Exploration of conjugate addition routes to advanced tricyclic components of mangicol A

Article abstract of DOI:10.1016/j.tet.2006.03.066

Two synthetic approaches to the cytotoxic marine natural product known as mangicol A are described. The starting material common to both pathways is the cyclopentenonecarboxylate 11. The first tactic involves the 1,4-addition to 11 of the cuprate derivable from iodide 10, while the second proceeds via base-promoted conjugate addition of the regiospecifically generated enolate anion of 41. The first strategy proceeds by a series of efficient steps to tricyclic aldol 21 and subsequently to β-diketone 7. The latter proved to be totally unresponsive to schemes aimed at introduction of a butenyl group. The second approach involves earlier introduction of this substituent as realized in stereocontrolled fashion via transition state 42. While further passage to 44 proved uneventful, this advanced intermediate and analogs thereof proved remarkably recalcitrant to cyclization in the precedented fashion. In no instance was generation of a suitable product realized. These studies serve to underscore the extent to which steric considerations can complicate matters and the extent to which they must be skirted. Finally, a direct enantioselective route to the side chain aldehyde 2 is detailed.

Full text of DOI:10.1016/j.tet.2006.03.066

Tetrahedron 62 (2006) 5791–5802
Exploration of conjugate addition routes to advanced tricyclic
components of mangicol A
*
Stefan Pichlmair, Manuel de Lera Ruiz, Ivan Vilotijevic and Leo A. Paquette
The Evans Chemical Laboratories, The Ohio State University, Columbus, OH 43210, United States
Received 22 January 2006; revised 10 March 2006; accepted 21 March 2006
Available online 2 May 2006
Abstract—Two synthetic approaches to the cytotoxic marine natural product known as mangicol A are described. The starting material com-
mon to both pathways is the cyclopentenonecarboxylate 11. The first tactic involves the 1,4-addition to 11 of the cuprate derivable from iodide
10, while the second proceeds via base-promoted conjugate addition of the regiospecifically generated enolate anion of 41. The first strategy
proceeds by a series of efficient steps to tricyclic aldol 21 and subsequently to b-diketone 7. The latter proved to be totally unresponsive to
schemes aimed at introduction of a butenyl group. The second approach involves earlier introduction of this substituent as realized in stereo-
controlled fashion via transition state 42. While further passage to 44 proved uneventful, this advanced intermediate and analogs thereof
proved remarkably recalcitrant to cyclization in the precedented fashion. In no instance was generation of a suitable product realized. These
studies serve to underscore the extent to which steric considerations can complicate matters and the extent to which they must be skirted.
Finally, a direct enantioselective route to the side chain aldehyde 2 is detailed.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
core.3 In the light of these early results, the decision was
made to evaluate alternative routes to 1 that feature conver-
gent Michael reactions as the mode of structural assembly.
As outlined in Scheme 1, the retrosynthesis is keyed to the
availability of enedione 6 whose role is ultimately to enter
into intramolecular photochemical [2+2] cyclization and
In a preceding paper,1 we reviewed background issues
concerning the antiinflammatory agent mangicol A (1)2
and reported on the possible application of intramolecular
[4+2] cycloaddition strategies for assembly of its central
Me
Me
Me
Me
H
TIPSO
O
+
Me
H
H
OR
Me
OH
Me
Me
OH
nucleophilic
addition
OTIPS
I
HO
HO
Me
MOMO
Me
OH
3
2
radical
fragmentation
Mangicol A (1)
OCS2Me
MeO2C
MeO2C
MeS2CO
MeS2CO
Me
OR
O
O
O
Me
Me
OR
O
H
H
H
OR
Me
Me
Me
TBDPSO
TBDPSO
TBDPSO
4
5
[2+2] photocyclization
6a, R = PMB
b, R = MOM
Scheme 1.
Keywords: Mangicols; Conjugate additions; Aldol reactions; Functionalized diquinanes; Steric constraints.
* Corresponding author. Tel.: +1 614 292 2520; fax: +1 614 292 1685; e-mail: paquette.1@osu.edu
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.03.066

5792
S. Pichlmair et al. / Tetrahedron 62 (2006) 5791–5802
deliver 4. Presently, the focus of our attention is the need to
develop a workable route to 6. Two subsequences aimed at
exploiting conjugate addition pathways to this enedione
form the subject of the present account. Also documented is
a brief synthesis of the enantiomerically enriched aldehyde 2.
fashion6 and reduced with sodium borohydride to furnish
the allylic alcohol 13 (Scheme 3). Formation of iodide 14
was most efficiently accomplished by treatment with lithium
chloride and methanesulfonyl chloride in DMF containing
2,6-lutidine, followed by the addition of sodium iodide.7
The targeted oxazolidinone was generated through reaction
of 14 with the sodium enolate of 15 with a de in excess of
95%.8 Removal of the chiral auxiliary was smoothly
achievedwithNaBH4. Thecarbinol16soformedwaschemo-
selectively reduced in a two-step process involving initial
production of the iodide followed by the action of tri-n-butyl-
tin hydride.9 Finally, the benzoate group was saponified and
the resultant chiral alcohol was transformed without event
into 10 by a closely related halogenation protocol.
2. Exploration of a cuprate-based approach
The first option to be explored for ultimately reaching 6a was
based on the suitability of engaging dienyl iodide 10 in 1,4-
conjugate addition to the enantiopure cyclopentenonecar-
boxylate 11.1 This step was anticipated to proceed with the
very dominant formation of 9 as a result of kinetically fa-
vored approach from that direction syn to the adjacent methyl
substituent (Scheme 2). As matters worked out, the doubly
activated nature of the Michael acceptor 11 facilitated cap-
ture of the organocopper species derived from 10.4 The con-
version of 9 to 8, or a variant of this system, was designed to
allow assembly of a highly functionalized cyclopentene,
whose cyclization under appropriate conditions would allow
the generation of 7. g-Alkylation of this cyclohexenone with
4-iodo-1-butene5 was to ensue, this step setting the stage for
more advanced functional group manipulation within 6a.
The coupling of iodide 10 to cyclopentenonecarboxylate 11
was most efficiently achieved using tert-butyllithium and
CuI to form the cuprate. Under these conditions, 9 was iso-
lated as a 16:1 mixture of diastereomers. With the benefit of
COSY and NOESY experiments, the major isomer could be
readily identified as the expected 9 (Scheme 4). Selective
borohydride reduction of the ketone carbonyl in this inter-
mediate gave rise to a 2:1 diastereomeric mixture of alco-
hols, which were protected as their benzoates. Subsequent
reductive ozonolysis furnished the keto aldehyde 18, which
was directly cyclized to cyclohexenone 19 under acidic
conditions. Recourse was next made to a Luche reduction
Aldehyde 12, prepared in two steps from commercially
available 1,4-butanediol, was methylenated in Mannich
MeO2C
O
MeO2C
O
MeO2C
O
O
O
Me
Me
Me
BzO
-alkylation
H
H
H
Me
OPMB
Me
OPMB
OPMB
Me
TBDPSO
TBDPSO
7
8
TBDPSO
6a
sequencial
aldol condensation
O
CO2Me
O
CO2Me
conjugate
addition
Me
Me
I
+
Me
H
Me
TBDPSO
TBDPSO
9
11
10
Scheme 2.
1. Me2NH, CH2O
2. NaBH4
1. LiCl, MsCl,
2,6-lutidine (83%)
OH
OH
I
O
(50%, 2 steps)
2. NaI
BzO
OBz
BzO
BzO
12
13
14
14
1. Ph3P, I2, im.
O
O
2. Bu3SnH, AIBN
3. K2CO3, MeOH
1. NaHMDS, THF
2. NaBH4
Me
N
O
(78%, 3 steps)
(79%)
Me
15
Ph
HO
16
17
1. MsCl, TEA
2. NaI
10
(62%, 2 steps)
Scheme 3.

S. Pichlmair et al. / Tetrahedron 62 (2006) 5791–5802
5793
CO2Me
O
BzO
CO2Me
OHC
1. NaBH4, MeOH
Me
Me
Me
t-BuLi, CuI
then 11
2. BzCl, TEA, CH2Cl2
I
3. O3, PPh3, CH2Cl2
(89%, 3 steps)
H
H
(66%, 88% de)
Me
Me
O
TBDPSO
TBDPSO
18
10
9
BzO
CO2Me
BzO
CO2Me
Me
1. NaBH4, CeCl3
Me
2. PMBBr, NaH
p-TsOH, 80 °C
(75%, 2 steps, de > 95%)
(64%)
H
Me
H
O
Me
TBDPSO
OPMB
TBDPSO
20
19
MeO2C
BzO
MeO2C
O
O
Me
OH
Me
1.O3, PPh3, CH2Cl2
2. AcOH, piperidine
1. K2CO3, MeOH
2. Dess-Martin, Py.
(65%, 2 steps)
(36%, 2 steps)
H
OPMB
H
OPMB
Me
Me
TBDPSO
TBDPSO
8
21
TBSOTf,
2,6-Lutidine
(100%)
Dess Martin, Py.
CH2Cl2
(37%)
MeO2C
O
Me
TBSO
7
H
OPMB
Me
TBDPSO
22
Scheme 4.
(de>95%) and subsequent formation of the p-methoxy-
benzyl ether. Ozonolytic of the trisubstituted double bond
and ensuing cyclization of the resulting keto aldehyde with
piperidine in acetic acid delivered 8 in 36% yield over the
two steps.
or triethylsilyl chlorides were met with failure, presumably
as a result of steric congestion, enhanced reactivity was
sought in the form of tert-butyldimethylsilyl triflate in the
presence of 2,6-lutidine. These conditions resulted in forma-
tion of acetal 22 (100% yield).
With aldehyde 8 in hand, the time had arrived for removal of
the benzoate protecting group and regeneration of the b-keto
ester functionality. Pleasingly, use of the Dess–Martin perio-
dinane and pyridine in CH2Cl2 proceeded with in situ cycli-
zation to generate the tricyclic aldol 21 as a 10:1 mixture of
diastereomers at the allylic alcohol center. The first-stage
oxidation is a very quick process (complete in less than
5 min) and cyclization ensues immediately. In contrast, the
second-stage oxidation involving 21 is a rather slow process,
which required 16 h to reach completion. The stereochem-
Our arrival at the tricyclic intermediates 7 and 21 was met
with the breakdown of many potentially useful transforma-
tions. For example, attempts to reduce the ester moiety in
21 with LiAlH4 or DIBAL-H led to rather complex product
mixtures. Similarly, while 21 could be chemoselectively
reduced with sodium triacetoxyborohydride by way of intra-
molecular hydride transfer10 with introduction of a b-hy-
droxyl subunit as in 23 (Scheme 5), controlled conversion
to a monoxanthate was not possible as a prelude to tin hy-
dride reduction.11 Direct reduction of the ketone carbonyl
in 21 via the formation of the tosylhydrazone12 promoted
wholesale degradation.
1
ical assignments to 21 are soundly based on H NMR, 13C
NMR, HMQC, and NOESY experiments (see A).
On a more positive note, treatment of 23 with triethylsilyl
chloride served as a means for achieving efficient monopro-
tection. As a result, arrival at 24 proved not to be a challenge.
Various options for the exploration of routes to the g-alkyl-
ated product 25 could now begin. The most direct route in-
volving generation of the extended anion of enone 24 and
reaction with 4-iodo-1-butene was disadvantaged from the
start due to a sensitivity of this intermediate to strong base.
Upon admixture with a variety of bases, decomposition
was seen to set in quickly in all cases. Experiments designed
to generate the conjugated silyl enol ether 2613 and to
MeO2C
HO
O
Me
H
H
H
H
H
H
H
H
Me
OPMB
TBDPSO
A
Transient protection of the secondary hydroxyl group in 21
came to be regarded as desirable. When initial attempts to
accomplish this transformation with tert-butyldimethylsilyl

5794
S. Pichlmair et al. / Tetrahedron 62 (2006) 5791–5802
1. NaBH(OAc)3
MeO2C
HO
MeO2C
OH
MeO2C
TESO
AcOH, CH3CN
Me
OH
O
Me
O
Me
NaBH(OAc)3
AcOH, CH3CN
2. TESCl, TEA, DMAP
3. Dess-Martin
(quant.)
(85%, 3 steps)
H
OPMB
H
Me
H
OPMB
OPMB
Me
TBDPSO
Me
Me
TBDPSO
TBDPSO
23
21
24
MeO2C
TESO
O
MeO2C
TESO
MeO2C
TESO
OTMS
O
Me
Me
1. base, then
TMSOTf, TEA
I
H
decomposition
OPMB
Me
H
H
TBDPSO
OPMB
OPMB
Me
Me
TBDPSO
TBDPSO
26
24
25
NBS, AIBN, CCl4
MeO2C
TESO
O
Me
H
OPMB
Me
Br
TBDPSO
27
Scheme 5.
brominate the g-position as in 275a,14 were likewise met with
failure. These and related problems prompted consideration
of means for introduction of the butenyl group earlier in the
synthesis.
ozonolysis of the remaining double bond in 29. The diol so
formed proved entirely accommodative of chemoselective
O-silylation at the primary site with TBSCl. Finally, the tar-
geted 30 was secured by the application of Swern oxidation
conditions.
3. Consideration of earlier incorporation
of the alkenyl chain
While the conversion of 20 to 30 proceeded quite satisfacto-
rily, 30 resisted alkylation via its enolate at every turn. These
attempts at functionalization included treatment with bases
such as NaHMDS, KHMDS, LDA, NaH, and KH, as well
as recourse to electrophiles exemplified by 4-bromo-1-
butene, 4-iodo-1-butene, allyl bromide, methyl iodide, and
gaseous formaldehyde. In the light of these developments,
the decision was made to advance on these intermediates
in an alternate fashion.
Ketone 30 was successfully prepared from 20 via the se-
quence outlined in Scheme 6.15 The benzoate group was ex-
cised with potassium carbonate, and the secondary hydroxyl
so uncovered was transformed into the mesylate. b-Elimina-
tion now became possible by treatment with DBU, and pro-
vided 28 in 90% overall yield. Next to be explored was
selective reduction of the double bond conjugated with the
carbomethoxy substituent. The utilization of NaBH4 and
NiCl2 in tandem16 proved quite amenable to this transforma-
tion. There followed the direct reduction to the primary car-
binol with DIBAL-H and formation of the benzoate to
furnish 29. Continued success was realized with reductive
4. The quest for more direct assembly
The prominent complications witnessed so far for achieving
proper alkylation of relevant advanced intermediates
1. NiCl2 · 6H2O,
NaBH4
2. DIBAL-H
BzO
CO2Me
1. K2CO3
CO2Me
Me
Me
2. MsCl, TEA,
then DBU
(90%, 2 steps)
3. BzCl, TEA
H
H
Me
Me
OPMB
OPMB
(81%, 3 steps)
TBDPSO
TBDPSO
BzO
28
20
OTBS
Me
BzO
1. O3, NaBH4
2. TBSCl, TEA
3. Swern
OTBS
Me
BzO
O
Me
O
(74%, 3 steps)
H
H
Me
Me
OPMB
H
OPMB
TBDPSO
Me
TBDPSO
OPMB
TBDPSO
29
30
31
Scheme 6.

S. Pichlmair et al. / Tetrahedron 62 (2006) 5791–5802
5795
emphasize the need for a strategic disconnection that makes
provision for the earlier introduction of a suitable side chain.
To this end, the involvement of 35 as a central building block
was considered attractive (Scheme 7). The implementation
of this plan called for site-specific deprotonation of 35 in ad-
vance of its Michael addition to 11. The appeal offered by
the framework where X was to be an appropriately config-
ured protected carbinol center defined our original thrust
in this direction. Further along the retrosynthetic pathway,
34 was to be cyclized to 33 in a manner paralleling the earlier
conversion of 19 to 20. The execution of a second-stage aldol
ring closure followed by minor functional group adjustments
was to lead to the tricyclic framework represented by 32.
a three-step sequence consisting of PMB protection, double
bond cleavage via the 1,2-diol, and oxidation with sodium
chlorite.17 The coupling of 37 to the Evans auxiliary (R)-4-
benzyl-2-oxazolidinone18 was mediated via the mixed anhy-
dride generated with pivaloyl chloride. The availability of
38 allowed for the implementation of an enantioselective
a-hydroxylation step involving the Davis oxaziridine.19
Although a diastereomeric excess of 5:1 was realized, the
enantiomeric purity could be readily enhanced to the
100% level by chromatographic separation of the pair of de-
rived MOM ethers on silica gel. Since attempts to transform
39 into its Weinreb amide resulted in destruction of the
material, an alternative pathway involving reduction to the
alcohol and Swern oxidation provided aldehyde 40 in 74%
overall yield. The route to 41 was then completed by 1,2-ad-
dition of 4-pentenylmagnesium bromide and exposure of the
resulting carbinol to the Dess–Martin periodinane reagent.20
MeO2C
O
O
Me
Me
O
Me
The deprotonation of 41 with KHMDS at ꢀ78 ꢁC proceeded
with high regioselectivity to generate the enolate anion de-
picted in 42 (Scheme 9). The intramolecular chelation involv-
ing the MOM substituent specifically defined in this transition
state serves to enhance steric biases and allows for the specific
formation of 43 as the only observed diastereomer. Chemo-
selective deprotection of the OPMB group was easily ac-
complished with DDQ in conventional fashion, thereby
making possible ensuing periodinane oxidation to the highly
functionalized diketo aldehyde 44 in 80% isolated yield.
H
H
OMOM
Me
OMOM
Me
OTBDPS
TBDPSO
32
33
Tandem aldol
condensation
OPMB
Me
O
MeO2C
O
Me
O
CO2Me
+
O
O
Michael
addition
When conditions previously developed by Hiranuma and
Hudlicky21 for intermolecular aldol condensations of the
projected 44/45 type were examined (see Table 1, experi-
ments 1 and 5), enamine formation was observed as in the
other cases (1H NMR analysis). However, this intermediate
underwent no further chemical reaction and resisted cycliza-
tion on prolonged heating. Treatment of 44 in the manner de-
vised in entries 2 and 3 of Table 1 resulted in clean amidation
of the carbomethoxy group to give 47, with this conversion
H
Me
X
OMOM
Me
OTBDPS
OTBDPS
11
34
35
Scheme 7.
The conversion of S-citronellol (36) to carboxylic acid 37,
shown in Scheme 8, was conveniently accomplished in
Me
PivCl, TEA;
1. NaH, PMBBr, THF
Me
OH
Li
N
2. OsO4, then NaIO4
HO
O
3. NaClO2, NaH2PO4,
PMBO
Ph
2-methyl-2-butene
(73%, 3 steps)
O
O
(94%)
Me
Me
S-(-) Citronellol (36)
37
Me
Me
1. NaHMDS, Davis
oxaziridine
PMBO
1. LiBH4, THF/H2O
PMBO
O
2. Swern oxidation
O
OMOM
Ph
2. MOMCl, NEt(iPr)2
(50%, 2 steps)
(74%, 2 steps)
N
N
O
O
Ph
O
O
38
39
Me
Me
MgBr
PMBO
1.
O
PMBO
O
OMOM
2. Dess-Martin, CH2Cl2
(73%, 2 steps)
OMOM
H
40
41
Scheme 8.

5796
S. Pichlmair et al. / Tetrahedron 62 (2006) 5791–5802
OPMB
MeO
OPMB
Me
OPMB
Me
K+
O
CO2Me
O
O
KHMDS,
O
Me
TBDPSO
O
Me
°
THF, -78 C
11
OMOM
H
OMOM
Me
CO2Me
O
OTBDPS
43
single diastereomer
proposed transition state
41
42
O
MeO2C
MeO2C
O
O
Me
Me
O
see
Table 1
1. Dess-Martin, CH2Cl2
2. DMP, CH2Cl2
O
(81%, 3 steps)
H
H
OMOM
OMOM
Me
OTBDPS
Me
OTBDPS
45
44
MeO2C
OH
O
Me
H
Me
OMOM
TBDPSO
46
Scheme 9.
Table 1. Representative conditions applied to 44 for attempted cyclization
and lactol ether 49 was noted when 44 was heated with CSA
in benzene under a Dean–Stark trap. A parallel direction
was followed when 50, the primary carbinol derived from
43 (Scheme 9), was further deprotected as in 51 and sub-
jected to periodinane oxidation. When processed in this
manner, lactone 52 was formed as the predominant product
(Scheme 10).
Expt. Reagent
Solvent
Result
1
2
3
4
5
6
7
Piperidine, HOAc Ether, reflux
Piperidine, HOAc Benzene, reflux Slow conversion to 47
Piperidine, HOAc Toluene, reflux Conversion to 47
Clean enamine formation
Martin sulfurane
Pyrrolidine, CSA
CSA (Dean–Stark) Benzene, reflux Formation of 48 and 49
CH2Cl2, rt
No reaction
Clean enamine formation
Ether, reflux
Et2NH$HCl
ClCH2CH2Cl,
70 ꢁC
MOM cleavage and
lactol formation 48
5. Determining the suitability of aldol cyclizations
for construction of the tricyclic core
occurring faster in refluxing toluene solution as expected.
Other notable transformations included MOM deprotection
and subsequent cyclization with lactol formation when
conditions such as those in experiment 7 were utilized. More
advanced dehydrative elimination to generate the pyran 48
We next investigated the possibility of forming a major
portion of the mangicol A framework through an aldol
ring-forming reaction. To enlist the proper regioselectivity,
the strategy was designed to involve an a-diketone such as
that resident in 54. Although several reaction parameters
O
N
O
Me
O
O
O
CO2Me
O
CO2Me
O
O
O
O
OMe
H
H
H
Me
OMOM
Me
Me
TBDPSO
TBDPSO
TBDPSO
Me
Me
47
49
48
OH
Me
OH
Me
O
O
O
CO2Me
O
CO2Me
O
CO2Me
3 N HCl, THF, 50 °C
(56%)
O
Dess-Martin, CH2Cl2
(59%)
O
O
H
H
H
Me
Me
Me
OMOM
OH
TBDPSO
TBDPSO
TBDPSO
Me
51
50
52
Scheme 10.

S. Pichlmair et al. / Tetrahedron 62 (2006) 5791–5802
5797
OPMB
Me
OBz
OBz
Me
Me
O
O
Bz
BzO
CO2Me
O
CO2Me
O
CO2Me
1. DDQ, CH2Cl2
2. BzCl, DMAP
O
1. BF3OEt2, PhSH
2. Dess-Martin, CH2Cl2
H
H
H
(74%)
OMOM
(77%)
O
OMOM
Me
OTBDPS
43
Me
OTBDPS
Me
OTBDPS
53
54
Me
OH
Me
MeO2C
CHO
OH
MeO2C
MeO2C
O
O
OH
Me
O
OH
TESO
K2CO3, MeOH
(86%)
Dess-Martin, CH2Cl2
(quant.)
O
O
H
H
H
H
Me
H
Me
Me
TBDPSO
TBDPSO
TBDPSO
55
57
56
Scheme 11.
are thereby introduced, the added complexity was expected
to shed light on which mode of cyclization would be kinet-
ically favored. Another advantage would stem from the
reversibility of aldol condensations, which could serve to
clarify available thermodynamic options as well. Thus, 43
was transformed into enol benzoate 53, thereby making
possible removal of the MOM protection group22 and mild
oxidation to gain access to 54 (Scheme 11). The benzoate
groups were then hydrolyzed with K2CO3 in methanol, a
process during which spontaneous passage to diquinane 55
materialized. Although this eventuality was not the desired
one, two positive consequences of this reaction pathway
were made apparent. First, our provisional assignments to
the configuration of several stereocenters could now be fully
corroborated by NOE correlations (see B).
were to no avail, nor was any reaction observed when 58
was comparably exposed to a range of basic conditions. In
fact, 56 and 58 are quite stable entities and can be stored
on the shelf for prolonged periods of time.
A final approach was envisioned to involve samarium
diiodide reduction of carbinol 50 and subsequent Dess–
Martin oxidation to obtain keto aldehyde 58. We attempted
the transformation of 58 to 59 by applying the same condi-
tions as from 44 to 45, with resultant in destruction of 58
(Scheme 12).
6. Synthesis of the side chain
The launching point for arrival at 2 was the allylic alcohol 60
previously shown by Mechelke and Wiemer23 to be available
in two steps from prenyl alcohol (Scheme 13). MOM protec-
tion of the hydroxyl group in 60 was followed by deacetyla-
tion and O-benzylation to deliver 63. With this functionality
in place, it proved straightforward to effect Sharpless asym-
metric dihydroxylation24 with AD-mix-a to afford 64 in
99% yield. This critical step was followed by reaction with
triisopropylsilyl chloride in the presence of NaH, thus mak-
ing possible chemoselective debenzylation25 and generation
of the primary alcohol 65. This very efficient two-step pro-
cess made possible the ultimate production of the targeted
2 via the Dess–Martin protocol.20
Me
OH
MeO2C
O
OH
O
H
H
Me
TBDPSO
B
Secondly, the availability of 55 allowed for one-step con-
version to aldehyde 56. However, all attempts to bring about
retroaldolization with 56 and alternative generation of 57
OH
O
H
MeO2C
O
OH
Me
CO2Me
O
CO2Me
Me
Me
O
O
1. SmI2, MeOH/THF
2. Dess-Martin, CH2Cl2
O
(55%, 2 steps)
H
H
H
Me
OTBDPS
Me
OMOM
Me
TBDPSO
OTBDPS
50
58
59
Scheme 12.

5798
S. Pichlmair et al. / Tetrahedron 62 (2006) 5791–5802
OAc
OAc
Me
1. SeO2, t-BuO2H,
p-MeOC6H4CO2H (39%)
1. LiOH, THF/H2O (100%)
2. MOMCl, (i-Pr)2NEt, THF (82%)
2. BnBr, TBAI, NaH,
THF, 0 °C (90%)
Me
Me
OR
60, R = H
61, R = MOM
OBn
OBn
1. NaH, THF,
TIPSCl, rt, (96%)
AD mix α, MeSO2NH2
HO
t-BuOH/H2O (1:1)
Me
OH
2. H2 (3 atm),
Pd/C, MeOH (93%)
(99%)
Me
OMOM
MOMO
62, R = H
63, R = Bn
64
OH
O
TIPSO
TIPSO
Dess-Martin, CH2Cl2
(87%)
H
Me
OTIPS
Me
OTIPS
MOMO
MOMO
65
2
Scheme 13.
7. Overview
Flash chromatography of the residue on silica gel (gradi-
ent–hexane–ethyl acetate¼5:1 to 3:1) afforded 7.4 g (79%)
of 16; IR (neat, cmꢀ1) 3434, 1721, 1641; 1H NMR
(250 MHz, CDCl3) d 8.05–7.97 (m, 2H), 7.63–7.29 (m,
3H), 5.92–5.72 (m, 1H), 5.13–4.87 (m, 2H), 4.45 (dt,
J¼1.6, 6.9 Hz, 2H), 5.58 (d, J¼5.3 Hz, 2H), 2.51 (t,
J¼6.8 Hz, 2H), 2.23–2.00 (m, 4H), 1.92–1.85 (m, 1H); 13C
NMR (75 MHz, CDCl3) d 166.3, 143.5, 136.3, 132.6,
129.9, 129.2 (2C), 128.0 (2C), 127.0, 126.6, 116.2, 113.0,
64.4, 63.0, 38.0, 37.6, 34.1, 34.4; HRMS ES m/z (M+Na)+
calcd 297.1461, obsd 297.1460.
In this paper, we have detailed the ability of cyclopentenone-
carboxylate 11 to serve as a Michael acceptor under two sets
of circumstances. In the first instances, the employment of
the cuprate derived from 10 as co-reagent proved to be bene-
ficial since subsequent conversion to the tricyclic aldol 21
and diketone 7 was efficiently realized. Our inability to bring
about the proper alkylation of these advanced intermediates
prompted examination of a second approach, which in-
volved the conjugate addition of the enolate of 41 to 11.
While the subsequent elaboration of 44 proceeded unevent-
fully, the intrinsic inability to bring about the appropriate cy-
clization at this stage highlighted the impact of a butenyl side
chain on impeding ring formation. While these strategies
ultimately prove unsuccessful, they are expected to facilitate
further investigation of the synthetic chemistry of the mangi-
cols. The availability of aldehyde 2 constitutes a positive
step in that direction.
8.2. Coupling of 10 with 11
A 250 mL round-bottomed flask was charged with 10
(4.06 g, 15.4 mmol) and ether (30 mL), and the solution
was cooled to ꢀ78 ꢁC while t-BuLi (18.2 mL, 1.7 M in pen-
tane) was introduced. The mixture was stirred in the cold for
5 min, then transferred to a suspension of CuI (1.46 g,
7.67 mmol) in dry ether (20 mL) being stirred at ꢀ20 ꢁC.
The mixture was stirred for another 5 min, cooled to
ꢀ30 ꢁC, and treated with 11 (3.25 g, 9.64 mmol) as a solu-
tion in dry ether (25 mL). The reaction mixture developed
a blue color while it was being stirred for another 50 min.
After quenching with saturated NH4Cl solution (200 mL),
the product was extracted into CH2Cl2 (3ꢂ300 mL), the
combined organic layers were dried, and the solvent was
evaporated under reduced pressure. The residue was purified
by flash chromatography on silica gel (gradient–hexane–
ethyl acetate¼40:1 to 20:1) to give 2.86 g (66%) of 9 as
8. Experimental
8.1. Alkylation of oxazolidinone 15 with iodide 14.
Reduction to the hydroxy benzoate
Oxazolidinone 15 (13.41 g, 51.7 mmol) was dissolved in dry
THF (85 mL) and cooled to ꢀ78 ꢁC at which point sodium
hexamethyldisilazide (52 mL, 1.51 M) was introduced.
The reaction mixture was stirred for 1 h when iodide 14
(12.3 g, 34.2 mmol) was added via cannula as a solution in
dry THF (65 mL). After 3 h, reaction was judged to be com-
pleted and water was added. After warming to rt, the product
was extracted into CH2Cl2 (3ꢂ100 mL) and the combined
organic layers were dried and freed of solvent to leave an or-
ange residue that was dissolved in MeOH (350 mL) at 0 ꢁC.
NaBH4 (9.85 g, 0.26 mmol) was added in portions over a pe-
riod of 30 min. The reaction mixture was stirred overnight,
quenched with water, and extracted with ethyl acetate
(3ꢂ350 mL). The combined organic layers were dried and
the solvent was evaporated to leave an oil that solidified.
1
a pale yellowish oil; IR (neat, cmꢀ1) 1757, 1730, 1640; H
NMR (250 MHz, CDCl3) d 7.70–7.60 (m, 4H), 7.60–7.32
(m, 6H), 5.87–5.65 (m, 1H), 5.07–4.92 (m, 2H), 4.72 (s,
2H), 3.75 (s, 3H), 3.60 (d, J¼10.2 Hz, 1H), 3.49 (dd,
J¼10.2, 1.0 Hz, 1H), 3.10–2.77 (m, 3H), 2.13–1.20 (series
of m, 9H), 1.10 (s, 9H), 0.88–0.78 (m, 6H); 13C NMR
(75 MHz, CDCl3) d 210.4, 170.4, 147.6, 137.1, 135.5
(4C), 133.1 (2C), 129.8 (2C), 127.7 (4C), 115.8, 110.7,
99.8, 67.3, 60.2, 52.4, 50.0, 43.9, 43.7, 42.7, 40.9, 34.4,
30.7, 28.1, 26.8 (3C), 19.3, 17.7; HRMS ES m/z (M+Na)+
calcd 583.3214, obsd 583.3229; [a]2D0 27.0 (c 1.07, C6H6).

S. Pichlmair et al. / Tetrahedron 62 (2006) 5791–5802
5799
8.3. Cyclohexenone 19
(gradient–hexane–ethyl acetate¼10:1 to 5:1) to give 4.48 g
1
(64%) of pure 19; IR (neat, cmꢀ1) 1721, 1677; H NMR
A solution of 9 (5.80 g, 10.3 mmol) in methanol (400 mL)
was cooled to 0 ꢁC, treated with NaBH4 (783 mg,
20.6 mmol) over a period of 5 min, stirred for another
10 min, quenched with water (250 mL), and extracted with
CH2Cl2 (3ꢂ400 mL). The combined organic layers were
dried and evaporated to give 6.0 g (100%) of the b-hydroxy
ester as a pale yellow foam which was not further purified;
(300 MHz, CDCl3, mixture of diastereomers) d 8.06–8.00
(m, 0.7H), 7.97–7.90 (m, 1.3H), 7.74–7.62 (m, 5H), 7.62–
7.28 (m, 8H), 6.69–6.60 (m, 1H), 5.66 (dd, J¼14.4,
6.4 Hz, 0.8H), 5.47–5.37 (m, 0.2H), 3.66 (s, 1H), 3.56 (dd,
J¼13.7, 9.9 Hz, 1H), 3.49 (s, 3H), 3.03 (dd, J¼10.9,
8.1 Hz, 0.7H), 2.83 (dd, J¼10.7, 5.3 Hz, 0.3H), 2.75 (dt,
J¼10.8, 4.3 Hz, 0.7H), 2.60–1.77 (series of m, 9H), 1.73
(dd, J¼14.2, 3.2 Hz, 0.3H), 1.10 (s, 2.1H) and 1.09 (s,
0.7H), 1.06 (s, 9H), 1.00 (d, J¼6.1 Hz, 3H); 13C NMR
(75 MHz, CDCl3, mixture of diastereomers) d 198.8 and
198.5, 174.5 and 171.7, 165.6 and 165.3, 145.3 and 144.8,
137.9 and 137.2, 135.5 (4C), 133.4 (2C), 132.7, 129.8,
129.5 (2C), 129.4 (2C), 129.1 (2C), 127.5 (2C), 77.4, 70.3
and 69.9, 56.7, 53.2, 51.7 and 51.4, 46.3 and 46.2, 45.5,
45.4 and 45.3, 44.3, 43.3 and 42.9, 34.3 and 34.2, 30.3 and
30.2, 29.7 and 29.4, 26.7, 21.2, 19.3, 19.2; HRMS ES m/z
(M+Na)+ calcd 675.3112, obsd 675.3129.
1
IR (neat, cmꢀ1) 3455, 1734, 1641; H NMR (250 MHz,
CDCl3, mixture of diastereomers) d 7.75–7.63 (m, 4H),
7.48–7.28 (m, 6H), 5.84–5.63 (m, 1H), 5.05–4.93 (m, 2H),
4.68 (d, J¼5.0 Hz, 2H), 4.42–4.25 (m, 1H), 3.77–3.68 (m,
3H), 3.41 (s, 1H), 3.35 (d, J¼1.1 Hz, 1H), 2.67–2.50 (m,
2H), 2.40–1.12 (series of m, 9H), 1.00–0.72 (m, 6H); 13C
NMR (75 MHz, CDCl3, mixture of diastereomers) d 175.7
and 173.9, 148.2 and 148.2, 137.3, 135.8 (4C), 133.5 (2C)
and 132.9 (2C), 129.8 (2C), 127.6 (4C), 115.8, 110.4 and
110.3, 100.0, 75.9, 72.5, 71.3, 70.5, 59.4, 56.3, 53.8, 51.7,
47.2, 45.4, 45.2, 43.7, 43.5, 41.0, 34.6 and 34.3, 30.7,
29.9, 29.4, 28.7 and 28.6, 26.9 and 26.8, 20.5, 20.2, 19.3;
HRMS ES m/z (M+Na)+ calcd 585.3371, obsd 585.3351.
8.4. Unsaturated aldehyde 8
A solution of 20 (1.55 g, 2.00 mmol) in 1:1 MeOH–CH2Cl2
was cooled to ꢀ78 ꢁC, ozonolyzed until a blue color per-
sisted, and purged with oxygen for 15 min. Triphenylphos-
phine (784 mg, 3.00 mmol) was added, the reaction
mixture was warmed to rt, the solvent was evaporated, and
the residue was dissolved in dry ether (125 mL) and repeat-
edly evaporated to dryness. At this point, dry ether (100 mL)
and piperidine (100 mL, 1.00 mmol) were added. After
5 min, AcOH (57 mL, 1.00) was introduced, and the mixture
was heated to reflux for 4 days. Flash column chromato-
graphy of the residue after solvent evaporation (hexane–
The alcohol above (117 mg, 0.21 mmol) was dissolved in
dry CH2Cl2 (5 mL), cooled to 0 ꢁC, and treated sequentially
with triethylamine (0.15 mL, 1.04 mmol), benzoyl chloride
(75 mL, 0.62 mmol), and DMAP (25 mg, 0.21 mmol). The
reaction mixture was stirred for 9 h, quenched with water
(20 mL), and extracted with CH2Cl2 (3ꢂ30 mL). The com-
bined organic layers were dried and evaporated to dryness.
The residue was purified by chromatography on silica gel
(gradient–hexane to hexane–ethyl acetate¼10:1) to give
124 mg (89%) of the benzoate as a colorless oil; IR (neat,
cmꢀ1) 1740, 1722; 1H NMR (250 MHz, CDCl3, mixture of
diastereomers) d 8.08–7.92 (m, 2H), 7.70–7.23 (series of
m, 13H), 5.82–5.62 (m, 1.6H) and 5.53–5.42 (m, 0.4H),
5.05–4.97 (m, 1H), 4.95 (s, 1H), 4.73–4.64 (m, 2H), 3.74
(s, 1.1H) and 3.57 (s, 1.6H), 3.56–3.43 (m, 2H), 2.98 (dd,
J¼10.3, 8.0 Hz, 0.6H), 2.84 (dd, J¼10.7, 5.3 Hz, 0.4H),
2.50–2.61 (m, 0.7H), 2.55 (dd, J¼14.2, 8.0 Hz, 0.4H),
2.44–2.30 (m, 0.5H), 2.22 (dd, J¼13.5, 4.9 Hz, 0.7H),
2.13–1.20 (series of m, 9H), 1.10 (s, 3H), 1.06 (s, 9H),
0.83 (d, J¼6.5 Hz, 3H); 13C NMR (75 MHz, CDCl3, mixture
of diastereomers) d 175.2 and 172.2, 165.7 and 165.5, 148.0
and 147.9, 135.6 (4C), 133.3 (2C), 130.1 and 129.6 (2C),
127.6 and 127.5 (4C), 115.7 and 115.7, 110.5 and 110.3,
74.3, 70.9 and 69.4, 57.4, 53.7, 53.3, 51.8 and 51.5, 46.8,
46.4, 45.1, 44.8, 43.6, 43.0, 42.6, 41.0, 34.4 and 34.1,
30.6, 28.6 and 28.4, 26.8 (3C), 19.3 and 19.0, 19.2; HRMS
ES m/z (M+Na)+ calcd 689.3633, obsd 689.3616.
1
ethyl acetate¼5:1) furnished 0.53 g (36%) of 8; H NMR
(300 MHz, CDCl3, mixture of diastereomers) d 10.06 (s,
0.5H) and 10.05 (s, 0.5H), 8.09–7.88 (m, 2H), 7.72–7.20
(m, 13H), 7.12–7.00 (m, 0.4H), 6.85 (d, J¼8.6 Hz, 1H),
6.75–6.66 (m, 0.4H), 5.72–5.60 (m, 0.4H), 5.56–5.47 (m,
0.4H), 4.57–4.27 (m, 3H), 3.82 (s, 0.4H), 3.78 (s, 2.6H),
3.76–3.70 (m, 0.4H), 3.60 (s, 1H), 3.49 (s, 1H), 3.46–3.30
(m, 1H), 3.28 (d, J¼10.0 Hz, 0.4H), 3.10–1.90 (series of
m, 10.6H), 1.70 (dd, J¼14.5, 2.49 Hz, 0.4H), 1.40–1.20
(m, 1.3H), 1.17 (s, 1.5H), 1.11 (s, 4.5H), 1.08 (s, 1.5H),
1.05 (s, 4.5H); 13C NMR (75 MHz, CDCl3, mixture of dia-
stereomers) d 189.0 and 188.9, 174.1 and 171.2, 165.9 and
165.6, 160.2 and 159.3, 143.3 and 143.0, 135.7 (4C),
133.1, 130.3, 130.2, 130.1, 129.9, 129.8, 129.7 (2C), 129.6
(2C), 129.5, 129.3, 129.2, 128.4, 127.8, 113.8, 99.7, 83.8,
77.2 and 73.6, 71.7 and 71.5, 70.1 and 69.7, 57.1, 55.2,
53.3, 52.0 and 51.8, 47.0 and 45.1, 45.9 and 44.2, 43.4 and
43.2, 37.1 and 37.0, 36.5 and 36.4, 26.9 (3C), 25.8 and
25.7, 20.5 and 20.4, 19.4 and 19.3, 19.1 and 18.7; HRMS
ES m/z (M+Na)+ calcd 811.3637, obsd 811.3635.
The benzoyl ester (7.2 g, 10.8 mmol) from above was dis-
solved in a 1:1 CH2Cl2–MeOH mixture (500 mL), cooled
to ꢀ78 ꢁC, and ozonolyzed until a blue color persisted.
After purging with oxygen, triphenylphosphine (4.25 g,
16.2 mmol) was added and the solution was warmed to rt.
The solvent was evaporated and the white solid was dis-
solved in dry benzene (150 mL). The benzene was evapo-
rated again and the remaining solid was redissolved in
dry benzene (600 mL). p-Toluenesulfonic acid (1.08 g,
5.6 mmol) was introduced and the mixture was heated at
75 ꢁC overnight, cooled to rt, and freed of solvent. The resi-
due was purified by flash chromatography on silica gel
8.5. a-Oxygenation of 38 and MOM protection
To a solution of 38 (48.9 mg, 0.12 mmol) in dry THF
(1.5 mL) was added sodium hexamethyldisilazide (141 mL,
0.141 mmol, 1 M in THF) at ꢀ78 ꢁC. The mixture was
stirred for 30 min, treated with the Davis reagent (39.9 mg,
0.15 mmol) dissolved in dry THF (1 mL) in a dropwise man-
ner and quenched after 10 min with a solution of camphor-
sulfonic acid (137 mg, 0.59 mmol) in THF (0.9 mL). The

5800
S. Pichlmair et al. / Tetrahedron 62 (2006) 5791–5802
cooling bath was removed, the white slurry was stirred for
20 min, water (20 mL) and ether were added, and the prod-
uct was extracted into CH2Cl2 (4ꢂ10 mL). The combined
organic layers were dried and evaporated to leave a residue
that was purified by flash chromatography on silica gel (gra-
dient–hexane–ethyl acetate¼7:1 to 4:1) to give a colorless
oil (37.7 mg, 74%) as a mixture of a-hydroxylated diastereo-
mers (de 76%); IR (neat, cmꢀ1) 1783, 1697, 1247; 1H NMR
(300 MHz, CDCl3) d 7.59–7.13 (m, 7H), 6.87 (dt, J¼8.6,
2.8 Hz, 2H), 5.07 (br d, J¼7.2 Hz, 1H), 4.78–4.70 (m,
1H), 4.43 (dd, J¼15.4, 11.5, 2H), 4.27–4.19 (m, 2H), 3.80
(s, 3H), 3.44–3.13 (m, 3H), 3.19 (dd, J¼13.5, 3.2 Hz, 1H),
2.83 (dd, J¼13.4, 9.3 Hz, 1H), 2.03–1.46 (series of m,
5H), 0.98 (d, J¼6.7 Hz, 3H); 13C NMR (75 MHz, CDCl3)
d 175.1, 159.1, 153.2, 134.9, 130.8, 129.5 (2C), 129.3
(2C), 129.1 (2C), 127.5, 113.7 (2C), 72.5, 69.4, 68.3, 66.9,
55.5, 55.3, 41.3, 37.5, 35.3, 27.1, 20.6; HRMS ES m/z
(M+Na)+ calcd 464.2044, obsd 464.2062; [a]2D0 ꢀ29.2
(c 1.89, CHCl3).
organic layers were dried and evaporated to dryness under
high vacuum to leave a residue that was purified by flash
chromatography on silica gel (gradient–hexane–ethyl
acetate¼25:1 to 10:1) to give 1.17 g (86%) of 41; IR (neat,
cmꢀ1) 1716, 1613, 1514; 1H NMR (300 MHz, CDCl3)
d 7.24 (d, J¼9.0 Hz, 2H), 6.87 (d, J¼9.0 Hz, 2H), 5.84–
5.67 (m, 1H), 5.06–4.94 (m, 2H), 4.59 (dd, J¼10.1,
6.9 Hz, 2H), 4.41 (dd, J¼11.6, 10.1 Hz, 2H), 4.07 (dd,
J¼8.1, 5.4 Hz, 1H), 3.80 (s, 3H), 3.57–3.48 (m, 2H), 3.33
(s, 3H), 2.47 (t, J¼7.2 Hz, 2H), 2.04 (q, J¼4.4 Hz, 2H),
1.38–1.33 (m, 5H), 0.93 (d, J¼6.7 Hz, 3H); 13C NMR
(75 MHz, CDCl3) d 138.0, 130.6, 129.2 (2C), 115.2, 113.7
(2C), 96.5, 81.0, 72.6, 67.9, 56.1, 55.3, 39.4, 37.2, 35.7,
33.1, 26.7, 22.3, 20.3; HRMS ES m/z (M+Na)+ calcd
401.2298, obsd 401.2301; [a]2D0 ꢀ24.2 (c 1.54, CHCl3).
8.7. Coupling of 41 to 11. Deprotection of 43
Ketone 41 (538 mg, 1.42 mmol) was dissolved in dry THF
(6.7 mL) and potassium hexamethyldisilazide (2.84 mL,
1.42 mmol, 0.5 M in toluene) was added at ꢀ78 ꢁC within
20 s. The mixture turned red after the addition of the first
drops and finally changed to an orange solution after a few
minutes. Stirring was maintained at ꢀ78 ꢁC for 1 h. A solu-
tion of 11 (500 mg, 1.18 mmol) in dry THF (3.6 mL) pre-
cooled at ꢀ78 ꢁC was transferred in and stirring was
maintained for 5 min prior to quenching with saturated
NH4Cl solution (20 mL). After extraction with ether
(3ꢂ20 mL), the combined organic layers were dried and
evaporated to leave an oil that was purified by rapid filtration
chromatography through silica gel to afford 716 mg of 43.
The above product (246 mg, 0.56 mmol) was dissolved in
dry CH2Cl2 (6 mL) and diisopropylethylamine (573 mL,
3.35 mmol) was added. The reaction mixture was cooled to
0 ꢁC and MOMCl (252 mL, 3.35 mmol) was introduced and
stirring was maintained overnight. Half-saturated NH4Cl
solution (5 mL) was introduced, the aqueous layer was ex-
tracted with CH2Cl2 (4ꢂ4 mL), and the combined organic
layers were filtered and evaporated to dryness. The residue
was purified by flash chromatography on silica gel (gradi-
ent–hexane–ethyl acetate¼7:1 to 4:1) to provide 180 mg
(67%) of 39 as a single diastereomer; IR (neat, cmꢀ1
)
1
1780, 1708, 1642; H NMR (300 MHz, CDCl3) d 7.49–
7.13 (m, 7H), 6.86 (d, J¼8.6 Hz, 2H), 5.36 (dd, J¼7.4,
5.5 Hz, 1H), 4.77 (d, J¼6.9 Hz, 1H), 4.57–4.20 (m, 4H),
4.42 (dd, J¼17.4, 11.5 Hz, 2H), 4.16 (d, J¼5.2 Hz, 2H),
3.79 (s, 3H), 3.54–3.44 (m, 2H), 3.38 (s, 3H), 3.32 (dd,
J¼13.1, 3.0 Hz, 1H), 2.79 (dd, J¼13.4, 9.5 Hz, 1H), 2.00–
1.78 (m, 2H), 1.72–1.57 (m, 3H), 1.46–1.30 (m, 1H), 0.96
(d, J¼6.6 Hz, 3H); 13C NMR (75 MHz, CDCl3) d 173.8,
159.0, 153.0, 135.2, 130.9, 129.5 (2C), 129.2 (2C), 129.0
(2C), 127.4, 113.7 (2C), 97.5, 74.5, 72.5, 68.0, 66.4, 56.3,
55.3 (2C), 40.4, 37.5, 35.3, 26.8, 20.5; HRMS ES m/z
(M+Na)+ calcd 508.2306, obsd 508.2320; [a]2D0 ꢀ27.2
(c 1.24, CHCl3).
Coupling product 43 (1.28 g, 1.60 mmol) was dissolved
in 10:1 CH2Cl2–H2O (47 mL) and DDQ (544 mg, 2.40
mmol) was introduced. The reaction mixture turned green-
black within seconds and after some minutes became intense
orange in color. The product was extracted into CH2Cl2
(3ꢂ20 mL), and the combined organic layers were dried
and evaporated to leave a residue that was purified by col-
umn chromatography on silica gel (gradient–hexane–ethyl
acetate¼8:1 to 3:1) to give a total of 826 mg (76%) of
pure carbinol as a colorless oil, along with 119 mg of a mixed
fraction which was further purified to give 878 mg (81%) of
50; IR (neat, cmꢀ1) 3512, 1751, 1726; 1H NMR (300 MHz,
CDCl3) d 7.58–7.18 (m, 4H), 7.50–7.34 (m, 6H), 5.73–5.55
(m, 1H), 5.00–4.90 (m, 2H), 4.55 (dd, J¼12.3, 6.8 Hz, 2H),
4.00 (t, J¼4.0 Hz, 1H), 3.80–3.55 (m, 5H), 3.73 (s, 3H),
3.49–3.30 (m, 2H), 3.30 (s, 3H), 2.94 (dd, J¼8.1, 5.7 Hz,
1H), 2.53 (dd, J¼17.6, 1.3 Hz, 1H), 2.30 (d, J¼17.6 Hz,
1H), 2.10–1.18 (m, 9H), 1.14 (s, 3H), 1.03 (s, 9H), 0.89 (d,
J¼6.5 Hz, 3H); 13C NMR (75 MHz, CDCl3) d 212.9,
208.8, 170.5, 137.2, 135.8 (4C), 132.6, 132.4, 130.0 (4C),
127.8 (2C), 115.6, 96.3, 80.3, 68.3, 60.3, 60.2, 56.2, 52.6,
51.7, 48.0, 45.9, 43.0, 38.8, 37.9, 30.3, 29.7, 29.2, 26.7
(3C), 26.1, 24.2, 20.3, 19.1; HRMS ES m/z (M+Na)+ calcd
703.3642, obsd 703.3637; [a]2D0 ꢀ18.4 (c 4.12, CHCl3).
8.6. Ketone 41
5-Bromopentene (2.19 mL, 19.7 mmol) was added to a sus-
pension of activated Mg turnings (478 mg, 19.7 mmol) in
dry ether (27 mL) and a crystal of I2. After 1 h, a solution
of 40 in ether cooled to ꢀ78 ꢁC was introduced via cannula.
The reaction mixture was stirred for 30 min at ꢀ78 ꢁC and
for 20 min at 0 ꢁC prior to quenching with saturated
NH4Cl solution and extraction with ether. The residue was
purified by flash chromatography on silica gel (gradient–
hexane–ethyl acetate¼7:1 to 5:1) to give the alcohol
(1.37 g, 85%). The alcohol (1.37 g) was immediately oxi-
dized to 41. Dissolution in dry CH2Cl2 (42 mL) and pyridine
(3.2 mL, 31.0 mmol) preceded addition of the Dess–Martin
periodinane (3.43 mg, 9.0 mmol). The reaction mixture was
stirred overnight and quenched with a 1:1 Na2S2O3–
NaHCO3 solution (40 mL). The biphasic mixture was stirred
for 3 h, and extracted with ether (3ꢂ30 mL). The combined
8.8. Diquinane 55
Diketone 54 (79.3 mg, 0.094 mmol) was dissolved in MeOH
(4.8 mL), potassium carbonate (172.9 mg, 1.25 mmol) was
added, and the suspension was stirred at rt for 4 h until the
reaction mixture was quenched with saturated NH4Cl

S. Pichlmair et al. / Tetrahedron 62 (2006) 5791–5802
5801
solution (5 mL), and extracted with ether (4ꢂ3 mL). The
combined organic layers were dried and evaporated to dry-
ness. The crude product was purified by flash chromato-
graphy on silica gel (gradient–hexane–ethyl acetate¼2:1 to
1:1) to yield 55 (48.1 mg, 80%); IR (neat, cmꢀ1) 3431,
d 212.6, 208.5, 202.2, 170.3, 137.2, 135.8 (4C), 132.4,
132.0, 130.2 (2C), 127.9 (4C), 115.6, 69.0, 60.8, 52.6,
52.2, 52.1, 50.9, 49.7, 42.7, 42.5, 31.0, 30.9, 29.7, 27.3,
27.0 (3H), 24.8, 19.6, 18.9; HRMS ES m/z (M+Na)+ calcd
641.3274, obsd 641.3297; [a]2D0 ꢀ40.0 (c 1.03, CHCl3).
1
1708, 1218; H NMR (500 MHz, CDCl3) d 7.65–7.55 (m,
4H), 7.48–7.34 (m, 6H), 5.73–5.62 (m, 1H), 5.0–4.90 (m,
2H), 3.78–3.58 (m, 3H), 3.64 (s, 3H), 3.38–3.30 (m, 2H),
2.50 (t, J¼6.5 Hz, 1H), 2.40 (d, J¼15.5 Hz, 1H), 2.34 (dd,
J¼12.5, 7.0 Hz, 1H), 2.10–1.95 (m, 2H), 1.84 (dt, J¼6.5,
6.5 Hz, 1H), 1.72–1.61 (m, 1H), 1.62–1.48 (m, 2H), 1.40–
1.30 (m, 1H), 1.08 (s, 9H), 1.06 (s, 3H), 0.87 (d,
J¼5.6 Hz, 3H); 13C NMR (75 MHz, CDCl3) d 210.9,
175.4, 167.3, 157.7, 147.8, 137.5, 135.7 (2C), 132.9 (2C),
129.8, 127.7 (2C), 115.4, 69.5, 60.5, 52.0, 51.3, 47.7, 42.4,
38.9, 38.2, 32.8, 30.7, 30.0, 29.7, 29.6, 26.9 (3C), 20.6,
20.1, 19.3; HRMS ES m/z (M+Na)+ calcd 657.3224, obsd
657.3206; [a]2D0 ꢀ31.12 (c 1.12, CHCl3).
8.10. Asymmetric dihydroxylation of 63
AD-mix-a [1.19 g containing 2.0 mg of K2OsO4$2H2O,
6.5 mg of (DHQ)2–PHAL, 0.35 g of K2CO3, and 0.83 g of
K3Fe(CN)6] was added to 8.6 mL of a 1:1 H2O–tert-butyl
alcohol solvent system. The mixture was stirred vigorously
at rt until dissolution was complete. Methanesufonamide
(81 mg, 0.85 mmol) was introduced, the reaction mixture
was cooled to 0 ꢁC, and 63 (200 mg, 0.85 mmol) was added
in one portion. After 20 h in the cold, the reaction mixture
was quenched with solid Na2SO3 (1.30 g, 12.62 mmol),
allowed to warm to rt for 1 h, and extracted with ethyl ace-
tate (5ꢂ10 mL). The combined organic extracts werewashed
with 2 M KOH (5 mL), dried, and concentrated prior to flash
chromatography on silica gel (hexane–ethyl acetate¼2:3).
There was isolated 230 mg (99%) of 64 as a colorless oil;
8.9. Aldehyde 58
Alcohol 50 (249.9 mg, 0.368 mmol) was dissolved in a 2:1
THF–MeOH mixture (13 mL) and SmI2 was added until
the deep blue color persisted (5.4 mL, 0.54 M suspension
in THF). Stirring was continued for another 15 min, and
the reaction mixturewas quenched with a 1:1 THF–H2O mix-
ture (10 mL) when the color disappeared. Then HCl (2 mL,
2 N solution) followed by water (30 mL) was added and
the mixture was extracted with CH2Cl2 (3ꢂ20 mL). The
combined organic layers were dried and evaporated to leave
a residue that was purified by flash chromatography on silica
gel (gradient–hexane–ethyl acetate¼4:1 to 3:1) to afford
189.4 (83%) of pure product; IR (neat, cmꢀ1) 3513, 1757,
1
IR (neat, cmꢀ1) 3460, 1454, 1110; H NMR (300 MHz,
CDCl3) d 7.40–7.25 (m, 5H), 4.63 (s, 2H), 4.57 (s, 2H),
3.83 (dt, J¼6.3, 4.0 Hz, 1H), 3.73–3.61 (m, 2H), 3.59 (d,
J¼9.7 Hz, 1H), 3.49 (d, J¼9.7 Hz, 2H), 3.36 (s, 3H), 3.03
(s, 1H), 2.96 (d, J¼4.4 Hz, 1H), 1.20 (s, 3H); 13C NMR
(75 MHz, CDCl3) d 137.7, 128.5 (2C), 127.8, 127.7 (2C),
99.7, 96.9, 73.6, 73.2, 73.0, 72.7, 71.1, 55.3, 21.0; HRMS
ES m/z (M+Na)+ calcd 293.1359, obsd 293.1356; [a]2D0
ꢀ10.1 (c 1.25, CHCl3).
1
1728, 1428; H NMR (300 MHz, CDCl3) d 7.72–7.54 (m,
Acknowledgements
4H), 7.52–7.33 (m, 6H), 5.65–5.48 (m, 1H), 4.99–4.86 (m,
2H), 3.78–3.64 (m, 1H), 3.70 (s, 3H), 3.59 (d, J¼10.4 Hz,
1H), 3.45–3.37 (m, 1H), 3.37 (d, J¼10.4 Hz, 1H), 2.83–
2.70 (m, 2H), 2.47 (d, J¼18.0 Hz, 1H), 2.33 (d, J¼18.0 Hz,
1H), 2.40–2.30 (m, 1H), 1.90–1.15 (series of m, 9H), 1.13
(s, 3H), 1.06 (s, 9H), 0.88 (d, J¼5.8 Hz, 3H); 13C NMR
(75 MHz, CDCl3) d 213.5, 208.7, 170.4, 137.3, 136.2,
135.7, 132.9, 132.2, 130.2, 127.9, 127.7, 115.5, 69.0, 60.8,
60.5, 52.6, 52.3, 52.1, 49.6, 43.2, 39.6, 30.9, 29.4, 28.8,
26.9, 24.8, 19.4, 17.9; HRMS ES m/z (M+Na)+ calcd
643.3451, obsd 643.3599; [a]2D0 ꢀ40.09 (c 1.39, CHCl3).
S.P. is indebted to the Austrian Science Fund (FWF) for the
award of an Erwin Schrodinger postdoctoral fellowship
¨
(project number: J2402).
Supplementary data
Experimental details for all compounds except for 8, 9, 16,
19, 39, 41, 43, 50, 55, 58, and 56. Supplementary data asso-
ciated with this article can be found in the online version, at
doi:10.1016/j.tet.2006.03.066.
The above alcohol (170.9 mg, 0.275 mmol) was dissolved in
dry CH2Cl2 (34 mL), Dess–Martin periodinane (209.1 mg,
0.550 mmol) was added, the mixture was stirred for 1.5 h,
quenched by the addition of saturated 1:1 Na2S2O3–
NaHCO3 solution (10 mL), and extracted with ether
(3ꢂ25 mL). The combined organic layers were dried and
evaporated to give a residue that was purified by flash chro-
matography on silica gel (hexane–ethyl acetate¼4:1) to give
pure 58 (164.6, 91%); IR (neat, cmꢀ1) 1757, 1726, 1428; 1H
NMR (300 MHz, CDCl3) d 9.73 (t, J¼2.0 Hz, 1H), 7.72–
7.52 (m, 4H), 7.52–7.32 (m, 6H), 5.66–5.47 (m, 1H),
5.96–4.84 (m, 2H), 3.72–3.62 (m, 1H), 3.67 (s, 3H), 3.58
(d, J¼10.5 Hz, 1H), 3.94–3.86 (m, 1H), 3.36 (d,
J¼10.5 Hz, 1H), 2.81–2.67 (m, 2H), 2.46 (d, J¼18.0 Hz,
1H), 2.42–2.30 (m, 2H), 2.23 (dd, J¼7.7, 5.4 Hz, 1H),
2.60–1.15 (series of m, 6H), 1.11 (s, 3H), 1.07 (s, 9H),
0.91 (d, J¼6.7 Hz, 3H); 13C NMR (75 MHz, CDCl3)
References and notes
1. Pichlmair, S.; de Lera Ruiz, M.; Basu, K.; Paquette, L. A.
Tetrahedron 2006, 62, 5178.
2. Renner, M. K.; Jensen, P. R.; Fenical, W. J. Org. Chem. 2000,
65, 4843.
3. For an alternative Diels–Alder approach, consult Araki, K.;
Saito, K.; Arimoto, H.; Uemura, D. Angew. Chem., Int. Ed.
2004, 43, 81.
4. (a) Ley, S. V.; Murray, P. J.; Palmer, B. D. Tetrahedron 1985,
41, 4765; (b) Trost, B. M.; Fleitz, F. J.; Watkins, W. J. J. Am.
Chem. Soc. 1996, 118, 5146.
5. (a) Yamamoto, K.; Hentemann, M. F.; Allen, J. G.; Danishef-
sky, S. J. Chem.—Eur. J. 2003, 9, 3242; (b) Mouaddib, A.;
Benoit, J.; Hasnaoui, A.; Merour, J.-Y.; Leonce, S. Hetero-
cycles 1999, 51, 2127; (c) Conia, J. Bull. Soc. Chim. Fr.

5802
S. Pichlmair et al. / Tetrahedron 62 (2006) 5791–5802
1954, 690; (d) Cory, R. M.; Burton, L. P. J.; Chan, D. M. T.;
McLaren, F. R.; Fred, R.; Rastall, M. H.; Renneboog, R. M.
Can. J. Chem. 1984, 62, 1908.
13. Fleming, I.; Goldhill, J.; Paterson, I. Tetrahedron Lett. 1979,
3209.
14. (a) Liu, G.; Wang, Z. Synthesis 2001, 119; (b) Allen, J. G.;
Danishefsky, S. J. J. Am. Chem. Soc. 2001, 123, 351.
15. De Lera Ruiz, M. Unpublished observations in this laboratory.
16. Kametani, T.; Kawamura, K.; Tsubuki, M.; Honda, T. J. Chem.
Soc., Perkin Trans. 1 1988, 193.
6. Heckendorn, R.; Allgeier, H.; Baud, J.; Gunzenhauser, W.;
Angst, C. J. Med. Chem. 1993, 36, 3721.
7. (a) Kotake, Y.; Iijima, A.; Yoshimatsu, K.; Tamai, N.; Ozawa,
Y.; Koyanagi, N.; Kyosuke, K.; Nomura, H. J. Med. Chem.
1994, 37, 1616; (b) Ho, W.; Tutwiler, G. F.; Cottrel, S. C.;
Morgans, D. J.; Tarhan, O.; Mohrbacher, R. J. J. Med. Chem.
1986, 29, 2184.
17. Hase, T.; Wahala, K. Encyclopedia of Reagents for Organic
¨ ¨ ¨
Synthesis; Paquette, L. A., Ed.; Wiley: Chichester, UK, 1995;
Vol. 7, pp 4533–4534.
8. (a) Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem. Soc.
1982, 104, 1737; (b) Wang, Z. Tetrahedron Lett. 1991, 32,
4631.
9. George, S. Y.; Yuan, L.-C.; Jakovac, I. J.; Jones, J. B. Tetra-
hedron 1984, 40, 1235.
18. Gage, J. R.; Evans, D. A. Org. Synth. Coll. 1993, VIII, 528.
19. Chen, B.-C.; Davis, F. A. Encyclopedia of Reagents for
Organic Synthesis; Paquette, L. A., Ed.; Wiley: Chichester,
UK, 1995; Vol. 6, pp 4054–4057.
20. Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155.
21. Hiranuma, S.; Hudlicky, T. Tetrahedron Lett. 1982, 23, 3431.
22. Kieczykowski, G. R.; Schlessinger, R. H. J. Am. Chem. Soc.
1978, 100, 1938.
23. Mechelke, M. F.; Wiemer, D. F. J. Org. Chem. 1999, 64, 4821.
24. Kolb, H. C.; Van Nieuwenhze, M. S.; Sharpless, K. B. Chem.
Rev. 1994, 94, 2483.
25. Heathcock, C. H.; Ratcliffe, R. J. Am. Chem. Soc. 1971, 93,
1746.
10. Evans, D. A.; Chapman, K. T. Tetrahedron Lett. 1986, 27,
5939.
11. (a) Mergott, D. J.; Frank, S. A.; Roush, W. R. Org. Lett. 2002, 4,
3157; (b) Hagiwara, H.; Kobayashi, K.; Miya, S.; Hoshi, T.;
Suzuki, T.; Okamoto, T.; Kobyashi, M.; Yamamoto, I.;
Oktsubo, S.; Kato, M.; Uda, H. J. Org. Chem. 2002, 67, 5969.
12. Corey, E. J.; Desai, M. C.; Engler, T. A. J. Am. Chem. Soc.
1985, 107, 4339.