement, access of the
necessary transition states 35a-b and/or 39a-b would seem
prohibitively high in energy due to the developing C13–C17
´
arrangement. Miljkovıc and Srivastava independently
studied the controlling influence of similar substitution
patterns in pyran ring systems. They observed a slight
computational preference for placement of the non-
glycoside linkage in an equatorial arrangement while
locating the acetal in an axial conformation.22

8968
R. G. Carter et al. / Tetrahedron 59 (2003) 8963–8974
Scheme 8. Possible explanation for the observed stereochemical preference.
diaxial interaction. It should be noted that a related
explanation is possible involving the oxonium ions 41–
44. In this case, a similar argument could be offered for the
inability to access oxonium ions 43a-b and 44a-b. This
kinetic-based argument necessitates that true cisoidal/
transoidal equilibrating conditions are not feasible on
systems containing C17 substituents under the prescribed
reaction conditions. The inability to observe any natural
transoidal bisspiroketal formation with C17 containing
compounds 16 and 27 appears to support this model.
In addition, the successful construction of the natural
transoidal bisspiroketals 24 and 32 under thermodynamic-
ally equilibrating conditions while observing that the
cisoidal bisspiroketal 33 is the predominate kinetic product
under the non-equilibrating conditions [0.003 M CSA, 210
to 48C, t-BuOH/PhMe (1:1)] provides further beneficial data
toward this explanation.
under equilibrating conditions on a species possessing a
C16 benzyloxy substituent. The application of this strategy
to the total synthesis of azaspiracid will be reported in due
course.
6. Experimental
6.1. General
6.1.1. Keto sulfone 15. To a stirred solution of 86a (224 mg,
0.371 mmol) in THF (1.5 mL) at 2788C was added LDA25
(420 mL, 0.42 mmol, 1.0 M in THF/hexanes) dropwise.
After 20 min, a precooled solution of 146a (134.4 mg,
0.445 mmol) in THF (0.4 mL) was added via cannula to the
yellow sulfone solution. An additional portion of THF
(2£100 mL) was added to rinse the aldehyde flask. After
25 min, the reaction removed from the cooling bath. After
2 min, the reaction was quenched with saturated aqueous
NH4Cl (25 mL), allowed to warm to room temperature and
extracted with Et2O (5£25 mL). The dried (MgSO4) extract
was concentrated in vacuo and purified by chromatography
over silica gel, eluting with 5–25% EtOAc/hexanes, to give
sequentially the least polar hydroxy sulfone 45a (64 mg,
0.071 mmol, 19%) followed by the more polar hydroxy
sulfones 45b and 45c (272 mg, 0.300 mmol, 81%) as a
colorless oils.
5. Conclusion
The systematic construction of bisspirocyclization pre-
cursors has been accomplished possessing the four possible
substitution patterns at C16 and C17 via a Julia coupling
strategy. Their subsequent bisspirocyclization led to the
observations of the dramatic and unprecedented controlling
influence of the C17 substituent. A possible working model
has been put forth to explain the observed results. Finally,
the successful construction of the natural transoidal
bisspiroketal stereochemistry has been accomplished
To a stirred solution of 45b/45c (39.8 mg, 0.0439 mmol)
˚
and powdered 4 A molecular sieves (200 mg) in CH2Cl2

R. G. Carter et al. / Tetrahedron 59 (2003) 8963–8974
8969
(0.7 mL) was added sequentially NMO (8.5 mg,
0.072 mmol) and TPAP (3.8 mg, 0.0108 mmol). An
additional portion of TPAP (3.7 mg, 0.0105 mmol) was
added during the course of the reaction. After 1 h, the
reaction was diluted with 30% EtOAc/hexanes (5 mL),
filtered through a small plug of silica gel (30% EtOAc/
hexanes rinse), concentrated in vacuo and purified by
chromatography over silica gel, eluting with 5–30%
EtOAc/hexanes, to give 15 as a colorless oil (32.0 mg,
0.0354 mmol, 81%). [a]2D3¼212.68 (c 1.65, CHCl3); IR
0.0021 mmol) in t-BuOH (1.6 mL) at room temperature was
added PPTS (6.8 mg, 0.027 mmol). After 5 h, the reaction
was quenched with solid NaHCO3 (100 mg), diluted with
sat. aq. NaHCO3 (50 mL) and extracted with Et2O
(3£50 mL). The dried (MgSO4) extract was concentrated
in vacuo and purified by chromatography over silica gel,
eluting with 5–20% EtOAc/hexanes, to give 17 as a
colorless oil (6.6 mg, 0.011 mmol, 51%). 1H NMR
(300 MHz, CDCl3) d 7.65–7.68 (m, 4H), 7.36–7.45 (m,
6H), 5.95–5.97 (m, 1H), 5.50–5.75 (m, 3H), 5.17 (dd,
J¼4.2, 5.5 Hz, 1H), 4.35–4.42 (m, 1H), 4.27–4.30 (m, 1H),
3.97 (m, 1H), 3.66 (t, J¼6.3 Hz, 2H), 3.39 (s, 3H), 1.83–
2.30 (m, 11H), 1.59–1.69 (m, 4H), 1.05 (s, 9H), 0.87 (d,
J¼6.6 Hz, 3H); 13C NMR (75 MHz, CDCl3) d 135.8, 134.3,
132.5, 130.5, 129.7, 128.6, 128.2, 127.8, 109.3, 105.6,
105.4, 75.1, 72.3, 63.5, 55.8, 41.6, 37.5, 32.3, 31.8, 31.1,
30.4, 29.9, 29.0, 27.1, 19.4, 15.9; HRMS (FABþ) calcd for
C37H50O6SiLi (MþLi) 625.3537, found 625.3516.
1
(neat) 3069, 2932, 1719, 1310, 111, 703 cm21; H NMR
(300 MHz, CDCl3) d 7.75–7.82 (m, 2H), 7.65–7.68 (m,
5H), 7.48–7.58 (m, 4H), 7.35–7.45 (m, 6H), 6.03 (m, 1H of
1 diastereomer), 5.93 (m, 1H of 1 diastereomer) 5.36–5.66
(m, 3H), 5.13 (t, J¼4.4 Hz, 1H of 1 diastereomer), 5.02 (dd,
J¼2.7, 5.3 Hz,1H of 1 diastereomer), 4.68 (t, J¼6.8 Hz, 1H
of 1 diastereomer), 4.61 (d, J¼10.2 Hz, 1H of 1 dia-
stereomer), 4.33–4.40 (m, 1H), 4.14–4.19 (m, 1H), 3.95–
4.01 (m, 1H), 3.65 (dd, J¼6.2, 11.3 Hz, 2H), 3.36 (s, 3H of 1
diastereomer), 3.33 (s, 3H of 1 diastereomer), 3.20–3.30
(m, 1H), 3.10 (s, 3H of 1 diastereomer), 3.04 (s, 3H of 1
diastereomer), 2.53 (dd, J¼10.6, 13.5 Hz, 1H of 1
diastereomer), 3.30–3.33 (m, 1.5H), 1.85–2.20 (m, 8H),
1.60–1.66 (m, 3H), 1.30–1.45 (m, 1H), 1.25 (d, J¼6.6 Hz,
3H of 1 diastereomer), 1.17 (d, J¼7.2 Hz, 3H of 1
diastereomer), 1.06 (s, 9H of 1 diastereomer), 1.05 (s, 9H
of 1 diastereomer), 0.99 (t, J¼8.0 Hz, 9H of 1 dia-
stereomer), 0.93 (t, J¼8.0 Hz, 9H of 1 diastereomer), 0.64
(q, J¼8.0 Hz, 6H of 1 diastereomer), 0.57 (q, J¼8.0 Hz, 6H
of 1 diastereomer); 13C NMR (75 MHz, CDCl3) d 205.3,
204.9, 137.2, 136.8, 135.7, 134.22, 134.16, 132.8, 132.6,
130.3, 129.9, 129.81, 129.75, 129.5, 129.1, 127.8, 104.1,
103.7, 97.2, 96.5, 78.4, 78.2, 72.6, 70.0, 69.9, 68.9, 68.7,
63.6, 63.4, 55.3, 55.2, 49.3, 49.2, 45.3, 45.1, 43.8, 35.7,
34.8, 32.3, 32.1, 31.4, 30.6, 30.42, 30.36, 29.9, 28.8, 27.0,
19.4, 16.0, 15.1, 7.1, 7.0, 5.0, 4.9; HRMS (FABþ) calcd for
C50H72O9SSi2Li (MþLi) 911.4596, found 911.4598.
6.1.4. Spiroketal sulfone 18. To a stirred solution of 15
(42 mg, 0.046 mmol) in MeCN (5.2 mL) at 08C was added
CSA (5.6 mg, 0.024 mmol). After 1 h, the reaction was
allowed to warm to room temperature. After 6.5 h, the
reaction was quenched with solid NaHCO3 (500 mg),
diluted with saturated aqueous NaHCO3 (25 mL), and
extracted with Et2O (4£25 mL). The dried (MgSO4) extract
was concentrated in vacuo and purified by chromatography
over silica gel, eluting with 5–25% EtOAc/hexanes, to give
18 as
a colorless oil (25 mg, 0.032 mmol, 72%).
[a]2D3¼238.88 (c 1.20, CHCl3); IR (neat) 3068, 2925,
2855, 1463, 1106 cm21; 1H NMR (300 MHz, CDCl3) d 7.93
(d, J¼7.4 Hz, 2H of 1 diastereomer), 7.90 (t, J¼7.4 Hz, 2H
of 1 diastereomer), 7.61–7.68 (m, 5H), 7.54 (t, J¼7.4 Hz,
2H of 1 diastereomer), 7.50 (t, J¼7.4 Hz, 2H of 1
diastereomer), 7.35–7.46 (m, 6H), 6.05–6.08 (m, 1H of 1
diastereomer), 5.92–5.97 (m, 1H of 1 diastereomer), 5.40–
5.70 (m, 3H), 5.16 (t, J¼4.8 Hz, 1H of 1 diastereomer), 4.81
(t, J¼4.8 Hz, 1H of 1 diastereomer), 4.30–4.35 (m, 1H of 1
diastereomer), 4.27 (dd, J¼7.2, 13.1 Hz, 1H of 1 dia-
stereomer), 4.09–4.16 (m, 1.5H), 3.88 (d, J¼1.8 Hz, 1H of
1 diastereomer), 3.82 (dd, J¼8.2, 12.3 Hz, 1H of 1
diastereomer), 3.60 (dd, J¼6.2, 13.9 Hz, 2H), 3.41 (s, 3H
of 1 diastereomer), 3.35 (s, 3H of 1 diastereomer), 2.68
(t¼12.3 Hz, 1H of 1 diastereomer), 2.40–2.55 (m, 2H), 2.31
(dd, J¼7.2, 12.3 Hz, 1H of 1 diastereomer), 1.89–2.19 (m,
7H), 1.56–1.85 (m, 7H), 1.15 (d, J¼6.4 Hz, 3H of 1
diastereomer), 1.04 (s, 9H), 0.93 (d, J¼6.7 Hz, 3H of 1
diastereomer); 13C NMR (75 MHz, CDCl3) d 140.6, 138.7,
135.7, 134.1, 133.9, 133.7, 133.1, 132.5, 130.8, 130.2,
129.9, 129.7, 129.2, 128.6, 127.8, 127.4, 126.7, 108.8,
106.4, 105.4, 105.2, 102.4, 74.6, 74.0, 73.6, 73.3, 72.3, 70.1,
69.7, 66.4, 66.1, 63.34, 63.30, 60.6, 55.9, 41.3, 40.7, 40.1,
38.5, 32.13, 32.10, 30.9, 30.2, 29.9, 29.8, 29.1, 28.92, 28.87,
28.3, 27.0, 21.3, 19.4, 17.2, 15.6, 15.5, 14.4; HRMS
(FABþ) calcd for C43H54O8SSiLi (MþLi) 765.3469,
found 765.3481.
6.1.2. Ketone 16. To a stirred solution of 15 (33 mg,
0.037 mmol) in THF (730 mL) and MeOH at 2108C
(2.2 mL) was added sequentially Na2HPO4 (21 mg) and
Na/Hg (500 mg, 5% Na). After 45 min, the reaction was
poured directly on a small plug of dry silica gel (40%
EtOAc/hexanes rinse), concentrated in vacuo and purified
by chromatography over silica gel, eluting with 7–30%
EtOAc/hexanes, to give 16 as a colorless oil (25 mg,
0.033 mmol, 88%). [a]2D3¼215.38 (c 1.18, CHCl3); IR
(neat) 2952, 1709, 1463, 1427, 1107 cm21 1H NMR
;
(300 MHz, CDCl3) d 7.66 (dd, J¼1.5, 7.5 Hz, 4H), 7.35–
7.45 (m, 6H), 5.95–6.01 (m, 1H), 5.62–5.73 (m, 2H), 5.51
(dd, J¼6.5, 15.5 Hz, 1H), 5.05 (t, J¼4.7 Hz, 1H), 4.24–4.37
(m, 2H), 3.91 (dt, J¼3.8, 10.0 Hz, 1H), 3.67 (t, J¼6.3 Hz,
2H), 3.32 (s, 3H), 3.25 (s, 3H), 2.68–2.77 (m, 1H), 2.50–
2.57 (m, 1H), 1.85–2.24 (m, 6H), 1.60–1.71 (m, 2H), 1.45
(ddd, J¼3.6, 6.8, 11.5 Hz, 1H), 1.11 (d, J¼6.9 Hz, 3H) 1.05
(s, 9H), 0.95 (t, J¼7.9 Hz, 9H), 0.59 (q, J¼7.9 Hz, 6H); 13C
NMR (75 MHz, CDCl3) d 214.1, 135.8, 134.2, 132.7, 130.3,
129.7, 128.6, 128.4, 128.0, 104.0, 98.2, 78.8, 72.6, 68.9,
63.5, 55.3, 48.4, 43.8, 43.6, 36.1, 32.4, 32.1, 30.6, 29.6,
28.9, 27.1, 19.4, 16.6, 7.0, 4.9; HRMS (FABþ) calcd for
C44H68O7Si2Li (MþLi) 771.4664, found 771.4682.
6.1.5. Enol ether 19. To a stirred solution of 18 (24.0 mg,
0.0317 mmol) in THF (0.7 mL) at 2788C was added n-BuLi
(16 mL, 0.040 mmol, 2.5 M in hexanes) dropwise. An
additional portion of n-BuLi (12 mL, 0.030 mmol, 2.5 M
in hexanes) was added during the course of the reaction.
After 1.25 h, the reaction was quenched with solid silica gel.
6.1.3. Spiroketal 17. To a stirred solution of 16 (16.0 mg,

8970
R. G. Carter et al. / Tetrahedron 59 (2003) 8963–8974
The reaction was allowed to warm to ambient temperature,
diluted with saturated aqueous NH4Cl (25 mL) and
extracted with Et2O (4£25 mL). The dried (MgSO4) extract
was concentrated in vacuo and purified by chromatography
over silica gel, eluting with 10–60% EtOAc/hexanes, to
give 19 as a colorless oil (19.9 mg, 0.0263 mmol, 83%).
[a]2D3¼þ29.18 (c 0.99, CHCl3); IR (neat) 3518, 3069, 2928,
To a stirred solution of 46 (40.0 mg, 0.0487 mmol) in
˚
CH2Cl2 (0.8 mL) with powdered 4 A mol. sieves
(<200 mg) was sequentially added NMO (13 mg,
0.11 mmol) and TPAP (6.8 mg, 0.019 mmol) at room
temperature An additional portion of TPAP (6 mg,
0.017 mmol) was added during the course of the reaction.
After 1.5 h, the reaction was diluted with 30% EtOAc/
hexanes (10 mL), filtered through a small plug of silica gel
(30% EtOAc/hexanes rinse), and concentrated in vacuo.
The resultant oil was purified by chromatography over silica
gel, eluting with 5–20% EtOAc/hexanes, to give 22 (29 mg,
0.035 mmol, 73%) as a colorless oil: [a]2D3¼þ17.38 (c 3.45,
CHCl3); IR (neat) 3070, 2952, 17.16, 1310, 1111 cm21; 1H
NMR (300 MHz, CDCl3) 7.76 (d, J¼7.5 Hz, 2H), 7.65–
7.68 (m, 5H), 7.51–7.59 (m, 2H), 7.36–7.46 (m, 6H), 6.01–
6.08 (m, 1H of a diastereomer), 5.01–5.96 (m, 1H of a
diastereomer), 5.31–5.64 (m, 3H), 4.61 (dd, J¼3.4, 3.4 Hz,
1H of a diastereomer), 4.47 (d, J¼9.0 Hz, 1H of a
diastereomer), 4.12–4.22 (m, 1H) 3.57–3.68 (m, 4H),
3.11 (s, 3H of a diastereomer), 3.06 (s, 3H of a
diastereomer), 2.92–3.03 (m, 1H), 2.60 (dd, J¼10.0,
13.8 Hz, 1H of a diastereomer), 2.27–2.38 (m, 1H of a
diastereomer), 2.25 (d, J¼6.2 Hz, 1H), 1.80–2.20 (m, 7H),
1.50–1.68 (m, 3H), 1.17 (d, J¼6.7 Hz, 3H of a dia-
stereomer), 1.11 (d, J¼7.2 Hz, 3H of a diastereomer), 1.06
(s, 9H of a diastereomer), 1.05 (s, 9H of a diastereomer),
0.93–0.98 (m, 9H), 0.55–0.65 (m, 6H); 13C NMR (75 MHz,
CDCl3) d 204.8, 204.6, 136.7, 135.8, 134.4, 134.3, 132.8,
132.3, 130.3, 129.8, 139.6, 129.2, 129.1, 127.9, 127.8,
127.7, 97.4, 96.5, 69.8, 69.3, 69.1, 68.8, 63.4, 63.0, 62.8,
49.4, 48.0, 47.6, 34.7, 34.2, 32.2, 32.1, 30.8, 30.5, 30.3,
29.9, 28.9, 28.8, 28.7, 27.1, 19.4, 16.4, 14.9, 7.1, 4.6.
1
2859, 1624 cm21; H NMR (300 MHz, C6D6) d 7.93 (m,
2H), 7.75 27.78 (m, 4H), 7.22–7.25 (m, 6H), 6.94–6.97
(m, 3H), 5.46–5.68 (m, 3H), 5.18 (dt, J¼1.1, 9.9 Hz, 1H),
4.44 (dt, J¼4.2, 11.5 Hz, 1H), 4.34–4.39 (m, 1H), 4.12–
4.18 (m, 1H and OH), 3.62 (t, J¼6.3 Hz, 2H), 3.25 (s, 3H),
3.10 (d, J¼15.0 Hz, 1H), 2.93 (d, J¼15.0 Hz, 1H), 1.70–
2.30 (m, 10H), 1.48–1.67 (m, 5H), 1.22 (d, J¼6.9 Hz, 3H),
1.17 (s, 9H); 13C NMR (75 MHz, C6D6) d 170.7, 144.1,
136.7, 135.0, 133.1, 130.7, 130.6, 130.4, 129.9, 129.6,
128.9, 129.6, 127.8, 126.8, 127.8, 126.8, 109.8, 107.3,
105.1, 79.4, 72.5, 71.1, 64.9, 55.9, 44.2, 43.3, 34.2, 32.9,
30.2, 29.6, 29.3, 27.8, 20.1, 19.7; HRMS (FABþ) calcd for
C43H54O8SSiLi (MþLi) 765.3469, found 765.3565.
6.1.6. Spiroketal 20. To a stirred solution of 19 (13.0 mg,
0.0172 mmol) in PhMe (2.4 mL) at 2788C was added CSA
(36 mg, 0.155 mmol). After 10 min, the reaction was
allowed to warm to ambient temperature over a period of
35 min. After an additional 90 min, the reaction was
quenched with solid NaHCO3, diluted with 33% EtOAc/
hexanes (10 mL), filtered through a small plug of silica gel
(33% EtOAc/hexanes rinse) and concentrated in vacuo. The
crude oil was purified by chromatography over silica gel,
eluting with 5–40% EtOAc/hexanes, to give sequentially 18
(6.3 mg, 0.0083 mmol, 49%) and 20 (5.6 mg, 0.074 mmol,
43%) as colorless oils. 20: [a]2D3¼245.58 (c 0.43, CHCl3);
1
IR (neat) 3070, 2925, 2854, 1460, 1308 cm21; H NMR
6.1.8. Ketone 23. To a stirred solution of 22 (86 mg,
0.103 mmol) in THF (0.6 mL) and MeOH (1.8 mL) at
2108C was added Na2HPO4 (70.6 mg, 0.493 mmol)
followed by Na/Hg (330 mg, 0.712 mmol, 5% Na). After
75 min, the reaction was diluted with 35% EtOAc/hexanes
(10 mL), filtered through a small plug of silica gel (35%
EtOAc/hexanes rinse), and concentrated in vacuo to give
crude 23 (0.103 mmol) as a colorless oil: 1H NMR
(300 MHz, CDCl3) d 7.67 (dd, J¼0.9, 6.2 Hz, 4H), 7.36–
7.45 (m, 6H), 5.96–6.02 (m, 1H), 5.69 (dt, J¼6.5, 15.5 Hz,
1H), 5. 63 (d, J¼9.9 Hz, 1H), 5.52 (dd, J¼6.6, 15.5 Hz, 1H),
4.22–4.29 (m, 1H), 3.67 (t, J¼6.2 Hz, 2H), 3.58 (t,
J¼5.8 Hz, 2H), 2.40–2.60 (m, 2H), 1.80–2.15 (m, 6H),
1.52–1.70 (m, 5H), 1.30–1.50 (m, 2H), 1.05–1.08 (m,
12H), 0.95 (t, J¼7.9 Hz, 9H), 0.58 (q, J¼7.9 Hz, 6H);
HRMS (FABþ) calcd for C40H61O4Si2 (Mþ2MeOH)
661.4108, found 661.4124.
(300 MHz, C6D6) d 7.89 (dd, J¼2.0, 8.0 Hz, 2H), 7.74–7.78
(m, 4H), 7.22–7.25 (m, 6H), 6.86–6.94 (m, 3H), 5.39–5.64
(m, 3H), 5.32 (d, J¼10.0 Hz, 1H), 5.20 (dd, J¼4.0, 5.6 Hz,
1H), 4.29 (dd, J¼6.6, 12.5 Hz, 1H), 4.09 (dd, J¼2.0, 5.0 Hz,
1H), 3.93–3.96 (m, 1H), 3.80–3.82 (m, 1H), 3.59 (t,
J¼6.3 Hz, 2H), 3.33 (s, 3H), 2.70 (dd, J¼12.2, 12.5 Hz,
1H), 2.52 (dd, J¼6.5, 12.2 Hz, 1H), 2.42–2.49 (m, 1H),
1.80–2.14 (m, 8H), 1.52–1.70 (m, 2H), 1.16 (s, 9H), 0.91
(d, J¼6.0 Hz, 3H); 13C NMR (125 MHz, C6D6) d 140.9,
136.4, 134.7, 133.6, 131.8, 131.2, 130.4, 130.2, 129.9,
129.3, 128.9, 128.2, 108.1, 105.8, 103.1, 75.6, 73.6, 73.1,
65.9, 63.9, 55.8, 41.5, 38.4, 32.7, 29.2, 28.9, 27.5, 23.5,
19.9, 16.0, 14.8; HRMS (FABþ) calcd for C43H54O8SSiLi
(MþLi) 765.3469, found 765.3466.
6.1.7. Keto sulfone 22. To a stirred solution of 86a (166 mg,
0.275 mmol) in THF (1.5 mL) at 2788C was added LDA24
(310 mL, 0.31 mmol, 1 M in THF/hexanes) dropwise via a
syringe. After 20 min, a precooled solution of the aldehyde
216d (75 mg, 0.344 mmol) in THF (0.3 mL) was added
rapidly via cannula to the yellow sulfone solution. After
25 min, the reaction was removed from the cooling bath.
After an additional 2 min, the reaction was quenched with
sat. aq. NH4Cl (25 mL) and extracted with EtOAc
(4£25 mL). The dried (MgSO4) extract was concentrated
in vacuo and purified by chromatography over silica
gel, eluting with 10–28% EtOAc/hexanes, to give the
hydroxy sulfone 46 (182 mg, 0.0.22 mmol, 81%) as a
colorless oil.
6.1.9. Spirocycles 24 and 25. To a stirred solution of crude
23 (0.103 mmol) in PhMe (7 mL) and t-BuOH (7 mL) was
added CSA (123 mg, 0.529 mmol). After 19 h, the reaction
was quenched with solid NaHCO3 (500 mg). After 10 min,
the solution was diluted with sat. aq. NaHCO3 (50 mL) and
extracted with Et2O (4£100 mL). The dried (MgSO4)
extract was concentrated in vacuo and purified by
chromatography over silica gel, eluting with 3–12%
EtOAc/hexanes, to give more polar transoidal 24 and less
polar cisoidal 25 (38 mg, 0.070 mmol, 68% over the 2 steps)
as colorless oils. transoidal 24: [a]2D3¼213.08 (c 0.185,
1
CHCl3); IR (neat) 2931, 2858, 1428, 1111, 702 cm21; H

R. G. Carter et al. / Tetrahedron 59 (2003) 8963–8974
8971
NMR (300 MHz, CDCl3) 7.65–7.68 (m, 4H), 7.34–7.43
(m, 6H), 5.94–6.0 (m, 1H), 5.62–5.72 (m, 2H), 5.50 (dd,
J¼6.0, 15.4 Hz, 1H), 4.36–4.43 (m, 1H), 3.92 (ddd, J¼3.0,
11.6, 11.6 Hz, 1H), 3.66 (t, J¼6.4 Hz, 2H), 3.57 (dt, J¼4.3,
11.2 Hz, 1H), 1.83–2.20 (m, 10H), 1.56–1.79 (m, 5H), 1.04
(s, 9H), 1.01 (d, J¼7.2 Hz, 3H); 13C NMR (75 MHz,
CDCl3) d 135.8, 134.2, 132.3, 130.7, 129.7, 128.2, 127.8,
110.0, 104.4, 69.0, 63.4, 62.8, 37.1, 36.5, 32.6, 32.1, 30.3,
28.9, 27.5, 27.0, 21.5, 19.4, 15.6; HRMS (FABþ) calcd for
C34H47O4Si (MþH) 547.3244, found 547.3228. cisoidal 25:
[a]2D3¼265.38 (c 0.19, CHCl3); IR (neat) 2960, 2930, 2855,
70%) followed by C10-epi 48 (5.0 mg, 0.0.073, 10%) as
colorless oils. [a]2D3¼þ62.48 (c 180, CHCl3); IR (neat)
1
3444, 2929, 2856, 1624, 1109 cm21; H NMR (300 MHz,
CDCl3) d 7.87 (d, J¼7.6 Hz, 2H), 7.66 (d, J¼6.7 Hz, 4H)
7.49–7.60 (m, 3H), 7.36–7.41 (m, 6H), 6.06–6.12 (m, 1H),
5.69 (d, J¼10.9 Hz, 1H), 5.65 (dt, J¼6.6, 15.7 Hz, 1H), 5.46
(dd, J¼6.0, 15.7 Hz, 1H), 4.35–4.43 (m, 1H), 3.65 (t,
J¼6.2 Hz, 2H), 3.55 (t, J¼5.7 Hz, 2H), 2.91 (d, J¼15.1 Hz,
1H), 2.83 (d, J¼15.1 Hz, 1H), 2.00–2.17 (m, 4H), 1.49–
1.67 (m, 7H), 1.14 (d, J¼6.9 Hz, 3H), 1.05 (s, 9H); 13C
NMR (75 MHz, CDCl3) d 170.3, 142.5, 135.8, 134.2, 133.1,
133.0, 130.4, 129.8, 129.4, 129.3, 127.8, 127.0, 126.0,
108.7, 106.4, 70.6, 63.4, 62.9, 42.1, 32.0, 31.4, 30.6, 30.4,
28.8, 27.0, 19.4, 18.6, 15.5; HRMS (FABþ) calcd for
C40H51O6SSi (MþH) 687.3176, found 687.3158.
1
1467, 1110 cm21; H NMR (300 MHz, CDCl3) 7.64–7.68
(m, 4H), 7.34–7.45 (m, 6H), 5.94–6.00 (m, 1H), 5.53–5.75
(m, 3H), 4.42–4.49 (m, 1H), 3.84 (ddd, J¼2.2, 11.1,
11.1 Hz, 1H), 3.60–3.67 (m, 1H), 3.66 (t, J¼6.2 Hz, 2H),
1.83–2.26 (m, 10H), 1.47–1.72 (m, 5H), 1.04 (s, 9H), 0.87
(d, J¼6.2 Hz, 3H); 13C NMR (75 MHz, CDCl3) d 135.8,
134.2, 132.4, 130.6, 129.7, 128.8, 128.0, 127.8, 109.6,
105.3, 70.1, 63.5, 62.0, 37.8, 37.5, 35.5, 32.1, 30.2, 28.9,
28.6, 27.0, 19.4, 16.8; HRMS (FABþ) calcd for C34H47O4Si
(MþH) 547.3244, found 547.3252.
6.1.12. Aldehyde 26. To a stirred solution of 48 (21.6 mg,
˚
0.0320 mmol) in CH2Cl2 (1.0 mL) with powdered 4 A mol.
sieves (250 mg) was sequentially added TPAP (2.5 mg,
6.3 mmol) and NMO (9.0 mg, 0.077 mmol) at room
temperature After 30 min, the reaction was diluted with
33% EtOAc/hexanes (10 mL), filtered through a small plug
of silica gel (33% EtOAc/hexanes rinse), and concentrated
in vacuo to give 26 (21.0 mg, 0.031 mmol, 97%) as a
colorless oil: [a]2D3¼þ52.48 (c 1.03, CHCl3); IR (neat) 2928,
6.1.10. Sulfone spirocycle 47. To a stirred solution of 22
(68 mg, 0.083 mmol) in MeCN (9 mL) at room temperature
was added CSA (8.8 mg, 0.038 mmol). After 2 h, the
reaction was quenched with solid NaHCO3 (250 mg). The
mixture was diluted with 35% EtOAc/hexanes (50 mL),
filtered through a small plug of silica gel (35% EtOAc/
hexanes rinse), and concentrated in vacuo. The resultant oil
was purified by chromatography over silica gel, eluting with
6–40% EtOAc/hexanes, to give 47 (52 mg, 0.076 mmol,
91%) as a colorless oil: IR (neat) 3069, 3046, 2930, 2857,
1
28.55, 1724, 1627, 1110, 599 cm21; H NMR (300 MHz,
CDCl3) d 9.64 (s, 1H), 7.87 (d, J¼6.8 Hz, 2H), 7.65 (dd,
J¼1.5, 6.0 Hz, 4H), 7.50–7.65 (m, 3H), 7.35–7.43 (m, 6H),
6.07–6.13 (m 1H), 5.68 (d, J¼10.5 Hz, 1H), 5.60 (dt,
J¼6.5, 15.7 Hz, 1H), 5.41, (dd, J¼6.5, 15.7 Hz, 1H), 4.29–
4.36 (m, 1H), 3.64 (t, J¼6.2 Hz, 2H), 3.51–3.59 (m, 1H),
2.84 (s, 2H), 2.32–2.43 (m, 2H), 1.95–2.15 (m, 4H), 1.73–
1.83 (m, 2H), 1.58–1.68 (m, 4H), 1.15 (d, J¼6.8 Hz, 3H),
1.05 (s, 9H); 13C NMR (75 MHz, CDCl3) d 201.8, 170.0,
135.8, 134.1, 133.2, 130.7, 129.8, 129.4, 129.2, 127.8,
127.0, 125.7, 109.9, 106.6, 70.9, 63.4, 42.0, 41.0, 41.7, 32.0,
30.0, 29.9, 28.8, 27.1, 26.3, 19.4, 18.5; HRMS (FABþ)
calcd for C40H49O6SSi (MþH) 685.3019, found 685.3020.
1471, 1446, 1427, 1307, 1150, 1111 cm21 1H NMR
;
(300 MHz, CDCl3) d 7.88–7.93 (m, 2H), 7.59–7.67 (m,
5H), 7.48–7.55 (m, 2H), 7.32–7.44 (m, 6H), 6.03–6.09 (m,
1H of a diastereomer), 5.88–5.93 (m, 1H of a diastereomer),
5.41–5.70 (m, 3H), 4.33–4.47 (m, 1H), 4.15 (dd, J¼7.0,
13.2 Hz, 1H of a diastereomer), 3.35–3.82 (m, 2H and 1H
of a diastereomer), 2.70 (dd, J¼12.4, 12.4 Hz, 1H of a
diastereomer), 2.52–2.60 (m 1H of a diastereomer), 2.3–
2.43 (m, 1H), 1.90–2.20 (m, 5H), 1.40–1.80 (m, 7H), 1.14
(d, J¼5.9 Hz, 3H of a diastereomer), 1.04 (s, 9H of a
diastereomer), 1.03 (s, 9H of a diastereomer), 0.90 (d,
J¼6.7 Hz, 3H of a diastereomer); 13C NMR (75 MHz,
CDCl3) d 141.0, 139.3, 135.7, 134.2, 133.9, 133.8, 133.2,
132.1, 130.8, 130.1, 130.0, 129.8, 129.4, 128.4, 128.2,
127.8, 127.3, 126.8, 109.9, 107.3, 102.53, 102.46, 73.8,
70.8, 69.3, 66.6, 63.5, 63.4, 61.3, 41.2, 38.7, 35.4, 34.8,
32.1, 32.0, 30.1, 29.9, 29.4, 28.9, 28.8, 28.5, 27.1, 26.0,
25.3, 19.4, 18.2, 16.4, 15.5; HRMS (FABþ) calcd for
C40H51O6SSi (MþH) 687.3176, found 687.3154.
6.1.13. Homoallylic alcohol 49. To a stirred solution of 26
(20.0 mg, 0.0292 mmol) in Et2O (1.1 mL) at 21008C was
added dropwise precooled (Ipc)2Ballyl26 (140 mL,
0.035 mmol, 0.25 M in pentane) via syringe. After 30 min,
the reaction was quenched with MeOH (50 mL) and warmed
to room temperature The solution was further quenched
with aq. phosphate buffer (800 mL, pH 7) and H2O2
(200 mL, 30% in H2O). After 30 min, the solution was
diluted with sat. aq. NaCl (25 mL) and extracted with Et2O
(4£25 mL). The dried (MgSO4) extract was concentrated in
vacuo and purified by chromatography over silica gel,
eluting with 7–40% EtOAc/hexanes, to give 49 (15.0 mg,
0.021 mmol, 71%) as a colorless oil. [a]2D3¼þ14.18 (c 0.68,
CHCl3); IR (neat) 3566, 2928, 2855, 2625, 1427,
6.1.11. Sulfone enol ether 48. To a stirred solution of 47
(51 mg, 0.75 mmol) in THF (2.3 mL) at 2788C was added
dropwise n-BuLi (35 mL, 0.0875 mmol, 2.5 M in hexanes).
An additional portion of n-BuLi (6 mL, 0.015 mmol, 2.5 M
in hexanes) was added during the course of the reaction.
After 70 min, the reaction was quenched with solid silica gel
(500 mg). After 5 min, the mixture was diluted with sat. aq.
NH4Cl (20 mL) and extracted with Et2O (4£25 mL). The
dried (MgSO4) extract was concentrated in vacuo and
purified by chromatography over silica gel, eluting with 6–
50% EtOAc/hexanes, to give 48 (35.8 mg, 0.0.22 mmol,
1
1110 cm21; H NMR (300 MHz, CDCl3) d 7.85–7.90 (m,
2H), 7.65 (dd, J¼1.4, 7.4 Hz, 4H) 7.47–7.61 (m, 3H), 7.34–
7.45 (m, 6H), 6.06–6.12 (m, 1H), 5.70–5.82 (m, 1H), 5.69
(d, J¼9.5 Hz, 1H), 5.66 (dt, J¼6.1, 15.7 Hz, 1H), 5.45 (dd,
J¼5.4, 15.7 Hz, 1H), 5.04–5.10 (m, 2H), 4.37–4.45 (m,
1H), 3.64 (t, J¼6.3 Hz, 2H), 3.52–3.62 (m, 1H), 2.92 (d,
J¼15.2 Hz, 1H), 2.84 (d, J¼15.2 Hz, 1H), 2.00–2.19 (m,
6H), 1.30–1.67 (m, 9H), 1.15 (d, J¼7.0 Hz, 3H), 1.05 (s,
9H); (75 MHz, CDCl3) d 170.3, 142.6, 135.8, 135.0, 134.1,
132.9, 132.7, 130.3, 129.8, 129.5, 129.3, 127.8, 126.9,

8972
R. G. Carter et al. / Tetrahedron 59 (2003) 8963–8974
126.0, 118.2, 108.6, 106.4, 70.6, 70.4, 63.4, 42.2, 42.1, 34.5,
32.0, 31.4, 30.2, 30.0, 28.8, 27.0, 19.4, 18.9; HRMS
(FABþ) calcd for C43H55O6SSi (MþH) 727.3489, found
727.3504.
1H NMR (300 MHz, CDCl3) d 7.74–7.78 (m, 2H), 7.60–
7.70 (m, 5H), 7.45–7.53 (m, 2H), 7.27–7.43 (m, 11H),
5.86–5.96 (m, 1H), 5.33–5.63 (m, 3H), 4.80–4.86 (m, 1H),
4.47–4.66 (m, 2H), 4.13–4.18 (m, 1H), 3.50–3.85 (m, 6H),
3.16–3.23 (m, 1H of a diastereomer), 3.10 (s, 3H of a
diastereomer), 3.07 (s, 3H of a diastereomer), 3.00–3.10 (m,
1H of a diastereomer), 2.65 (dd, J¼10.0, 13.8 Hz, 1H of a
diastereomer), 1.72–2.32 (m, 7H), 1.55–1.70 (m, 3H), 1.3–
1.52 (m, 4H), 1.09 (s, 9H of a diastereomer), 1.07 (s, 9H of a
diastereomer), 1.05–1.10 (m, 3H), 0.99 (t, J¼9.4 Hz, 6H of
1 diastereomer), 0.97 (t, J¼9.4 Hz, 6H of 1 diastereomer),
0.58–0.72 (m, 6H); 13C NMR (75 MHz, CDCl3) d 205.3,
204.7, 139.2, 139.0, 137.1, 136.9, 135.7, 134.24, 134.17,
132.9, 132.3, 130.5, 129.8, 129.6, 129.5, 129.14, 129.06,
128.53, 128.48, 128.3, 128.1, 127.8, 127.7, 127.6, 97.5,
96.5, 73.0, 72.3, 69.8, 69.3, 69.1, 68.7, 66.5, 66.2, 63.51,
63.47, 49.4, 49.0, 44.6, 44.5, 35.3, 34.7, 34.4, 33.6, 32.2,
32.1, 30.1, 28.9, 28.8, 27.1, 19.4, 15.6, 14.5, 7.04, 7.02, 4.6;
HRMS (FABþ) calcd for C53H71O7SSi2 (M2MeOH)
907.4459, found 907.4452.
6.1.14. Desulfonylated enol ether 27. To a stirred solution
of 49 (12.5 mg, 0.0175 mmol) in THF (0.3 mL) and MeOH
(0.6 mL) at 2108C was sequentially added Na2HPO4
(36 mg) and Na/Hg (224 mg, 5% in Hg). After 20 min, the
reaction was warmed to 08C. After 1 h, the reaction was
diluted with Et2O, filtered through a small pad of Celitew
(Et2O rinse) and concentrated in vacuo. The crude product
27 was used immediately in subsequent manipulations.
6.1.15. C17 Spirocycle 28. To a stirred solution of crude 27
(0.0175 mmol) in PhMe (1.0 mL) and t-BuOH (1.0 mL) was
added CSA (17.1 mg, 0.0737 mmol) at room temperature
After 14.5 h, the reaction was quenched with solid NaHCO3
(250 mg). After 10 min, the solution was diluted with 25%
EtOAc/hexanes (25 mL), filtered through a small plug of
silica gel (25% EtOAc/hexanes rinse), and concentrated in
vacuo. The resultant oil was purified using preparative thin
layer chromatography over silica gel, eluting with 20%
EtOAc/hexanes, to give 28 (7.6 mg, 0.013 mmol, 76% over
2 steps) as a colorless oil: [a]2D3¼250.08 (c 0.38, CHCl3); IR
6.1.17. Ketone 31. To a stirred solution of 30 (16.0 mg,
0.017 mmol) in THF (0.3 mL) and MeOH (1.0 mL) at
2108C was added Na2HPO4 (16 mg, 0.112 mmol). After
5 min, Na/Hg amalgam (118 mg, 0.257 mmol, 5% in Hg)
was added. After 2 h, the reaction was diluted with 20%
EtOAc/hexanes, filtered through a small plug of silica gel
(20% EtOAc/hexanes rinse) and concentrated in vacuo to
yield crude 31 (0.017 mmol) which was used without any
further purification. [a]D23¼20.58 (c 2.55, CHCl3); IR (neat)
2931, 1714, 1393, 1109 cm21; 1H NMR (300 MHz, CDCl3)
d 7.65–7.70 (m, 4H), 7.20–7.45 (m, 11H), 5.93–6.00 (m,
1H), 5.63–6.81 (dt, J¼6.4, 15.5 Hz, 1H), 5.48–5.60 (m,
2H), 4.73 (d, J¼11.7 Hz, 1H), 4.52 (d, J¼11.7 Hz, 1H),
4.21–4.28 (m, 1H), 3.65–3.75 (m, 3H), 3.58 (dd, J¼5.2,
10.3 Hz, 1H), 3.44–3.51 (m, 1H), 3.21 (s, 3H), 2.64–2.80
(m, 1H), 2.43–2.55 (m, 2H), 1.65–2.20 (m, 9H), 1.47–1.55
(m, 1H), 1.27 (s, 9H), 1.00 (d, J¼6.8 Hz, 3H), 0.98 (t, J¼
7.8 Hz, 9H), 0.62 (q, J¼7.8 Hz, 6H); 13C NMR (75 MHz,
CDCl3) d 214.0, 138.9, 135.8, 134.2, 132.7, 130.3, 129.8,
128.5, 128.4, 128.2, 127.8, 98.2, 77.8, 72.5, 68.9, 65.9, 63.5,
48.4, 43.2, 35.7, 35.4, 32.1, 30.6, 29.5, 28.9, 27.1, 19.4,
16.5, 7.0, 4.6; HRMS (FABþ) calcd for C47H67O5Si2
(M2MeOH) 767.4527, found 767.4512.
(neat) 3037, 2928, 2865, 1461, 1110 cm21 1H NMR
;
(300 MHz, CDCl3) d 7.65–7.68 (m, 4H), 7.34–7.42 (m,
6H), 5.93–6.97 (m, 1H), 5.59–5.93 (m, 4H), 4.94–5.04 (m,
2H), 4.38–4.43 (m, 1H), 3.80–3.86 (m, 1H), 3.66 (t,
J¼6.2 Hz, 2H), 1.99–2.25 (m, 12H), 1.5–1.78 (m, 5H),
1.04 (s, 9H), 0.87 (d, J¼6.2 Hz, 3H); 13C NMR (125 MHz,
CDCl3) d 136.0, 134.4, 132.0, 131.0, 129.9, 129.0, 128.2,
128.0, 116.5, 110.1, 105.4, 70.9, 70.0, 63.8, 41.1, 37.9, 36.0,
32.4, 31.5, 30.6, 30.1, 29.2, 28.9, 27.3, 19.6, 16.7; HRMS
(FABþ) calcd for C37H49O4Si (MþH) 585.3400, found
585.3397.
6.1.16. Keto sulfone 30. To a stirred solution of sulfone 86a
(76.0 mg, 0.126 mmol) in THF (0.8 mL) at 2788C was
added LDA25 (140 mL, 1.0 M in THF) dropwise. After
25 min, a solution of the aldehyde 296e (53.9 mg,
0.160 mmol) in precooled THF (0.2 mL) was added via
cannula to the orange sulfone solution. After 25 min, the
reaction was removed from the cooling bath. After 2 min,
the reaction was quenched with sat. aq. NH4Cl (25 mL) and
extracted with Et2O (4£30 mL). The dried (MgSO4) extract
was concentrated in vacuo to give the crude 50 (125 mg) as
a colorless oil. The crude hydroxy sulfone 50 was used
immediately; chromatography of the crude mixture resulted
in spirocyclization at C10 to a complex mixture of isomers.
6.1.18. C16 Spirocycles 32 and 33. To a stirred solution of
crude 31 (0.017 mmol) in PhMe (1.1 mL) and t-BuOH
(1.1 mL) was added CSA (19 mg, 0.0818 mmol). After
17 h, the reaction was quenched with solid NaHCO3
(200 mg). After 5 min, the solution was diluted with 40%
EtOAc/hexanes, filtered through a small plug of silica gel
(40% EtOAc/hexanes rinse) and concentrated in vacuo. The
crude oil was purified by preparative TLC (15% EtOAc/
hexanes) to give the less polar transoidal spiroketal 32
(3.3 mg, 0.0051 mmol, 30%) and more polar cisoidal
spiroketal 33 (5.5 mg, 0.0084 mmol, 50%) as colorless
oils. transoidal 32: [a]2D3¼234.78 (c 0.265, CHCl3); IR
To a stirred solution of crude 50 (0.126 mmol) in CH2Cl2
˚
(1.0 mL) with powdered 4 A mol. sieves (100 mg) was
sequentially added NMO (19.0 mg, 0.162 mmol) and TPAP
(18.0 mg, 0.0512 mmol). An additional portion of TPAP
(17.8 mg, 0.0506 mmol) was added during the course of the
reaction. After 3.25 h, the reaction was diluted with 25%
EtOAc/hexanes (5 mL), filtered through a small plug of
silica gel (25% EtOAc/hexanes rinse) and concentrated in
vacuo and purified by chromatography over silica gel,
eluting with 7–40% EtOAc/hexanes (with 0.5% Et3N), to
give 30 (70.4 mg, 0.0751 mmol, 60% over 2 steps) as a
1
(neat) 2960, 2926, 2853, 1427, 1110, 702 cm21; H NMR
(300 MHz, CDCl3) d 7.67 (d, J¼6.1 Hz, 4H), 7.26–7.43 (m,
11H), 5.93–5.97 (m, 1H), 5.61–5.69 (m, 2H), 5.50 (dd,
J¼6.0, 15.8 Hz, 1H), 4.51–4.55 (m, 2H), 4.35–4.39 (m,
1H), 3.68–3.83 (m, 3H), 3.66 (t, J¼6.2 Hz, 2H), 1.59–2.17
(m, 13H), 1.05 (s, 9H), 1.03 (d, J¼7.1 Hz, 3H); 13C NMR
colorless oil: IR (neat) 2931, 1721, 1448, 1310, 1110 cm21
;

R. G. Carter et al. / Tetrahedron 59 (2003) 8963–8974
8973
Gronemeyer, M. A.; Tschumper, G. S. Org. Lett. 2002, 4,
2181–2184.
(75 MHz, CDCl3) d 138.9, 135.8, 134.2, 132.3, 130.7,
129.74, 129.66, 128.6, 128.0, 127.81, 127.76, 108.9, 104.7,
70.7, 69.5, 69.2, 64.2, 63.4, 36.5, 33.6, 33.14, 33.08, 32.1,
30.3, 28.9, 27.0, 19.4, 16.3; HRMS (FABþ) calcd for
C41H53O5Si (MþH) 653.3662, found 653.3658. cisoidal 33:
[a]2D3¼258.08 (c 0.40, CHCl3); IR (neat) 3069, 3032, 2961,
2921, 2851, 1467, 1111 cm21; 1H NMR (300 MHz, CDCl3)
d 7.66 (dd, J¼1.7, 7.5 Hz, 4H), 7.26–7.45 (m, 11H), 5.94–
6.00 (m, 1H), 5.52–5.82 (m, 3H), 4.61 (d, J¼12.4 Hz, 1H),
4.52 (d, J¼12.4 Hz, 1H), 4.36–4.42 (m, 1H), 3.79–3.90 (m,
2H), 3.66 (t, J¼6.2 Hz, 2H), 3.41–3.43 (m, 1H), 1.60–2.24
(m, 13H), 1.04 (s 9H), 0.87 (d, J¼6.8 Hz, 3H); 13C NMR
(75 MHz, CDCl3) d 139.0, 135.8, 134.2, 132.2, 130.5,
129.7, 128.7, 128.5, 127.9, 127.8, 127.6, 109.5, 105.3, 77.4,
72.0, 70.1, 63.5, 63.2, 37.4, 35.1, 32.6, 32.2, 32.0, 30.2,
28.9, 27.0, 19.4, 16.4; HRMS (FABþ) calcd for C41H53O5Si
(MþH) 653.3662, found 653.3672.
7. For synthetic efforts by other laboratories toward azaspiracid,
see: (a) Dounay, A. B.; Forsyth, C. J. Org. Lett. 2001, 3,
975–978. (b) Aiguade, J.; Hao, J.; Forsyth, C. J. Org. Lett.
2001, 3, 979–982. (c) Aiguade, J.; Hao, J.; Forsyth, C. J.
Tetrahedron Lett. 2001, 42, 817–820. (d) Hao, J.; Aiguade, J.;
Forsyth, C. J. Tetrahedron Lett. 2001, 42, 821–824.
(e) Nicolaou, K. C.; Pihko, P. M.; Diedrichs, N.; Zou, N.;
Bernal, F. Angew. Chem. Int. Ed. 2001, 40, 1262–1265.
(f) Buszek, K. R. 221st National ACS Meeting, San Diego,
ORGN-570. (g) Forsyth, C. J.; Hao, J.; Aiguade, J. Angew.
Chem. Int. Ed. 2001, 40, 3663–3667. (h) Sasaki, M.; Iwamuro,
Y.; Nemoto, J.; Oikawa, M. Tetrahedron Lett. 2003, 44,
6199–6207.
8. Satake, M.; Ofuji, K.; Naoki, H.; James, K. J.; Furey, A.;
McMahon, T.; Silke, J.; Yasumoto, T. J. Am. Chem. Soc. 1998,
120, 9967–9968.
9. After the submission of this manuscript, the total synthesis
of the proposed structure of azaspiracid was reported.
(a) Nicolaou, K. C.; Li, Y.; Uesaka, N.; Koftis, T. V.;
Vyskocil, S.; Ling, T.; Govindasamy, M.; Qian, W.; Bernal,
F.; Chen, D. Y.-K. Angew. Chem. Int. Ed. 2003, 42,
3643–3648. (b) Nicolaou, K. C.; Chen, D. Y.-K.; Li, Y.;
Qian, W.; Ling, T.; Vyskocil, S.; Koftis, T. V.; Govindasamy,
M.; Uesaka, N. Angew. Chem. Int. Ed. 2003, 42, 3649–3653.
10. MacMahon, T.; Silke, J. Harmful Algae News 1996, 14, 2.
11. Ito, E.; Satake, M.; Ofuji, K.; Kurita, N.; McMahon, T.; James,
K.; Yasumoto, T. Toxicon 2000, 38, 917–930.
Acknowledgements
National Institutes of Health (GM63723) is gratefully
acknowledged for support of this work. The authors would
also like to thank Professor Max Deinzer (Oregon State
´
University) and Dr Jeff Morre (Oregon State University) for
mass spectra data, Professor David E. Graves (University of
Alabama-Birmingham) and Roger Kohnert (Oregon State
University) for NMR assistance as well as Dr Roger
Hanselmann (Rib-X Pharmaceuticals) for his helpful
discussions.
12. (a) James, K. J.; Sierra, M. D.; Lehane, M.; Brana Magdalena,
A.; Furey, A. Toxicon 2003, 41, 277–283. (b) Ofuji, K.;
Satake, M.; McMahon, T.; James, K. J.; Naoki, H.; Oshima,
Y.; Yasumoto, T. Biosci. Biotechnol. Biochem. 2001, 65,
740–742. (c) Ofuji, K.; Satake, M.; McMahon, T.; Silke, J.;
James, K. J.; Naoki, H.; Oshima, Y.; Yasumoto, T. Nat. Toxins
1999, 7, 99–102.
References
1. For a recent review of bisspiroketals: (a) Brimble, M. A.;
´
Fares, F. A. Tetrahedron 1999, 55, 7661–7706. For other
13. James, K. J.; Furey, A.; Lehane, M.; Ramstad, H.; Aune, T.;
Hovgaard, P.; Morris, S.; Higman, W.; Satake, M.; Yasumoto,
T. Toxicon 2002, 40, 909–915.
synthetic work towards bisspiroketals not included in the
review: (b) Furkert, D. P.; Brimble, M. A. Org. Lett. 2002, 4,
3655–3658. (c) Nakamura, S.; Inagaki, J.; Kudo, M.;
Sugimoto, T.; Obara, K.; Nakajima, M.; Hashimoto, S.
Tetrahedron 2002, 58, 10353–10374. (d) Caprio, V.; Brimble,
M. A.; Furkert, D. P. Tetrahedron 2001, 57, 4023–4034.
(e) Allen, P. R.; Brimble, M. A.; Prabaharan, H. J. Chem. Soc.,
Perkin Trans. 1 2001, 379–389.
14. James, K. J.; Moroney, C.; Roden, C.; Satake, M.; Yasumoto,
T.; Lehane, M.; Furey, A. Toxicon 2003, 41, 145–151.
15. (a) Hanessian, S.; Grillo, T. A. J. Org. Chem. 1998, 63,
´
1049–1057. (b) Solladıe, G.; Almario, A.; Colobert, F. Synlett
1992, 167–168. (c) Smith, A. B.; Hale, K. J. Tetrahedron Lett.
1989, 30, 1037–1040. (d) O’Connor, S. J.; Williard, P. G.
Tetrahedron Lett. 1989, 30, 4637–4640. (e) Beau, J.-M.;
Sinay¨, P. Tetrahedron Lett. 1985, 26, 6193–6196. (f) Trost,
B. M.; Lynch, J.; Renaut, P.; Steinman, D. H. J. Am. Chem.
Soc. 1986, 108, 284–291.
2. For a recent review of the anomeric effect: Graczyk, P. P.;
Mikolajczyk, M. Top. Stereochem. 1994, 21, 159–349.
3. (a) McGarvey, G. J.; Stepanian, M. W. Tetrahedron Lett.
1996, 37, 5461–5464. (b) McGarvey, G. J.; Stepanian, M. W.;
Bressette, A. R.; Ellena, J. F. Tetrahedron Lett. 1996, 37,
5465–5468. (c) Brimble, M. A.; Rush, C. J. J. Chem. Soc.,
Perkin Trans. 1 1994, 497–500.
´
16. (a) Alvarez, E.; Cuvigny, T.; Herve du Penhoat, C.; Julia, M.
Tetrahedron 1988, 44, 119–126. (b) Ashwell, M.; Jackson,
R. F. W. J. Chem. Soc., Perkin Trans. 1 1989, 835–847.
(c) Ichikawa, Y.; Isobe, M.; Masaki, H.; Kawai, T.; Goto, T.
Tetrahedron 1987, 43, 4759–4766.
4. The terms transoidal and cisoidal refer to the relationship of
the two pyran oxygens to one another.
17. Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P.
Synthesis 1994, 639–666.
5. For a recent hydrogen-bonding example: Nicolaou, K. C.;
Qian, W.; Bernal, F.; Uesaka, N.; Pihko, P. M.; Hinrichs, J.
Angew. Chem. Int. Ed. 2001, 40, 4068–4071.
18. Tidwell, T. T. In Handbook of Reagents for Organic
Synthesis: Oxidizing and Reducing Agents; Burke, B. D.,
Danheiser, R. L., Eds.; Wiley: New York, 1999; pp 154–156.
19. Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113,
7277–7287.
6. (a) Carter, R. G.; Weldon, D. J. Org. Lett. 2000, 2, 3913–3916.
(b) Carter, R. G.; Weldon, D. J.; Bourland, T. C. 221st
National ACS Meeting, San Diego, ORGN-479. (c) Carter,
R. G.; Graves, D. E. Tetrahedron Lett. 2001, 42, 6035–6039.
(d) Carter, R. G.; Bourland, T. C.; Graves, D. E. Org. Lett.
2002, 4, 2177–2179. (e) Carter, R. G.; Graves, D. E.;
20. The terms natural and nonnatural refer to relationship of the
stereochemistry at C10 and C13 to the proposed stereochemistry

8974
R. G. Carter et al. / Tetrahedron 59 (2003) 8963–8974
of azaspiracid as shown in compound 3. A natural transoidal
24. (a) Juaristi, E.; Cuevas, G. The Anomeric Effect; CRC: Boca
Raton, FL, 1995. (b) For an alternate explanation involving the
Principle of Least Molecular Deformation: Sinnott, M. L. Adv.
Phys. Org. Chem. 1988, 24, 113–204.
bisspiroketal possesses the same stereochemical relationship
as 3 whereas a nonnatural transoidal spiroketal would possess
the opposite C10, C13 stereochemistry to 3. These terms are
used instead of the traditional Cahn–Ingold–Prelog R/S
nomenclature as the priority rankings change due to the
presence of the C12 sulfone.
25. The 1.0 M LDA solution was prepared fresh immediately prior
to use: to a stirred solution of N,N-diispropyl amine (404 mg,
560 mL, 4.0 mmol) in THF (1.84 mL) at 2788C was added
n-BuLi (1.6 mL, 4.0 mmol, 2.5 M in hexanes) dropwise. After
5 min, the white suspension was warmed to 2108C. After
30 min, the solution was employed in the relevant reaction.
26. The (Ipc)2Ballyl was prepared as a stock solution in pentane
from the commercially available (2)-Ipc2BOMe and
allylMgBr in accord with the low salt protocol developed by
Brown and co-workers Racherla, U. S.; Brown, H. C. J. J. Org.
Chem. 1991, 56, 401–404.
21. It should be noted that decomposition of the bisspirocycles 18
and 20 was a competitive process upon extended reaction
times.
´
22. (a) Miljkovıc, M.; Yeagley, D.; Deslongchamps, P.; Dory,
Y. L. J. Org. Chem. 1997, 62, 7597–7604. (b) Srivastava,
R. M.; Pava˜o, A. C.; Seabra, G. M.; Brown, R. K. J. Mol.
Struct. 1997, 412, 51–58.
23. This hypothesis assumes that no additional stabilizing
functions are present (e.g. Nicolaou and co-workers use of a
C9 hydroxyl function). See Ref 5.