Rapid Assembly of Polycyclic Substances by a Multicomponent Cascade (4 + 2)-(2 + 2) Cycloadditions: Total Synthesis of the Proposed Structure of Paesslerin A

Article abstract of DOI:10.1021/ja039808k

A versatile method for stereoselective formation of multisubstituted bicyclo[4.2.0]octane framework has been developed by means of catalytic (4 + 2)-(2 + 2) cycloaddition reactions. Its application to the synthesis of a substance reported to be the cytotoxic sesquiterpene, paesslerin A, is described, and it has been made clear that a revision of the structure of natural paesslerin A is required. Copyright

Full text of DOI:10.1021/ja039808k

Published on Web 01/20/2004
Rapid Assembly of Polycyclic Substances by a Multicomponent Cascade
(4 + 2)-(2 + 2) Cycloadditions: Total Synthesis of the Proposed Structure of
Paesslerin A
Kazato Inanaga, Kiyosei Takasu,* and Masataka Ihara*
Department of Organic Chemistry, Graduate School of Pharmaceutical Sciences, Tohoku UniVersity,
Aobayama Sendai 980-8578, Japan
Received November 26, 2003; E-mail: mihara@mail.pharm.tohoku.ac.jp
Table 1. Cascade (4 + 2)-(2 + 2) Cycloadditiona
A large variety of organic substances having complicated
polycyclic structures and a wide assortment of stereogenic centers
are found in nature. These compounds often exhibit attractive and
specific biological activities. From the synthetic chemists’ point
of view, ideal strategies for preparing these structurally complex
substances would involve sequences in which stereocontrolled
formation of multiple carbon-carbon bonds occur in a single step
1
2
3
entry
diene (R , R )
acrylate (R )
product
yield (%)
drb,c
starting with simple, readily available materials. As a result, great
attention has been given to the development of multicomponent
reactions (MCR)1 because of their high degree of atom economy
and their applications in combinatorial chemistry as well as
diversity-oriented syntheses. Despite the intense interest in MCR,
only a limited effort has been given to applying these processes to
the synthesis of stereochemically complex polycyclic compounds.2
Recent studies in our laboratory have led to development of a
hard Lewis acid (e.g., EtAlCl2)-catalyzed, intermolecular Michael
aldol-like, (2 + 2)-cycloaddition reaction.3 The process affords
substituted cyclobutanes starting with silyl enol ethers and R,â-
unsaturated esters. In an extension of this work, we envisaged that
a Lewis acid-promoted cascade process involving sequential Diels-
Alder reaction and (2 + 2)-cycloaddition between 2-siloxydiene
and R,â-unsaturated ester partners would provide complex poly-
cyclic products (Scheme 1). In this communication, we describe
the results of an investigation of this novel catalytic cascade (4 +
2)-(2 + 2) cycloaddition process, which leads to the construction
of bicyclo[4.2.0]octanes, and its application to the synthesis of a
substance reported to be the cytotoxic sesquiterpene paesslerin A
(1) by Palermo and co-workers.4
1
2
3
4
5
6
2a (H, TBS)
2a
2b (Me, TBS)
2b
2c (Me, TIPS)
2c
3a (Me)
3b (CH(CF3)2)
3a
3b
3a
3b
4a
4b
4c
4d
4e
4f
64
66
82
79
79
75
88:9:3
88:12:0
85:15:0
91:9:0
94:6:0
97:3:0
a
Reaction conditions: 2 (1 equiv), 3 (4 equiv), 50 mol % EtAlCl2,
b
CH2Cl2, -78 °C, 1h. Stereochemistry was not determined for the minor
isomers. Diastereomeric ratio was determined by H NMR.
c
1
Scheme 2
Scheme 1
acrylate (3a) was found to afford a mixture of three diastereomeric
trisubstituted bicyclo[4.2.0]octanes 4a in 64% yield (entry 1). In
contrast, reaction of the 3-methyl-2-siloxydienes 2b and 2c resulted
in more efficient production of 4c and 4d, respectively (entries 3
and 4). The enhanced yields in these cases might be a result of the
higher stabilities of the tetrasubstituted silyl enol ether intermediates.
A slight improvement in diastereoselectivity was observed when
triisopropylsiloxydiene 2c was employed as a reactant (entries 5
and 6). The relative configuration of the major diastereomeric
product 4e was confirmed by using NMR spectroscopy.5 The
excellent levels of diastereoselectivity observed in these cascade
reactions was likely a result of stereo-electronically controlled axial
attack by acrylate in the second Michael aldol-like, (2 + 2)-
cycloaddition step. The stereochemistry at the cyclobutyl ester center
in the products was established in the final aldol addition process.3
To probe an application of this novel MCR to the preparation
of the complex target, paesslerin A (1), cascade (4 + 2)-(2 + 2)
cycloadditons of cyclic siloxydienes were investigated (Scheme 2).
However, reaction of 2-siloxycycloheptadiene 5 with acrylate 3a
did not result in the formation of an MCR product even when it
Initial optimization to carry out desired cascade (4 + 2)-(2 +
2) cycloaddition reactions showed that highest yields were obtained
when 1 equiv of the siloxydiene, 4 equiv of the acrylate, and 0.5
equiv of EtAlCl2 were used for 1 h reactions at -78 °C. When
lesser amounts of either the acrylate or EtAlCl2 were employed,
incomplete reaction was observed. The results of the cascade (4 +
2)-(2 + 2) cycloaddition process are summarized in Table 1.
Reaction of 2-tert-butyldimethyl-siloxybutadiene (2a) with methyl
9
1352
J. AM. CHEM. SOC. 2004, 126, 1352-1353
10.1021/ja039808k CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Scheme 3. Total Synthesis of 1a
revealed that the substances are not identical. The 2D-NMR data
for synthetic 1 are fully consistent with the structure of the target.
Finally, the structure of synthetic 1 was unambiguously assigned
by using X-ray crystallographic analysis of the derived p-bromo-
benzoate 16.13 The results clearly demonstrate that a revision of
the structure of natural paesslerin A is required.
In conclusion, this study has led to the development of a novel
cascade (4 + 2)-(2 + 2) cycloaddition, which allows for the rapid
construction of polysubstituted bicyclo[4.2.0]octanes starting from
three simple components. It is noteworthy that the MCR process
is accompanied by formation of four carbon-carbon bonds and
four stereogenic centers in a single operation. In addition, a short
stereoselective synthesis of a substance proposed paesslerin A (1)
has been accomplished.
Acknowledgment. We thank Dr. J. A. Palermo for his kind
discussion about paesslerin A and Dr. C. Kabuto for crystallographic
characterization. The work was supported by the Grant-in-Aid for
Scientific Research from the Ministry of Education, Culture, Sports,
Science, and Technology, Japan.
a
Reagent: (a) TIPSOTf, NEt3, -78 °C, 90%. (b) Methyl propiolate,
EtAlCl2, -40 °C to rt, 92%. (c) DIBAL-H (5 equiv), -78 °C, 74%. (d)
ClCSOPh, Py, rt, then Bu3SnH, AIBN, 80 °C. (e) t-BuOK, H2O-THF, 70%
for two steps. (f) HOTT,10 Et3N, DMAP, rt, then t-dodecanethiol, 85 °C.
(g) TBAF, reflux, 81% for two steps. (h) Ac2O, cat. Sc(OTf)3, 99%. (i) For
15, BuLi, rt, then p-bromobenzoyl chloride, 41%.
Supporting Information Available: Experimental procedures,
characterization data, observed NOESY correlations of 4e, 7, and 11,
and crystallographic structures of 10 and 16 (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
was conducted at higher temperature. In this case, the desilylated
(4 + 2) adduct 6 was formed as a single diastereomer. A plausible
explanation for this observation is that steric hindrance retards the
second (2 + 2)-cycloaddition step. As expected, cascade (4 + 2)-
(2 + 2) cycloaddition of 5 with methyl propiolate in the presence
of EtAlCl2 did take place to furnish tricyclo[4.3.2.02,5]undecane 7
as a single diastereomer. A higher temperature was required for
this (2 + 2) cycloaddition reaction. 2D-NMR analysis revealed that
the adduct 7 has the same relative stereochemistry as found in the
substance reported to be paesslerin A (1).
The route for synthesis of paesslerin A (1), employing the novel
cascade (4 + 2)-(2 + 2) cycloaddition process, is shown in Scheme
3. Siloxydiene 9 was prepared from the known enone 8.6 The key
cascade reaction of 9 with methyl propiolate generated the tricyclic
intermediate 10 in excellent yield (92%) and with complete
diastereoselectivity. The relative stereochemistry of 10 was deter-
mined by using X-ray crystallography.7 A critical issue in this
sequence was regioselective reduction of 10 at the C-12 ester group.
We anticipated that this would be favored over reduction at the
C-16 ester because of steric crowding of the latter by the bulky
triisopropylsiloxy group at C-5. In the event, treatment of 10 with
DIBAL-H at -78 °C led to selective 1,2-reduction of the ester at
C-12, accompanied by unexpected 1,4-reduction of the cyclobutene
carboxylate to afford 11 in 74% yield.
References
(1) (a) Zhu, J. Eur. J. Org. Chem. 2003, 1133-1144. (b) Hulme, C.; Gore,
V. Curr. Med. Chem. 2003, 10, 51-80. (c) Ugi, I. Pure Appl. Chem.
2001, 73, 187-191. (d) Do¨mling, A.; Ugi, I. Angew. Chem., Int. Ed. 2000,
39, 318-3210. (e) Weber, L.; Illgen, K.; Almstetter, M. Synlett 1999,
366-374. (f) Tietze L. F. Chem. ReV. 1996, 96, 115-136.
(2) Recently, high pressure-promoted (4 + 2)-(4 + 2)-(3 + 2) cascade
cycloaddition was reported: van Berkom, L. W. A.; Kuster, G. J. T.;
Kalmoua, F.; Gelder, R. D.; Scheeren, H. W. Tetrahedron Lett. 2003, 44,
5091-5093.
(3) Takasu, K.; Ueno, M.; Inanaga, K.; Ihara, M. J. Org. Chem. 2004, 69,
517-521.
(4) Brasco, M. F. R.; Seldes, A. M.; Palermo, J. A. Org. Lett. 2001, 3, 1415-
1417.
(5) For detailed NMR assignment of the major diastereomer of 4e, see the
Supporting Information. Stereochemistries of the other major isomer of 4
were inferred on the analogy of 1H and 13C NMR.
(6) Paquette, L. A.; Ham, W. H. J. Am. Chem. Soc. 1987, 109, 3025-3036.
(7) Crystal data for 10. C27H44O5Si, monoclinic, space group C2/c, a )
30.980(9) Å, b ) 8.652(2) Å, c ) 20.436(6) Å, â ) 94.433(4)°, V )
5461.3(2) Å3, Z ) 8, D ) 1.160 g/cm3, R ) 0.045, Rw ) 0.112, GOF )
1.11.
(8) (a) Oba, M.; Nishiyama, K. Tetrahedron 1994, 50, 10193-10200. (b)
Crich, D.; Quintero, L. Chem. ReV. 1989, 89, 1413-1432.
(9) Barton, D. H. R.; Crich, D.; Motherwell, W. B. Tetrahedron 1985, 41,
3901-3924.
(10) Abbreviation of HOTT: S-(1-oxido-2-pyridinyl)-1,1,3,3-tetramethyl-
thiouroniumhexafluorophosphate. Garner, P.; Anderson, J. T.; Dey, S.;
Youngs, W. J.; Galat, K. J. Org. Chem. 1998, 63, 5732-5733.
(11) Ishihara, K.; Kubota, M.; Kurihara, H.; Yamamoto, H. J. Org. Chem.
1996, 61, 4560-4567.
(12) Spectral data for synthetic 1. Colorless oil, IR (neat) 1732 cm-1, 1H NMR
(600 MHz, CDCl3): δ 5.69 (dt, 1H, J ) 7.1, 1.8 Hz), 3.00 (t, 1H, J ) 7.1
Hz), 2.17 (ddd, 1H, J ) 14.3, 11.9, 7.7 Hz), 2.02 (s, 3H), 2.02 (ddd, 1H,
J ) 14.3, 10.1, 5.3 Hz), 1.82 (d, 3H, J ) 1.8 Hz), 1.70 (d, 1H, J ) 7.1
Hz), 1.55 (d, 1H, J ) 14.9 Hz), 1.43 (ddd, 1H, J ) 11.9, 11.4, 5.3 Hz),
1.43 (d, 1H, J ) 14.9 Hz), 1.37 (ddd, 1H, J ) 14.9, 7.1, 2.4 Hz), 1.26
(ddd, 1H, J ) 11.4, 10.1, 7.7 Hz), 1.19 (s, 3H), 1.17 (ddd, 1H, J ) 14.9,
7.1, 2.4 Hz), 0.96 (s, 3H), 0.80 (s, 3H). 13C NMR (150 MHz, CDCl3): δ
169.6, 141.5, 122.2, 85.3, 49.3, 45.3, 38.7, 36.7, 35.7, 35.6, 34.2, 31.8,
29.4, 28.6, 23.9, 21.8, 21.6, LRMS m/z: 262 (M+).
Next, alcohol 11 was transformed into 12 by reductive dehy-
droxylation via the nonisolated intermediate xanthate.8 Hydrolysis
of the ester group in 12 to form 13 was followed by decarboxylation
to generate 14 by using an improved Barton’s method9 employing
HOTT.10 Treatment of 14 with TBAF furnished alcohol 15, which
was converted into the proposed structure of paesslerin A (1) by
using Sc-catalyzed acetylation.11 The sequence for the synthesis
of 1 from 8 was accomplished in 34% overall yield (eight steps).
(13) Crystal data for 16. C22H27BrO2, monoclinic, space group P21/c, a )
7.3901(8) Å, b ) 12.005(1) Å, c ) 22.278(5) Å, â ) 91.6346(5)°, V )
1975.5(4) Å3, Z ) 4, D ) 1.356 g/cm3, R ) 0.038, Rw ) 0.040, GOF )
1.75.
1
Surprisingly, comparisons of the H and 13C NMR data of the
synthetic compound with those reported for the natural product
JA039808K
9
J. AM. CHEM. SOC. VOL. 126, NO. 5, 2004 1353