Synthetic studies toward sordarin: Building blocks for the terpenoid core and for analogues thereof

Article abstract of DOI:10.1021/jo801911s

Avenues to bi- and tricyclic building blocks for the elaboration of sordaricin and its analogues are described. The target molecules were obtained through inter- and intramolecular Diels-Alder reactions of a number of previously unknown cyclopentadienes. Unusual properties of 3-cyanoenones and 1-cyanocyclopen-tadienes have been unveiled and circumvented.

Full text of DOI:10.1021/jo801911s

Synthetic Studies toward Sordarin: Building Blocks for the
Terpenoid Core and for Analogues Thereof
Arnaud Schule´,† Huan Liang,‡ Jean-Pierre Vors,§ and Marco A. Ciufolini*,†,‡
Laboratoire de Synthe`se et Me´thodologie Organiques, CNRS UMR 5181, UniVersite´ Claude Bernard Lyon
1 and Ecole Supe´rieure de Chimie, Physique, Electronique de Lyon, 43, Bd. du 11 NoVembre 1918,
69622 Villeurbanne, France, Department of Chemistry, UniVersity of British Columbia,
2036 Main Mall, VancouVer, BC V6T 1Z1, Canada, and Bayer CropScience SA, Rue Pierre Baizet,
69005 Lyon, France
ciufi@chem.ubc.ca
ReceiVed September 10, 2008
Avenues to bi- and tricyclic building blocks for the elaboration of sordaricin and its analogues are described.
The target molecules were obtained through inter- and intramolecular Diels-Alder reactions of a number
of previously unknown cyclopentadienes. Unusual properties of 3-cyanoenones and 1-cyanocyclopen-
tadienes have been unveiled and circumvented.
Introduction
The incessant onslaught of fungal pathogens renders the
search for new antifungal agents an important priority both in
medicine and in agriculture. Indeed, fungi are responsible for a
number of human pathologies, especially in immunocompro-
mised subjects (AIDS and cancer patients, recipients of organ
transplants),1 as well as for major agricultural losses.2
FIGURE 1. Structure of sordarin (1) and congeners.
The potency and selectivity of 1 have elicited considerable
activity in both the biological and the chemical arena. In the
latter connection, three landmark total syntheses have been
recorded.5 In the medicinal chemistry domain, efforts have been
largely confined to the glycoside segment of the molecule.6
Antifungal agents that operate by novel mechanisms are of
special interest, and in that regard, a remarkable family of natural
products isolated from Sordaria araneosa, and known as the
sordarins3 (Figure 1), have captured significant attention, given
their unique mode of action. Thus, sordarin, 1, and its congeners
block protein synthesis by inhibiting the fungal elongation factor
2.4
(4) (a) Justice, M. C.; Hsu, M. J.; Tse, B.; Ku, T.; Balkovec, J. M.; Schmatz,
D.; Nielsen, J. J. Biol. Chem. 1998, 273, 3148. (b) Capa, L.; Mendoza, A.;
Lavandera, J. L.; Gomez de las Heras, F.; Garcia-Bustos, J. F. Antimicrob. Agents
Chemother. 1998, 42, 2694. (c) Shastry, M.; Nielsen, J.; Ku, T.; Hsu, M.-J.;
Liberator, P.; Anderson, J.; Schmatz, D.; Justice, M. C. Microbiology 2001, 147,
383.
(5) (a) Kato, N.; Kusakabe, S.; Wu, X.; Kamitamari, M.; Takeshita, H.
J. Chem. Soc., Chem. Commun. 1993, 1002. (b) Mander, L. N.; Thomson, R. J.
Org. Lett. 2003, 5, 1321. (c) Mander, L. N.; Thomson, R. J. J. Org. Chem.
2005, 70, 1654. (d) Kitamura, M.; Chiba, S.; Narasaka, K. Chem. Lett. 2004,
33, 942. (e) Chiba, S.; Kitamura, M.; Narasaka, K. J. Am. Chem. Soc. 2006,
128, 6931.
(6) Cf (a) Bueno, J. M.; Chicharro, J.; Fiandor, J. M.; Gomez de las Heras,
F.; Huss, S. Bioorg. Med. Chem. Lett. 2002, 12, 1697. (b) Serrano-Wu, M. H.;
St Laurent, D. R.; Carroll, T. M.; Dodier, M.; Gao, Q.; Gill, P.; Quesnelle, C. A.;
Marinier, A.; Mazzucco, C. E.; Regueiro-Ren, A.; Stickle, T. M.; Wu, D.; Yang,
H.; Yang, Z.; Zheng, M.; Zoeckler, M. E.; Vyas, D. M.; Balasubramanian, B. N.
Bioorg. Med. Chem. Lett. 2003, 13, 1419, and references cited therein.
† Universite´ Claude Bernard Lyon 1 and Ecole Supe´rieure de Chimie,
Physique, Electronique de Lyon.
‡ University of British Columbia.
§ Bayer CropScience SA.
(1) Cf. (a) Rupp, S. Fut. Microbiol. 2007, 2, 141. (b) Klepser, M. E.
Pharmacotherapy 2006, 26, 68. (c) Onyewu, C.; Heitman, J. Anti-Infec. Agents
Med. Chem. 2007, 6, 3. (d) Richardson, M. D.; Warnock, D. W. Fungal Infection:
Diagnosis and Management, 3rd ed.; Blackwell Publishing: Malden, MA, 2003.
(2) Cf. Modern Crop Protection Compounds; Kraemer, W.; Schirmer, U.,
Eds.; Wiley-VCH: Weinheim, Germany, 2007.
(3) (a) Hauser, D.; Sigg, H. P. HelV. Chim. Acta 1971, 54, 1178, and
references cited thereinRecent review (b) Liang, H. Beilst. J. Org. Chem. 2008,
4, 31.
10.1021/jo801911s CCC: $40.75
Published on Web 01/26/2009
2009 American Chemical Society
J. Org. Chem. 2009, 74, 1587–1597 1587

Schule´ et al.
SCHEME 1
FIGURE 2. Structures of known (11 and 12) and yet undocumented
(9 and 10) silyloxycyclopentadienes.
SCHEME 2a
Indeed, little appears to be in the public domain concerning the
structure-activity relationship of the terpenoid core. It is known
that analogue 2, wherein a cyano functionality is present in lieu
of a formyl group, is essentially equipotent to 1.7 It is also
established that the hexose unit may be replaced by a saccharide
analogue or even by a hydrophobic alkyl chain.7 However, the
aglycone of sordarin, which is termed sordaricin, 3, is inactive.
Our involvement in the sordarin area8 was motivated by a
desire not only to achieve a total synthesis, but also to address
the structure-activity relationship of the terpenoid core. We
thus launched a synthetic program that would also generate
building blocks for the assembly of a diversity of analogues of
sordaricin nitrile, 4. Such analogues, shown in Scheme 1 as 5,
would display diverse substituents R1-R4 on the main bicyclic
system and various modifications of rings A and B. Of special
interest were compounds (i) amenable to the construction of
diverse variants of ring A, (ii) possessing a contracted ring B
or no ring B at all, (iii) such that R4 contains an ether linkage
wherein the oxygen atom is connected to the bicyclic core by
a short bridge consisting of 1 and 4 carbon atoms,9 (iv) bearing
H, or an oxygenated functionality, or a cyano substituent, in
lieu of a carboxylic group, at the bridgehead position (R3), and
(v) possibly incorporating small alkyl groups or an oxygenated
functionality as the R1 substituent. Frameworks of general
structure 6 satisfied such desiderata. Accessing the target
molecules from cyclopentadienes 7 and 8 through bimolecular
or intramolecular Diels-Alder (IMDA)10 reactions, respectively,
seemed to be the best solution. At this early stage, we
disregarded issues of absolute stereocontrol, focusing instead
on devising an approach that would be of facile execution. All
chiral compounds described herein are, therefore, racemic.
a (a) Base ) LDA, R-X ) o-Cl-C6H4CH2Br (14a) or allyl bromide
(14b); (b) base ) LDA, DMPU as cosolvent, R-X ) 4-iodo-1-butene (14c)
or chloromethyl ethyl ether (14d); (c) base ) KHMDS (2.4 equiv), R-X
) EtOOC-CN (2.4 equiv), DMPU as cosolvent.
to have been employed only as nucleophiles. For instance, a
cyclopentadiene such as 11 was used in a nucleophilic addition
to a sulfonium salt12 and as a terminating element in cationic
olefin cyclization reactions,13 while one of the type 12 was added
to Eschenmoser’s salt during a sequence leading to the introduc-
tion of an exomethylene group,14 or hydroxyethylated by
reaction with ethylene oxide/TiCl4.15 Preliminary experiments
thus aimed to explore the Diels-Alder reactivity of 9 and 10.
We initially investigated some symmetrically substituted
cyclopentadienes, Diels-Alder reactions of which were unaf-
fected by issues of facial selectivity. The preparation of these
dienes commenced with a double alkylation of the kinetic
enolate of the known 1316 (Scheme 2). The only point worthy
of note here is that the modest yield recorded in the second
alkylation of 14d was probably due to the tendency of the
corresponding enolate to undergo elimination of ethoxide ion
and consequent formation of an exomethylene derivative, which
probably degraded under the conditions of the reaction. Still, it
was pleasing to observe that the second alkylation step produced
at least some of the desired 15d. It should also be noted that
compound 15e was prepared in one step from 13 by double
acylation with the Mander reagent.17
Results
Among cyclopentadienes of general structure 7, dioxygenated
species of the type 9 and 10 (Figure 2) were regarded as
especially useful for the objectives of the present work.
Surprisingly, a search of the Chemical Abstracts Service
database found no recorded instances of such cyclic analogues
of the Danishefsky diene.11 Mono-oxygenated variants such as
11 and12 are known, but they are rare, and their Diels-Alder
chemistry remains unexplored. Indeed, such compounds appear
Enol silylation of enones 15 afforded cyclopentadienes 16-20
(Scheme 3). These substances proved to be sensitive and prone
to desilylation; therefore, they were not extensively purified prior
(7) Tse, B.; Balkovec, J. M.; Blazey, C. M.; Hsu, M.-J.; Nielsen, J.; Schmatz,
D. Bioorg. Med. Chem. Lett. 1998, 8, 2269.
(8) Liang, H.; Schu¨le´, A.; Vors, J.-P.; Ciufolini, M. A. Org. Lett. 2007, 9,
4119.
(9) Interest in such compounds was motivated by the results described in
ref.
(10) The Kato (ref 5a) and Mander (ref 5b) total syntheses of 1 indeed rely
on IMDA routes to the terpenoid nucleus.
(12) Kodpinid, M.; Siwapinyoyos, T.; Thebtaranonth, Y. J. Am. Chem. Soc.
1984, 106, 4862.
(13) Corey, E. J.; Wood, H. B., Jr. J. Am. Chem. Soc. 1996, 118, 11982.
(14) Trost, B. M.; Curran, D. P. J. Am. Chem. Soc. 1981, 103, 7380.
(15) Lalic, G.; Petrowski, Z.; Galonic, D.; Matovic, R.; Saicic, R. N.
Tetrahedron 2001, 57, 583.
(11) Reviews: (a) Danishefsky, S. Acc. Chem. Res. 1981, 14, 400. (b)
Danishefsky, S. J.; DeNinno, M. P. Angew. Chem., Int. Ed. Engl. 1987, 26, 15.
(c) Danishefsky, S.; Kitahara, T.; Schuda, P. F. Org. Synth. 1990, Vol. VII, 312.
(16) Best made as detailed by: Clerici, A.; Pastori, N.; Porta, O. Tetrahedron
2001, 57, 217.
(17) Crabtree, S. R.; Mander, L. N.; Sethi, P. S. Org. Synth. 1991, 70, 256.
1588 J. Org. Chem. Vol. 74, No. 4, 2009

Synthetic Studies toward Sordarin
SCHEME 3a
SCHEME 4
SCHEME 5
a (a) 99% conversion for 16; 91% conv. for 17; 95% conv. for 18; 63%
conv. for 19; 92% conv. for 20. (b) Toluene, 65 °C, 12 h, 53% from 15b
via 17 for 21; 22% from 15c via 18 for 22. (c) Neat, 60 °C, 15 h, 40%
from 15d via 19. (d) Neat, rt, 12 h, 27% from 15e via 20.
to Diels-Alder reaction. In the latter respect, TIPS enol ethers
were sufficiently stable to undergo Diels-Alder reactions under
purely thermal conditions, while TMS and TES ethers were
labile and unsuitable for that purpose.18 Consequently, this study
utilized TIPS enol ethers exclusively. The dienophile destined
to combine with 16-20 would have to incorporate functionality
suitable for the elaboration of ring B of sordarin-like scaffolds
through ionic or radical reactions. A halogen-substituted dieno-
phile such as 2-chloroacrylonitrile nicely satisfied this require-
ment. Gratifyingly, the Diels-Alder reaction of 17-20 with
2-chloroacrylonitrile proceeded normally to furnish the expected
cycloadducts as single isomers, albeit in moderate yield (Scheme
3). Thus, such previously unexplored cyclopentadienes are
competent in Diels-Alder reactions. On the other hand,
sterically hindered diene 16 failed to react. The dienes were
also inert toward less reactive dienophiles such as acrylonitrile
or acrolein at temperatures up to 65 °C.19 No effort was made
to promote such reactions with Lewis acid activators.
desilylation triggered retroaldol fragmentation of the bicyclic
system and consumption of the transient 26 through a multitude
of side reactions (Scheme 4). On the other hand, treatment of
21 and 22 with warm (55 °C) aqueous 4 N HCl in THF
promoted only hydrolysis of the vinyl ether to furnish ketones
27 and 28 (Scheme 5). These proved to be good substrates for
radical cyclization reactions leading to tricyclic sordarin building
blocks 29 and 31.24 As anticipated on the basis of the Beckwith
model,25 the reaction occurred diastereoselectively to give
products wherein the configuration of the methyl-bearing
stereogenic carbon coincides with that of the corresponding
carbon atom of sordarin. This assignment is supported by a
strong nuclear Overhauser effect observed between the methyl
hydrogens and the neighboring exo hydrogen. Contrary to the
case of 21-24, TBAF release of the TIPS group from tricyclic
intermediates 29 and 31 proceeded uneventfully to give alcohols,
30 and 32.
Ozonolytic cleavage of the surviving double bond in such
tricyclic systems permitted the elaboration of apical ether
functionalities of the type found in a number of bioactive
analogues of 1. Thus, reduction of 29 (NaBH4, Scheme 6)
provided the exo alcohol diastereoselectively,26 which was
benzylated prior to ozonolytic formation of aldehyde 35. The
latter was advanced to 36-42 through conventional reaction
sequences, certain aspects of which are nonetheless worthy of
comment. First, debenzylation of 37 and 38 was best ac-
complished with the Pearlman catalyst.27 Second (Scheme 7),
and contrary to the case of 29, NaBH4 reduction of 43, obtained
The structure of the cycloadducts was assigned as 21-2420
on the following basis. First, the regioselectivity anticipated on
electronic grounds21 found support in the diagnostic coupling
constants observed between the bridgehead (Hb) proton and the
neighboring exo (Hx) and endo (He) hydrogens (Jbx ) 3-4 Hz,
Jbe ) 0 Hz).22 We presume that these regioisomers were formed
as the endo diastereomers (Alder rule),23 as shown, because it
seemed unlikely that the reaction might have generated exclu-
sively exo cycloadducts.
Cleavage of the TIPS ether in these adducts and functional-
ization of the bridgehead OH group might have provided
interesting sordarin analogues. However, attempted desilylation
of 21-24 resulted in complete degradation. We suspect that
(18) TMS and TES silyl ethers underwent desilylation back to the starting
enones upon heating in the presence of dienophiles.
(19) An MNDO calculation (Hyperchem package) indicated that the LUMO
energy of 2-chloroacrylonitrile is about 0.5 eV lower than that of acrylonitrile
or acrolein.
(20) Adducts 22 and 24 were not fully characterized. For the series of
compounds emanating from 22, full characterization occurred at the stage of
derivative 28.
(21) This would be consequence of the electronic similarity between the
cyclopentadienes in question and the Danishefsky diene (ref 11).
(22) Bowey, F. A. Nuclear Magnetic Resonance Spectroscopy; Academic
Press: New York, NY, 1969.
(24) Compound 31 was not characterized beyond a crude 1H NMR. In crude
form, it was directly advanced to 32, which was purified and fully characterized.
(25) (a) Beckwith, A. L. J.; Schiesser, C. H. Tetrahedron 1985, 41, 3925.
(b) Beckwith, A. L. J.; Easton, C. J.; Lawrence, T.; Serelis, A. K. Aust. J. Chem.
1983, 36, 545.
(26) As anticipated based on the behavior of the sterically similar
camphor: Brown, H. C.; Muzzio, J. J. Am. Chem. Soc. 1966, 88, 2811.
(27) Pearlman, W. M. Tetrahedron Lett. 1967, 8, 1663.
(23) (a) Alder, K. Justus Liebigs Ann 1951, 571, 157. Theoretical study: (b)
Imade, M.; Hirao, H.; Omoto, K.; Fujimoto, H. J. Org. Chem. 1999, 64, 6697.
J. Org. Chem. Vol. 74, No. 4, 2009 1589

Schule´ et al.
SCHEME 6
SCHEME 9
SCHEME 10
SCHEME 7
ppm; however, none of the new entities were sufficiently potent.
Possible reasons for this may have been the lack of an isopropyl,
other small alkyl group, at the position adjacent to the carbonyl
functionality and/or the absence of the bridgehead COOH group.
Concerning the first issue, we considered generating an enolate
of ketone 27 to effect the desired alkylation either in the aldol
or in the SN2 mode. In practice, this proved not to be possible,
because the ketone resisted deprotonation with LDA. In this
respect, 27 differs from the structurally related camphor, which
undergoes deprotonation normally with LDA. To wit, the
documented aldol addition of the zinc enolate of camphor to
acetone29 was duplicated without difficulty, as was the ensuing
formation of isopropylidene camphor. Evidently, the bridgehead
silyloxy functionality in 27 and related ketones shields the
neighboring methylene group from the action of external bases.
Attempts to achieve alkylation of cycloadducts 21-24 by
harnessing the nucleophilic character of the vinyl ether func-
tionality also failed. For instance (Scheme 9), exposure of 21
to 2,2-dimethoxypropane and catalytic TiCl4 produced none of
the desired 50.30
SCHEME 8
in 98% yield by ozonolysis of 29, was nonstereoselective,
leading to a mixture of epimers of alcohol 45. We believe that
rapid reduction of the formyl group leads to an intermediate
such as 44. Intramolecular hydride delivery to the keto carbonyl
may now compete with the bimolecular reaction, accounting
for the poor diastereoselectivity.
Encouraged by these results, we contemplated the possibility
of inducing a facially selective Diels-Alder reactions of
cyclopentadienes 7 wherein R4 ) COOEt and R5 ) linear alkyl.
To test this hypothesis, 14b was advanced to 46 by sequential
Mander acylation17 and silyl enol ether formation (Scheme 8).
However, no facial selectivity was observed in the reaction of
the emerging cyclopentadiene with 2-chloroacrylonitrile, and
this line of research was abandoned.28
A plausible cure for such ills was to install the desired alkyl
substituent on the starting cyclopentadiene. This possibility was
investigated using diene 54 (Scheme 10). Surprisingly, 54 failed
to combine with 2-chloroacrylonitrile (65 °C, toluene, K2CO3;
(28) Because of this disappointing outcome, compounds 47 and 48 were not
purified to homogeneity or fully characterized.
(29) (a) Lightner, D. A.; Flores, M. J.; Crist, B. V.; Gawronski, J. K. J. Org.
Chem. 1980, 45, 3518. (b) Schenato, R. A.; dos Santos, E. M.; Tenius, B. S. M.;
Costa, P. R. R.; Caracelli, I.; Zukerman-Schpector, J. Tetrahedron: Asymmetry
2001, 12, 579. (c) Frimer, J. J.; Afri, M.; Baumel, S. D.; Gilinsky-Sharon, P.;
Rosenthal, Z.; Gottlieb, H. E. J. Org. Chem. 2000, 65, 1807.
(30) This reaction furnished instead ketal 51, which was not thoroughly
characterized:
Several of the bicyclic and tricyclic compounds shown above
were tested for activity against a broad range of phytopathogenic
fungi. Substances 34 and 35 showed marginal efficacy at 50
1590 J. Org. Chem. Vol. 74, No. 4, 2009

Synthetic Studies toward Sordarin
SCHEME 11
SCHEME 13
SCHEME 12
conditions or upon contact with silica gel. Fortunately, the
Michael step afforded product of excellent quality, enabling the
conduct of a subsequent Baylis-Hillman reaction34 without
purification. The kinetics of the latter process are notoriously
slow, and in fact formation of 64 required 2 days even by
operating in neat acrylonitrile. The product emerged as a ca.
60:40 mixture of two unassigned alcohol diastereomers, which
were not separated, since the OH group is destined to undergo
ultimate oxidation to a ketone. Instead, the mixture was directly
subjected to double silylation in the presence of TIPS-OTf and
Hu¨nig base (CH2Cl2, 0 °C). The 1H NMR spectra of the resultant
reaction mixtures revealed the presence of nearly equimolar
amounts of diene 65 and cycloadduct 66. The latter again
emerged as a single regioisomer (1H NMR), as anticipated on
the basis of bond polarization. Structure 66 was initially assigned
on the basis of the coupling constants observed between protons
Hx, He, and Hb (Jbx ) 4.1 Hz; Jbe ) 0.0 Hz; Jex ) 13.6 Hz).
However, it was also apparent that the cycloadduct had formed
as a 3:1 mixture of unassigned epimers of the lateral OTIPS
group. This signaled that one of the two stereoisomers of 65
(initially a 60:40 mixture of diastereomers) undergoes IMDA
faster than the other. Upon standing at room temperature for
12 h, the diene component in the mixture was completely
converted to the cycloadduct, which was now present as a 60:
40 mixture of isomers. The sequence leading to 66 was
ultimately modified by effecting the silylation step at 0 °C and
then allowing the reaction mixture to stir at 25 °C for 12 h.
This resulted in direct formation of a 60:40 mixture of
diastereomers of 66 (72%).
The application of the same approach to the assembly of
cycloadducts displaying a bridgehead COOH group, or an
equivalent functionality, raised an interesting issue. Derivati-
zation of kinetic enolates of enones 67 as seen thus far would
lead to compounds 68 (Scheme 14). These could undoubtedly
be advanced to structures 69 and thence to the ultimate 70. But
if one could access the thermodynamic enolates of 67 and
thereby produce 71, the preparation of an intermediate displaying
an equivalent of the carboxy group at an appropriate position
would be greatly facilitated. For instance, reaction of 71 with a
source of cyanide ion may result in formation of 72, which could
be expeditiously converted into 70. Fortunately, important work
by Koreeda35 had established that LHMDS deprotonates enones
higher temperatures caused desilylation back to 53). Given that
the steric effect of a mere methyl group seemingly suffices to
suppress the Diels-Alder reactivity of 54, analogous reactions
of dienes incorporating even more sterically demanding alkyls,
such as an isopropyl group, were not pursued.31 On the other
hand, pioneering work by Kato5a and Mander5b,c (Scheme 11)
had demonstrated efficient IMDA reactions of sterically con-
gested or electronically deactivated cyclopentadienes. We thus
refocused on an IMDA approach to the assembly of building
blocks for 1 displaying alkyl and bridgehead carboxy (or
equivalent) groups.
Important observations by Cuevas32 suggest that the presence
of at least a bicyclic system is necessary for activity. We thus
concentrated on the preparation of tricyclic ketones 58 (Scheme
12), which we felt could be elaborated into a variety of
sordaricin congeners, including bicyclo[2.2.1]- (59), bicy-
clo[3.2.1]- (60), tricyclic, and tetracyclic (61) analogues, the
latter in both the homo- and the heterocyclic series. Compounds
58 were envisioned to derive from 62, wherein the indicated
C-C bonds would be established through Michael and
Baylis-Hillman reactions. To test this approach, we targeted
66 (Scheme 13), which incorporates a bridgehead oxygen
functionality while lacking an alkyl substituent corresponding
to the isopropyl group of 1. Thus, the Michael addition of the
known 14e33 to acrolein provided 63 in quantitative yield. The
reaction proceeded best in the presence of 2% DBU. The use of
larger amounts of base caused an apparent stalling of the
reaction, resulting in diminished yields. This appears to be due
to the fact that greater quantities of basic agents accelerate a
competing retro-Michael fragmentation of 63. Indeed, this
aldehyde is a sensitive material that degrades easily under basic
(31) Because of this failure, compound 54 was not purified to homogeneity
or fully characterized.
(32) (a) Cuevas, J. C.; Martos, J. L. Tetrahedron Lett. 1998, 39, 8553. (b)
Cuevas, J. C.; Lavandera, J. L.; Martos, J. L. Bioorg. Med. Chem. Lett. 1999, 9,
103.
(33) (a) Irie, H. Katakawa, J. Tomita, M. Mizuno, Y. Chem. Lett. 1981, 637.
(b) Boschelli, D. Smith, A. B., III. Stringer, O. D. Jenkins, R. H., Jr. Davis,
F. A. Tetrahedron Lett. 1981, 22, 4385. Prepared in 52% yield by reaction of
13 with LHMDS followed by MeO2COCN.
(34) Reviews: (a) Ciganek, E. Org. React. 1997, 51, 201. (b) Basavaiah, D.;
Rao, A. J.; Satyanarayana, T. Chem. ReV. 2003, 103, 811. (c) Awad, L.; Demange,
R.; Zhu, Y.-H.; Vogel, P. Carb. Res. 2006, 341, 1235.
J. Org. Chem. Vol. 74, No. 4, 2009 1591

Schule´ et al.
SCHEME 14
SCHEME 16
SCHEME 15
and dehydration of the intermediate 85 with the Burgess
reagent43 produced 86.
67 in which R ) alkyl to furnish thermodynamic enolates.
Accordingly, the known 7336 was converted into 74, which upon
exposure to LHMDS followed by MeO2C-CN indeed afforded
a product, 75 (81%), formally ensuing through acylation of the
thermodynamic enolate (Scheme 15). The Michael addition of
75 to acrolein proceeded as expected, but for reasons that remain
unclear, the subsequent Baylis-Hillman reaction of the resultant
76 was abnormally slow. Several days were required to convert
the substrate into a nearly 50:50 mixture of alcohol diastereo-
mers of 77 (68% chromatographed yield) under the conditions
previously employed for 63 (neat acrylonitrile, DABCO).
Aggarwal conditions (La(OTf)3 and triethanolamine as cocata-
lysts)37 enabled a 9-fold rate acceleration, but at the cost of a
reduced yield of 77 (50% chromatographed). The product was
obtained as a 65:35 ratio of unassigned alcohol diastereomers.
We were now in a position to evaluate the effect of the isopropyl
group on the IMDA reaction of a silyl enol ether derivative of
77, and to that end, the compound was subjected to double
silylation as detailed earlier. Cycloadducts 78-80 were obtained
without incident.38 In accord with previous findings, TIPS ethers
performed most efficiently in this step (Scheme 16). Further-
more, desilylation of 78 proceeded efficiently to furnish 81 as
a mixture of isopropyl epimers. Swern oxidation39 of the alcohol
surrendered diketone 82, thereby demonstrating a critical
desilylation-oxidation step. More importantly, treatment of the
TBS ether40 of 77 with diethylaluminum cyanide (Nagata
reagent)41 afforded dinitrile 84. Luche reduction42 of the enone
Distressingly, compound 86 resisted IMDA reaction at
temperatures as high as 160 °C, above which it decomposed.
This behavior contrasts with that of 56, which cyclizes easily
and with the correct regioselectivity44 to give 57. In either case,
electron- withdrawing groups are present on both the dienic and
the dienophilic segments of the molecule. Consequently, the
failure of 86 to undergo IMDA reaction is not imputable solely
to an electronic incompatibility between diene and dienophile.
It is known that cyclic cyanodienes are poor components of
Diels-Alder reactions.45 On the other hand, 1-cyanocyclopen-
tadiene seems to be more reactive than acyclic cyanodienes in
that it combines with tetracyanoethylene (admittedly a highly
reactive dienophile) even at -25 °C.46 Consequently, an IMDA
of a cyanocyclopentadiene such as 86 was expected to be even
more facile. We surmised that the failure of this reaction might
be due to an abnormal lowering of the HOMO energy of the
diene operated by the cyano substituent. An indication that this
may be the case emerges from the fact that the Hammett σ
values for a CN group are more positive than those of a COOMe
group.47 However, lack of knowledge about the F value for the
IMDA reaction in question disallows an accurate evaluation of
the effect that this may have on overall kinetics. The matter
was thus addressed computationally48 using COOMe- and CN-
(42) Gemal, A. L.; Luche, J. L. J. Am. Chem. Soc. 1981, 103, 5454.
(43) Atkins, G. M.; Burgess, E. M. J. Am. Chem. Soc. 1968, 90, 4744.
(44) The cyclopentadiene unit in 86 lacks an oxygen substituent that might
direct formation of the correct regioisomer of the adduct. However, 56 produces
57 only (Scheme 11 and ref 5c). More significantly, 2,4-pentadienoic acid reacts
with acrylic acid to give a cycloadduct that displays vicinal COOH groups
(Davalian, D.; Garratt, P.; Koller, W.; Mansuri, M. J. Org. Chem. 1980, 45,
4183). Thus, we anticipated good regioselectivity in the IMDA cyclization of
86.
(35) (a) Koreeda, M.; Liang, Y.; Akagi, H. J. Chem. Soc., Chem. Commun.
1979, 449. (b) Koreeda, M.; Liang, Y. Tetrahedron Lett. 1981, 22, 15.
(36) Majetich, G.; Song, J.-S.; Ringold, C.; Nemeth, G. A.; Newton, M. G.
J. Org. Chem. 1991, 56, 3973, and refs cited therein.
(37) Aggarwal, V. K.; Mereu, A.; Tarver, G. J.; McCague, R. J. Org. Chem.
1998, 63, 7183.
(38) The structure of these molecules again rests on the coupling constants
between protons He, Hx and Hb (Jbx ) 4.4 Hz; Jbe ) 0 Hz; Jex ) 13.1 Hz).
(39) Mancuso, A. J.; Swern, D. Synthesis 1981, 3, 165.
(40) Protection of 77 as, e.g., a TES ether furnished intermediates that reacted
poorly in subsequent steps.
(45) Snyder, H. R.; Poos, G. I. J. Am. Chem. Soc. 1950, 72, 4104.
(46) Banert, K.; Koehler, F.; Meier, B. Tetrahedron Lett. 2003, 44, 3781.
(47) Cf. (a) Smith, M. B.; March, J. March’s AdVanced Organic Chemistry,
5th ed.; John Wiley & Sons: New York, NY, 2001; pp 368 ff.
(48) Calculations were carried out with the Hyperchem package, available
from Hypercube, Inc. (www.hyper.com).
(41) Nagata, W.; Yoshioka, M. Org. React. 1984, 25, 255.
1592 J. Org. Chem. Vol. 74, No. 4, 2009

Synthetic Studies toward Sordarin
SCHEME 17
SCHEME 18
substituted cyclopentadienes 88 and 89 (Scheme 17) as better
tractable mimics of the actual systems. Semiempirical calcula-
tions detected insignificant differences in the FMO energies for
the two molecules.49 However, DFT calculations (6-31G*)
carried out on MM+-optimized structures revealed that the
HOMO energy of 88 is about 1 eV lower than that of the ester
analogue. This may account for the problematic reaction of 86.
A plausible way to circumvent such unfavorable electronic
properties was to attempt the reaction with a silyl enol ether
derivative of 84. The silyloxy substituent was anticipated to raise
the HOMO energy of the diene system, thereby enhancing its
reactivity.
SCHEME 19
To our complete surprise, 84 was recovered unchanged upon
countless attempts to achieve enol silylation under a variety of
conditions,50 even by the application of the Corey-Gross
method.51 The behavior of 84 stood in strident contrast to that
of other cyanoenones, such as 4-oxo-2-pentene-carbonitrile,
which are reported to undergo enol silylation without incident.52
The reasons for such a failure are mystifying. We posited
that 84 might suffer from an inherent barrier to enolization. To
address this hypothesis, the compound was treated with NaH
in THF, whereupon it was slowly converted into a mixture of
products containing 93 (mixture of diastereomers, 29% yield,
chrom.) as the major component (Scheme 18). This material
clearly results from a formal intramolecular Michael addition
of the enolate of the enone to the lateral conjugated nitrile. Of
course, we cannot exclude the possible occurrence of a dienolate
Diels-Alder reaction followed by retro-Michael fragmentation
(cf. 90 to 91 to 92). While the latter step would be disfavored
on stereoelectronic grounds (it corresponds to an anti-Baldwin
retro-5-endotrig process),53 the significant release of strain
associated with the breakup of the bicyclo[2.2.1] system may
well overcome such barriers.
SCHEME 20
Regardless, the formation of 93 proved that the enolate of
84 is indeed accessible. Given the “enormous mechanistic
complexity” of ketone deprotonation,54 it was imprudent to
formulate simplistic explanations as to why enol silylation of
84 failed. On the other hand, the foregoing observations were
consistent with an insufficient kinetic reactivity of the base
employed for deprotonation. A practical way to circumvent such
difficulties was to harness the so-called “LiX effect”;55 that is,
the ability of lithium halides to enhance the kinetic reactivity
of amide bases. Indeed, exposure of 84 to an excess56 of
LHMDS and LiCl in THF-HMPA, in the presence of TBSCl,
delivered the long-sought 94 in 82% yield (Scheme 19).57 In
contrast to its congeners, diene 94 displayed no inclination
whatsoever to cyclize to 95 at room temperature: cyclization
(49) Numerical results are provided as Supporting Information.
(50) Over a range of temperatures, compound 84 was immune to the action
of trialkylsilyl halides/tertiary amines in the presence of Lewis acids) Danishefsky,
S.; Kitahara, T. J. Am. Chem. Soc. 1974, 96, 7807), trialkylsilyl triflates (Review:
Bach, T.; Brummerhop, H. J. Prak. Chem. 1999, 341, 410), bis(trimethylsilyl)
acetamide (Cameron, D. W.; Feutrill, G. I.; Perlmutter, P. Tetrahedron Lett.
1981, 22, 3273), bis(trimethylsilyl) trifluoroacetamide ( Fattori, D.; de Gucht-
eneere, E.; Vogel, P. Tetrahedron Lett. 1989, 30, 7415), or TMSI/HMDS ( Miller,
R. D.; McKean, D. R. Synthesis 1979, 73.
(51) Corey, E. J.; Gross, A. W. Tetrahedron Lett. 1984, 25, 495.
(52) Hosokawa, T.; Aoki, S.; Murahashi, S. Synthesis 1992, 558.
(53) (a) Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734. (b)
Baldwin, J. E.; Thomas, R. C.; Kruse, L. I.; Silberman, L. J. Org. Chem. 1977,
42, 3846.
(55) (a) Seebach, D. Angew. Chem., Int. Ed. Engl. 1988, 27, 1624. (b)
Jackman, L. M.; Rakiewicz, E. F. J. Am. Chem. Soc. 1991, 113, 1202. (c) Bunn,
B. J.; Simpkins, N. S. J. Org. Chem. 1993, 58, 533. (d) Lipshutz, B. H.; Wood,
M. R.; Lindsley, C. W. Tetrahedron Lett. 1995, 36, 4385. (e) Pratt, L. M.;
Newman, A.; St. Cyr, J.; Johnson, H.; Miles, B.; Lattier, A.; Austin, E.;
Henderson, S.; Hershey, B.; Lin, M.; Balamraju, Y.; Sammonds, L.; Cheramie,
J.; Karnes, J.; Hymel, E.; Woodford, B.; Carter, C. J. Org. Chem. 2003, 68,
6387.
(56) A significant excess of Li-base was essential in the present case, even
though excess base may actually inhibit enolization, at least in the case of
esters: Sun, X.; Collum, D. B. J. Am. Chem. Soc. 2000, 122, 2452.
(57) The corresponding TMS enol ether was too labile, while the TES enol
ether afforded poor yields in the subsequent IMDA.
(54) Hall, P. L.; Gilchrist, J. H.; Collum, D. B. J. Am. Chem. Soc. 1991,
113, 9571.
J. Org. Chem. Vol. 74, No. 4, 2009 1593

Schule´ et al.
cooled to -78 °C. A solution of the above compound (2.4 g, 15.5
mmol) in dry THF (8 mL) was introduced dropwise, and the mixture
was stirred at -78 °C for an additional 30 min after all of the ketone
had been added. Neat allyl bromide (3.4 mL, 38.8 mmol) was then
introduced dropwise, and the mixture was stirred for an additional
2 h at -78 °C. The solution was then allowed to warm to 0 °C and
treated with water (5 mL). The THF was removed in vacuo, and
the aqueous residue was extracted with EtOAc (3 × 40 mL). The
combined extracts were washed with saturated aq NaCl solution,
dried (MgSO4), and evaporated. Chromatographic purification of
the residue (15% EtOAc in cyclohexane) gave 2.1 g (10.9 mmol,
had to be forced by heating a toluene solution of 94 in a sealed
tube at 140 °C for 12 h. Fortunately, the expected 95 was
obtained as a single regioisomer, but as a 50:50 mixture of
epimers of the lateral OTBS group, in 77% chromatographed
yield. We consider the reluctance of 94 to undergo IMDA
reaction as another manifestation of the HOMO-lowering
influence of the CN substituent. As seen earlier for similar
compounds, the diagnostic coupling constants observed for
protons He, Hx, and Hb (Jbx ) 3.9 Hz; Jbe ) 0 Hz; Jex ) 13.5
Hz) support the structure of the new cycloadduct. Treatment of
95 with pyridine-HF induced only release of the nonenolic TBS
group. The resultant mixture of diastereomeric alcohols 96
underwent smooth Dess-Martin oxidation to 97 (single isomer),
thereby achieving the preparation of a building block for sordarin
analogues wherein a cyano group replaces the bridgehead
COOH functionality. We believe that intermediates of the type
97 are the key to the exploration of the structure-activity
relationship of the terpenoid core of 1. In addition they may
possibly be useful for a total synthesis of the natural product.
In conclusion, this work has established avenues to intermedi-
ates that could serve as building blocks for the synthesis of
sordaricin and congeners. In the process, the preparation and
the Diels-Alder reactivity of a number of previously unknown
cyclopentadienes have been explored. Unusual properties of
cyano-substituted enones and cyclopentadienes have been
unveiled and circumvented. The successful sequence leading
to tricyclic ketone 97 requires 10 steps from the known dione
73 (available in 3 steps from methyl isobutyl ketone, 98),36 and
it proceeds in an overall yield of 6%. Investigations aiming to
improve the synthesis of compounds such as 97 and to elaborate
them to sordaricin and its analogues continue in our laboratory.
1
70%) of product. H NMR: 5.63 (m, 2H); 5.24 (t, J ) 1.0, 1H);
5.05 (m, 4H); 3.81 (s, 3H); 2.43 (d, J ) 1.0, 2H); 2.36 (ddt, J )
13.6, J ) 6.5, J ) 1.3, 2H); 2.15 (dd, J ) 13.6, J ) 8.2, 2H). 13
C
NMR: 208.5; 189.0; 133.2; 118.2; 103.6; 58.4; 50.4; 41.2; 36.9.
MS (CI): 193 [M + H]+. IR: 1695, 1597. HRCIMS: calcd for
C12H17O2 193.1229, found 193.1230.
Representative Procedure for Alkylation of 13 with Less
Reactive Electrophiles: 5-(1-Buten-4-yl)-3-methoxy-2-cyclopenten-
1-one (14c). A 2.3 M solution of n-BuLi in hexanes (14.0 mL, 32.7
mmol) was added dropwise to a cold (-40 °C) solution of dry
diisopropylamine (4.6 mL, 32.7 mmol) in anhydrous THF (39 mL),
with good stirring under Ar. The mixture was allowed to warm
slowly to -10 °C, then it was cooled to -78 °C. A solution of 13
(2.5 g, 21.8 mmol) and dry DMPU (10.5 mL, 87.1 mmol) in dry
THF (5 mL) was introduced dropwise, and the mixture was stirred
at -78 °C for an additional 40 min after all of the ketone had been
added. A second portion of DMPU (10.5 mL, 87.1 mmol) was
added, and the mixture was stirred for another 5 min. Neat 4-iodo-
1-butene (19.8 g, 109 mmol) was then injected, and the mixture
was stirred for an additional 3 h, during which time it was allowed
to warm to 0 °C. The solution was treated with water (5 mL) and
extracted with EtOAc (3 × 30 mL). The combined extracts were
washed with saturated aq NaCl solution, dried (Na2SO4), and
evaporated. Chromatographic purification of the residue (40%
EtOAc in cyclohexane) gave 2.3 g (13.9 mmol, 64%) of product.
1H NMR: 5.78 (m, 1H); 5.30 (s, 1H); 5.00 (m, 2H); 3.84 (s, 3H);
2.75 (dd, J ) 17.8, J ) 7.1, 1H); 2.51 (m, 1H); 2.29 (dd, J ) 17.8,
J ) 3.0, 1H); 2.12 (m, 2H); 1.97 (m, 1H); 1.45 (m, 1H). 13C NMR:
208.2; 190.2; 138.2; 115.6; 104.1; 59.0; 45.3; 35.1; 31.8; 31.0. MS
(EI): 166 M+, 151, 112. HREIMS: calcd for C10H14O2 166.0994,
found 166.0994.
5,5-Bis(1-buten-4-yl)-3-methoxy-2-cyclopenten-1-one (15c). A 2.3
M solution of n-BuLi in hexanes (5.6 mL, 13.0 mmol) was added
dropwise to a cold (-40 °C) solution of dry diisopropylamine (1.9
mL, 13.0 mmol) in anhydrous THF (17 mL), with good stirring
under Ar. The mixture was allowed to warm slowly to -10 °C,
then it was cooled to -78 °C. A solution of the above compound
(1.8 g, 10.9 mmol) and dry DMPU (5.2 mL, 43.4 mmol) in dry
THF (5 mL) was introduced dropwise, and the mixture was stirred
at -78 °C for an additional 40 min after all of the ketone had been
added. A second portion of DMPU (5.2 mL, 43.4 mmol) was added,
and the mixture was stirred for another 5 min. Neat 4-iodo-1-butene
(9.9 g, 54.2 mmol) was then injected and the mixture was stirred
for an additional 3 h, during which time it was allowed to warm to
0 °C. The solution was treated with water (3 mL), and the THF
was removed in vacuo. The aqueous residue was extracted with
EtOAc (3 × 30 mL). The combined extracts were washed with
saturated aq NaCl solution, dried (Na2SO4), and evaporated.
Chromatographic purification of the residue (40% EtOAc in
cyclohexane) gave 792 mg (3.6 mmol, 33%) of product. 1H NMR:
5.76 (m, 2H); 5.28 (s, 1H); 4.96 (m, 4H); 3.84 (s, 3H); 2.49 (s,
2H); 1.94 (m, 4H); 1.61 (m, 4H). 13C NMR: 209.9; 189.6; 138.6;
115.0; 104.6; 59.0; 51.1; 39.1; 36.6; 28.9. MS (CI): 221 [M + H]+.
HRCIMS: calcd for C14H21O2 221.1542, found 221.1544.
Experimental Section58
Representative Procedure for the Alkylation of 13 with Reac-
tive Electrophiles: 5-Allyl-3-methoxy-2-cyclopenten-1-one (14b). A
2.5 M solution of n-BuLi in hexanes (10.7 mL, 26.8 mmol) was
added dropwise to a cold (-40 °C) solution of dry diisopropylamine
(3.8 mL, 26.8 mmol) in anhydrous THF (36 mL), with good stirring
under Ar. The mixture was allowed to warm slowly to -10 °C,
then it was cooled to -78 °C. A solution of 13 (2.5 g, 22.3 mmol)
in dry THF (8 mL) was introduced dropwise, and the mixture was
stirred at -78 °C for an additional 30 min after all of the ketone
had been added. Neat allyl bromide (4.8 mL, 55.8 mmol) was then
introduced dropwise, and the mixture was stirred for an additional
2 h at -78 °C. The solution was then allowed to warm to 0 °C and
treated with water (5 mL). The THF was removed in vacuo, and
the aqueous residue was extracted with EtOAc (3 × 40 mL). The
combined extracts were washed with saturated aq NaCl solution,
dried (MgSO4), and evaporated. Chromatographic purification of
the residue (30% EtOAc in cyclohexane) gave 2.4 g (16.1 mmol,
1
72%) of product. H NMR: 5.73 (m, 1H); 5.27 (t, 1H, J ) 1.1);
5.05 (m, 2H); 3.82 (s, 3H); 2.69 (ddd, 1H, J ) 17.5, 7.1, 1.1); 2.57
(m, 2H); 2.32 (ddd, 1H, J ) 17.5, 2.6, 1.1); 2.17 (m, 1H). 13C
NMR: 207.5; 190.4; 135.6; 117.4; 104.1; 59.0; 44.9; 35.8; 34.2.
MS (CI): 153 [M + H]+. HRCIMS: calcd for C9H13O2 153.0916,
found 153.0913.
5,5-Diallyl-3-methoxy-2-cyclopenten-1-one (15b). A 2.5 M solu-
tion of n-BuLi in hexanes (7.4 mL, 18.6 mmol) was added dropwise
to a cold (-40 °C) solution of dry diisopropylamine (2.6 mL, 18.6
mmol) in anhydrous THF (23 mL), with good stirring under Ar.
The mixture was allowed to warm slowly to -10 °C, then it was
5,5-Bis(carbomethoxy)-3-methoxy-2-cyclo-penten-1-one (15e). A
solution of 13 (506 mg, 4.5 mmol) in dry THF (2 mL) was added
dropwise to a cold (-78 °C) solution of KHMDS (commercial 0.5
M solution in toluene, 10.7 mL, 5.4 mmol) in dry THF (7 mL).
(58) Unless otherwise indicated, NMR spectra were recorded from CDCl3
solutions and IR spectra were obtained from thin films deposited on NaCl plates.
More detailed experimental protocols are provided as Supporting Information.
1594 J. Org. Chem. Vol. 74, No. 4, 2009

Synthetic Studies toward Sordarin
After all of 13 had been added, the mixture was stirred at -78 °C
for 30 min, then dry DMPU (0.6 mL, 4.5 mmol) was injected,
followed by neat ethyl cyanoformate (0.5 mL, 5.4 mmol; CAU-
TION: source of highly toxic HCN). The mixture was stirred for
75 min at -78 °C, then a second portion of KHMDS (0.5 M
solution in toluene, 9.0 mL, 4.5 mmol) was added, and the mixture
was stirred at -78 °C for another 30 min. Dry DMPU (0.6 mL,
4.5 mmol) was injected, followed by neat ethyl cyanoformate (0.5
mL, 5.4 mmol). The solution was stirred at -78 °C for 2 h, then
it was allowed to warm to rt, and it was finally quenched with
water (4 mL). The THF was evaporated, and the residue was
extracted with EtOAc (3 × 20 mL). The combined extracts were
washed with saturated aq NaCl solution, dried (MgSO4), and
evaporated. Chromatographic purification of the residue (30%
EtOAc in cyclohexane) furnished 491 mg (1.9 mmol, 42%) of
product. 1H NMR: 5.28 (s, 1H); 4.30-4.12 (m, 4H); 3.88 (s, 3H);
3.24 (s, 2H); 1.28 (t, J ) 7.2, 6H). 13C NMR: 197.7; 172.3; 171.7;
91.1; 73.5; 59.2; 51.3; 27.4; 13.4. MS (CI): 257 [M + H]+. HRMS:
calcd for C12H16O6 [M + Na]+ 279.0845, found 279.0843.
1.6, 1H); 2.60 (dd, J ) 14.4, J ) 3.7, 1H); 2.40 (dd, J ) 14.0, J
) 7.5, 1H); 2.32 (dd, J ) 15.1, J ) 9.1, 1H); 2.06 (d, J ) 3.7,
1H); 1.65 (d, J ) 14.3, 1H); 1.3-1.1 (m, 21H). 13C (C6D6): 163.8;
135.5; 135.1; 119.9; 117.9; 117.8; 98.8; 98.2; 65.9; 64.5; 56.3; 46.4;
44.4; 36.7; 36.4; 18.8; 14.2. IR: 2323. MS (CI): 492 [M + C4H9]+;
436 [M + H]+; 400 [M - Cl]+. HRCIMS calcd for C24H39ClNO2Si
436.2439, found 436.2436.
7,7-Bis(allyl)-2-exochloro-2-endocyano-1-triisopropylsilyloxy-
bicyclo[2.2.1]heptan-5-one (27). A solution of 21 (708 mg, 1.6
mmol) in THF (12 mL) containing 4 N aq HCl (5.8 mL) was stirred
at 45 °C for 105 min, then it was cooled to rt and diluted with
diethyl ether (30 mL). The organic phase was separated and
sequentially washed with saturated aq NaHCO3 and saturated aq
NaCl solutions, then it was dried (Na2SO4) and evaporated to
provide 658 mg (96%) of colorless crystalline 27, mp 58-59 °C.
1H NMR (C6D6): 6.17 (m, 1H); 5.52 (m, 1H); 5.08-4.99 (m, 4H);
3.03 (m, 1H); 2.65 (m, 1H); 2.56-2.42 (m, 3H); 2.25 (dd, J )
15.5, J ) 9.5, 1H); 1.98 (dd, J ) 14.0, J ) 6.0, 1H); 1.88 (d, J )
5.5, 1H); 1.55 (m, 1H); 1.1-1.0 (m, 21H). 13C NMR (C6D6): 205.4;
134.4; 133.1;119.7; 118.7; 119.5; 90.1; 64.6; 55.7; 54.1; 46.5; 40.4;
36.1; 34.2; 18.6; 14.0. IR: 2359, 1754, 1637. MS (CI): 422 [M +
H]+. HRCIMS: calcd for C23H37ClNO2Si 422.2282, found 422.2281.
Representative Procedure for Radical Cyclization: Compound
29. A solution of AIBN (757 mg, 4.6 mmol) and tris(trimethylsi-
lyl)silane (5.2 mL, 16.9 mmol) in dry degassed toluene (5 mL)
was added dropwise over 45 min to a hot (85 °C) solution of 27
(6.5 g, 15.4 mmol) in dry degassed toluene (30 mL). The resulting
mixture was stirred at 85 °C for an additional 2 h, then it was cooled
and evaporated. Chromatography of the residue (5% EtOAc in
pentane) provided 3.8 g (9.9 mmol, 65%) of colorless crystalline
Representative Procedure for the Preparation of Siloxycyclo-
pentadienes: 5,5-Diallyl-1-triisopropylsilyloxy-3-methoxy-1,3-cy-
clopentadiene (17). Neat TIPSOTf (4.4 mL, 16.3 mmol) was added
to a cold (0 °C) solution of enone 15b (2.4 g, 12.6 mmol) and
DIPEA (3.5 mL, 20.1 mmol) in dry CH2Cl2 (50 mL), with good
stirring under Ar. The cold bath was removed, and the solution
was stirred for 3 h at rt, then it was diluted with a 3:1 mixture of
pentane and CH2Cl2. The colorless precipitate that appeared was
filtered off, and the filtrate was sequentially washed with saturated
aq NaHCO3 and with saturated aq NaCl solutions, then dried
(MgSO4) and evaporated. The oily residue consisted (1H NMR) of
a mixture of 17 and starting enone in a 91:9 ratio. The mixture
was used as such for the following step. The desired 17 possessed
1
29, mp 73-74 °C. H NMR (C6D6): 5.35 (m, 1H); 4.93 (m, 2H);
2.91 (m, 1H); 2.75 (d, J ) 18.2, 1H); 2.29 (dd, J ) 18.2, J ) 1.4,
1H); 2.13 (dd, J ) 13.6, J ) 6.8, 1H); 1.96 (d, J ) 5.9, 1H); 1.82
(dd, J ) 13.4, J ) 7.7, 1H); 1.50 (m, 2H); 1.05 (m, 22H); 0.77 (d,
J ) 6.7, 3H); 0.65 (m, 1H). 13C NMR (C6D6): 206.6; 133.7; 120.8;
118.1; 89.4; 56.9; 54.2; 49.7; 41.1; 37.3; 35.2; 34.3; 25.2; 18.6;
13.7; 12.8. IR: 2234, 1759, 1640. MS (CI): 388 [M + H]+.
HRCIMS: calcd for C23H38NO2Si 388.2672, found 388.2676.
Compound 30. A solution of 29 (1.0 g, 2.6 mmol) and TBAF
(2 g, 6.5 mmol) in THF (10 mL) was stirred at rt for 2 h, then it
was partitioned between ether (30 mL) and water (5 mL). The
organic phase was separated, and the aqueous one was extracted
with more ether. The combined organic phases were dried (Na2SO4)
and evaporated. The residue was rapidly filtered over silica gel (30%
EtOAc in pentane) to give 395 mg (66%) of colorless crystalline
1
the following spectral characteristics. H NMR (C6D6): 5.83 (m,
2H); 5.20 (d, J ) 1.9, 1H); 5.15-5.00 (m, 4H); 4.32 (d, J ) 1.9,
1H); 3.37 (s, 3H); 2.40 (m, 4H); 1.15-0.90 (m, 21H). 13C NMR
(C6D6): 167.4; 160.7; 136.1; 116.9; 101.6; 92.3; 55.9; 55.4; 40.6;
18.5; 13.0. MS (CI): 405 [M + C4H9]+; 349 [M + H]+. HRCIMS:
calcd for C21H37O2Si 349.2563, found 349.2562.
5,5-Bis(1-buten-4-yl)-1-triisopropylsilyloxy-3-methoxycyclopenta-
1,3-diene (18). Neat TIPSOTf (1.1 mL, 4.1 mmol) was added to
a cold (0 °C) solution of the enone 15c (687 mg, 3.1 mmol) and
DIPEA (0.9 mL, 5.0 mmol) in dry CH2Cl2 (12.5 mL), with good
stirring under Ar. The cold bath was removed, and the solution
was stirred for 2 h at rt, then it was diluted with a 3:1 mixture
of pentane and CH2Cl2 (30 mL). The colorless precipitate that
appeared was filtered off, and the filtrate was sequentially washed
with saturated aq NaHCO3 and with saturated aq NaCl solutions,
then dried (MgSO4) and evaporated. The oily residue consisted
(1H NMR) of a mixture of 18 and starting enone in a 95:5 ratio.
The mixture was used as such for the following step. The desired
18 possessed the following spectral characteristics. 1H NMR
(C6D6): 5.89 (m, 2H); 5.20 (d, 1H, J ) 2.0); 5.11 (ddapp, J )
17.0, J ) 2.0, 1H); 4.99 (ddapp, J ) 10.2, J ) 2.2, 1H); 4.15 (d,
J ) 2.0, 1H); 3.38 (s, 3H); 2.25-2.00 (m, 4H); 1.75 (m, 4H);
1.15-0.85 (m, 21H). 13C NMR (C6D6): 166.9; 160.1; 140.2;
114.5; 101.7; 92.3; 55.9; 55.8; 36.1; 29.7; 18.5; 13.1. MS (CI):
377 [M + H]+. HRCIMS: calcd for C23H41O2Si 377.2876, found
377.2878.
Representative Procedure for the Bimolecular Diels-Alder
Reaction: 7,7-Bis(allyl)-6-exochloro-6-endocyano-1-triisopropylsi-
lyloxy-3-methoxybicyclo[2.2.1]hept-2-ene (21). A solution of 17 (5.2
mmol) and 2-chloroacrylonitrile (1.3 mL, 15.7 mmol; CAUTION:
cancer suspect agent) in dry toluene (10 mL) containing suspended
solid K2CO3 (100 mg) was stirred at 65 °C for 12 h. The mixture
was filtered over Celite and evaporated. Chromatography of the
residue (5% EtOAc in cyclohexane) furnished 1.2 g (2.8 mmol,
53% over 2 steps) of 21. 1H NMR (C6D6): 6.18 (m, 1H); 5.66 (m,
1H); 5.10-4.96 (m, 4H); 4.64 (d, J ) 1.6, 1H); 3.09 (s, 3H); 2.75
(dd, J ) 14.0, J ) 7.5, 1H); 2.62 (dddapp, J ) 15.1, J ) 9.1, J )
1
30, mp 97-98 °C. H NMR (C6D6): 5.39 (m, 1H); 4.89 (m, 2H);
2.93 (s, 1H); 2.78 (m, 1H); 2.64 (d, J ) 18.0, 1H); 2.06 (dd, J )
18.0, J ) 1.0, 1H); 2.01 (dd, J ) 14.0, J ) 7.5, 1H); 1.91 (d, J )
6.7, 1H); 1.68 (dd, J ) 14.0, J ) 7.5, 1H); 1.46 (m, 2H); 1.07 (dd,
J ) 14.0, J ) 2.0, 1H); 0.72 (d, J ) 7.0, 3H); 0.55 (dd, J ) 13.0,
J ) 8.0, 1H). 13C NMR (C6D6): 207.9; 133.9; 120.8; 118.0; 87.4;
55.5; 55.2; 49.1; 40.9; 37.3; 34.0; 33.8; 24.9; 12.7. IR: 3458, 2239,
1749, 1638. MS (CI): 288 [M + C4H9]+; 232 [M + H]+; 214 [(M
- H2O) + H]+. HRCIMS: calcd for C14H18NO2 232.1338, found
232.1331.
2-Isopropyl-3-methoxycyclopent-2-en-1-one (74). Neat Me2SO4
(4.5 g, 36.0 mmol; CAUTION: toxic, corrosiVe, cancer suspect
agent) was added at rt to a vigorously stirred suspension of 73
(5.0 g, 36.0 mmol) and K2CO3 (4.5 g, 36.0 mmol) in acetone (60
mL). The mixture was then heated to 60 °C for 8 h with continued
stirring, then it was cooled and concentrated. The residue was
partitioned between EtOAc (10 mL) and aq 1 M NaOH (15 mL),
the layers were separated, and the aqueous phase was further
extracted with EtOAc (3 × 30 mL). The combined extracts were
washed with brine (30 mL), dried (MgSO4), and concentrated.
Chromatography of the residue (25% EtOAc/hexanes) gave 5.5 g
1
(35.0 mmol, 99%) of 74 as an oil. H NMR: 3.91 (s, 3H); 2.73
(sept, J ) 7.0, 1H); 2.62 (m, 2H); 2.39 (m, 2H); 1.11 (d, J ) 7.0,
6H). 13C NMR: 204.8; 184.6; 125.8; 56.5; 33.8; 24.4; 23.1; 20.4.
J. Org. Chem. Vol. 74, No. 4, 2009 1595

Schule´ et al.
IR: 1682, 1621. HRCIMS: calcd for C9H15O2 [M + H]+ 155.1072,
found 155.1078.
2324, 1745, 1645. HRCIMS: calcd for C35H64NO5Si2 [M + H]+
634.4323, found 634.4322.
Ketone 83. A solution of 77 (800 mg, 2.5 mmol), imidazole (253
mg, 3.7 mmol), DMAP (30 mg, 245 µmol), and TBSCl (747 mg,
5 mmol) in dry DMF (2 mL) was stirred at 60 °C for 11 h, then it
was cooled to rt, diluted with saturated aq NH4Cl (20 mL), and
extracted with EtOAc (3 × 10 mL). The combined extracts were
washed with brine (15 mL), dried (MgSO4), and concentrated.
Chromatography of the residue (gradient 0 to 25% EtOAc/hexanes)
gave 758 mg (70%) of a nearly 60:40 mixture of diastereomers of
83 as a light yellow oil. 1H NMR: 5.98-5.89 (m, 2H); 4.30-4.20
(m, 1H); 3.98-3.93 (m, 3H); 3.72 (s, 3H); 3.01-2.91 (m, 1H);
2.66 (dd, J ) 17.5, 4.2, 1H); 2.26 (dd, J ) 17.5, 6.6, 1H); 1.95-1.82
(m, 2H); 1.56-1.43 (m, 2H); 1.22 (dd, J ) 6.9, 4.3, 6H); 0.90 (s,
9H); 0.06-0.04 (m, 6H). 13C NMR: 202.5; 179.8, 179.7; 173.5;
129.6; 126.6; 126.5; 126.4; 116.8; 99.5; 72.0; 71.8; 59.2; 59.1; 52.6;
52.0; 45.0; 44.9; 30.5; 30.3; 28.0; 27.6; 25.5; 24.1; 24.0; 20.3; 19.97;
19.95; 17.95; -4.96; -4.98; -5.16; -5.18. IR: 2223, 1736, 1619.
HRESIMS: calcd for C23H38NO5Si [M + H]+ 436.2516, found
436.2519.
4-Carbomethoxy-2-isopropyl-3-methoxycyclopent-2-en-1-one (75).
A solution of 74 (306 mg, 2 mmol) in THF (2 mL) was added
dropwise to a cold (-78 °C) solution of LHMDS (2.1 mL of
commercial 1 M THF solution, 2.1 mmol, diluted in additional 2
mL of THF), and the mixture was stirred at -78 °C for 2 h. Neat
MeOOCCN (160 µL, 2 mmol; CAUTION: source of toxic HCN)
was injected, and the solution was stirred -78 °C for 30 min. The
reaction was quenched at -78 °C by addition of aq 1 N HCl (4
mL; CAUTION: formation of HCN), allowed to warm to rt, and
extracted with EtOAc (3 × 15 mL). The combined extracts were
washed with brine (10 mL), dried (Na2SO4), and concentrated.
Chromatography of the residue (50:50 EtOAc/hexanes) gave 340
1
mg (81%) of 75 as an oil. H NMR: 3.90 (s, 3H); 3.83 (dd, J )
7.6, 2.3, 1H); 3.78 (s, 3H); 2.79 (dt, J ) 6.7, 1H); 2.72 (d, J ) 7.4,
1H); 2.47 (dd, J ) 17.8, 2.3, 1H); 1.14 (d, J ) 6.8, 6H). 13C NMR:
202.2; 179.5; 172.1; 128.1; 57.4; 53.2; 42.7; 39.3; 23.4; 20.2. IR:
1736, 1702, 1620. HRCIMS: calcd for C9H17O4 [M + H]+
213.1127, found 213.1131.
Cyanoenone 84. Commercial Et2AlCN solution (1.0 M in toluene,
3.3 mL, 3.3 mmol; CAUTION: source of toxic HCN) was added
at rt to a stirred solution of 83 (729 mg, 1.7 mmol, 60:40 mixture
of diastereomers) in dry benzene (15 mL). The mixture was stirred
at rt for 1.5 h, then it was quenched with 1 M NaOH (4 mL;
CAUTION: exothermic reaction, formation of cyanide) and
concentrated. The residue was taken up with aq saturated NaHCO3
(20 mL) and with EtOAc (20 mL). The organic phase was
recovered, and the aqueous phase was further extracted with EtOAc
(3 × 10 mL). The combined extracts were washed with brine (15
mL), dried (MgSO4), and concentrated. The residue was chromato-
graphed (25:75 EtOAc/hexanes) to give 508 mg (70%) of a nearly
60:40 mixture of diastereomers of 84 as a light yellow oil. 1H NMR:
6.05-5.95 (m, 2H); 4.35-4.25 (m, 1H); 3.85-3.77 (m, 3H); 3.07
(dd, J ) 19.0, 3.0, 1H); 3.01-2.95 (m, 1H); 2.37 (dd, J ) 19.0,
3.0, 1H); 2.25-2.15 (m, 1H); 1.85-1.45 (m, 3H); 1.30-1.23 (m,
6H); 0.95-0.85 (m, 9H); 0.12-0.05 (m, 6H). 13C NMR: 203.1;
171.6; 171.5; 161.8; 161.7; 136.0; 135.9; 130.5; 126.3; 126.1; 117.0;
116.9; 113.8; 113.7; 71.9; 71.7; 53.7; 53.6; 43.8; 43.6; 31.5; 31.3;
31.0; 30.6; 26.7; 25.9; 25.8; 20.2; 20.1; -4.60; -4.8. IR: 2225,
1726. HRESIMS: calcd for C23H34N2O4Si [M + Na]+ 453.2183,
found 453.2186.
Aldehyde 76. Neat DBU (13 µL, 87 µmol) was added to a cold
(0 °C) solution of 75 (920 mg, 4.3 mmol) and acrolein (308 µL,
4.6 mmol; CAUTION: toxic, cancer suspect agent) in MeCN (9
mL). The mixture was stirred at 0 °C for 30 min, then at rt for 4 h,
and finally it was concentrated. The residue was treated with aq
saturated NH4Cl (20 mL) and extracted with EtOAc (4 × 10 mL).
The combined extracts were washed with brine (15 mL), dried
(MgSO4), and concentrated to afford crude 76 (1.1 g, 95%), which
was used without further purification.1H NMR: 9.76 (s, 1H); 3.99
(s, 3H); 3.74 (s, 3H); 2.99 (sept, J ) 6.8, 1H); 2.69 (d, J ) 17.8,
1H); 2.48 (m, 1H); 2.32-2.28 (m, 3H); 2.23 (d, J ) 17.7, 1H);
1.23 (dd, J ) 7.1, 4.3, 6H). 13C NMR: 202.8; 200.9; 180.2; 173.6;
126.6; 59.7; 53.1; 52.2; 45.3; 39.0; 26.1; 24.5; 20.7; 20.4. IR: 1730,
1692, 1611. HRCIMS: calcd for C14H21O5 [M + H]+ 269.1389,
found 269.1390.
Nitrile 77. A solution of 76 (65 mg, 242 µmol), DABCO (27
mg, 242 µmol), triethanolamine (16 µL, 121 µmol), and La(OTf)3
(7 mg, 12 µmol) in acrylonitrile (0.5 mL; CAUTION: cancer
suspect agent) was stirred at rt for 11 h, then it was concentrated.
The residue was taken up with aq saturated NaHCO3 (20 mL) and
extracted with EtOAc (3 × 10 mL). The combined extracts were
washed with brine (15 mL), dried (MgSO4), and concentrated.
Chromatography of the residue (50:50 EtOAc/hexanes) gave a ca.
Silyl Enol Ether 94. Commercial LHMDS solution (1.0 M in
THF, 0.3 mL, 0.3 mmol) was slowly added to a cold (-78 °C),
stirred solution of 84 (133 mg, 0.3 mmol), LiCl (347 mg, 8.7 mmol),
and TBSCl (924 mg, 6.2 mmol) in dry THF (2 mL) and HMPA
(0.7 mL; CAUTION: cancer suspect agent). The mixture was
stirred at -78 °C for 30 min, then it was heated to 50 °C. An
additional five portions of LHMDS (1.0 M in THF, each portion
0.3 mL, 0.3 mmol) were added at regular intervals over 8 h. Finally,
the mixture was cooled, carefully quenched with aq saturated NH4Cl
(2 mL), and concentrated. The residue was partitioned between
EtOAc (20 mL) and aq saturated CuSO4 (20 mL), the organic phase
was recovered, and the aqueous phase was extracted with EtOAc
(3 × 10 mL). The combined extracts were washed with brine (15
mL), dried (MgSO4), and concentrated. The residue was chromato-
graphed (1:9 EtOAc/hexanes) to give 138 mg (82%) of a ca. 50:50
mixture of diastereomers of diene 94 as a light yellow oil. 1H NMR:
6.00-5.95 (m, 2H); 5.36-5.30 (m, 1H); 4.30-4.20 (m, 1H);
3.74-3.70 (m, 3H); 3.09-2.85 (m, 1H); 2.26-2.08 (m, 1H);
2.02-1.80 (m, 2H); 1.72-1.47 (m, 1H); 1.34-1.25 (m, 6H);
1.03-0.84 (m, 18H); 0.37-0.02 (m, 12H). 13C NMR: 172.3; 172.2;
171.0; 170.8; 157.4; 156.5; 131.3; 128.0; 127.9; 118.3; 116.7; 116.6;
114.6; 114.5; 113.4; 73.8; 73.5; 54.6; 54.3; 32.2; 31.1; 29.6; 29.3;
27.1; 27.0; 21.9; 21.8; 21.6; 21.5; 21.4; 21.3; 19.5; -3.3; -3.4;
-3.5; -3.7. IR: 2209, 1736. HRESIMS: calcd for C29H48N2O4Si2
[M + Na]+ 567.3054, found 567.3050.
1
60:40 mixture of diastereomers of 77 (39 mg, 50%) as an oil. H
NMR: 6.04 (d, J ) 10.2, 2H); 4.29 (m, 1H); 4.01 (d, J ) 3.0, 3H);
3.71 (s, 3H); 3.01 (m, 1H); 2.69 (dd, J ) 17.8, 3.5, 1H); 2.30 (d,
J ) 17.7, 1H); 2.05-1.98 (m, 2H); 1.75-1.50 (m, 2H); 1.24 (m,
6H). 13C NMR: 203.7; 181.1; 180.8; 174.1; 174.0; 130.7; 126.9;
126.8; 126.7; 117.4; 117.3; 72.0; 71.5; 59.9; 59.8; 52.7; 45.4; 45.3;
30.3; 30.1; 29.3; 28.8; 24.6; 20.9; 20.8. IR: 3367, 2223, 1732, 1688,
1606. HRCIMS: calcd for C17H24NO5 [M + H]+322.1654, found
322.1654.
Compound 80. Neat TIPSOTf (190 µL, 710 µmol) was added
to a cold (0 °C) CH2Cl2 (1.3 mL) solution of 77 (104 mg, 320
µmol; 50:50 mixture of diastereomers) and Hu¨nig base (140 µL,
810 µmol). The solution was warmed to rt and stirred for 12 h,
then it was diluted with a 3:1 mixture of pentane/CH2Cl2 (10 mL),
resulting in precipitation of a colorless solid which was filtered
off. The filtrate was washed with aq saturated NaHCO3 (3 mL),
dried (Na2SO4), and concentrated. Chromatography of the residue
(5:95 EtOAc/hexanes) gave 123 mg (61%) of a mixture of
1
diastereomers of 80 as a colorless oil. H NMR: 4.57 (ddapp, J )
8.6, 6.1, 1H); 4.41 (t, J ) 8.3, 1H); 3.91 (s, 3H); 3.77 (s, 3H); 3.64
(s, 3H); 3.59 (s, 3H); 2.75-2.65 (m, 2H); 2.58 (m, 2H); 2.49-2.42
(m, 2H); 2.16 (d, J ) 12.9, 2H); 2.05-1.80 (m, 6H); 1.75-1.40
(m, 2H); 1.34 (d, J ) 6.7, 12H); 1.23-1.00 (m, 21H). 13C NMR:
173.8; 156.6; 154.4; 124.2; 123.9; 123.8; 122.9; 94.8; 94.2; 75.9;
69.8; 66.6; 66.1; 56.8; 56.5; 52.1; 51.9; 49.9; 49.2; 48.8; 42.5; 33.4;
27.8; 27.7; 26.5; 25.5; 25.1; 22.7; 21.3; 20.6; 18.3; 13.5; 12.9. IR:
Cycloadduct 95. A solution of 94 (80 mg, 147 µmol) in dry
toluene (1 mL) was stirred at 140 °C in a 20 mL glass pressure
1596 J. Org. Chem. Vol. 74, No. 4, 2009

Synthetic Studies toward Sordarin
vessel for 12 h, then it was cooled to rt and applied to a silica
silica gel column (10 g). Elution (gradient 0 to 15% EtOAc/hexanes)
gave 62 mg (77.5%) of a nearly 50:50 mixture of diastereomers of
95 as a light yellow oil. 1H NMR: 4.43-4.32 (m, 1H); 3.76-3.66
(m, 3H); 2.85-2.78 (m, 1H); 2.65-2.49 (m, 2H); 2.26-1.89 (m,
3H); 1.81 (dd, J ) 12.0, 4.0, 1H); 1.62 (d, J ) 12.0, 1H); 1.24 (d,
J ) 8, 3H); 1.15 (d, J ) 8, 3H); 0.98-0.86 (m, 18 H); 0.21-0.09
(m, 12H).13C NMR: 172.1; 171.7; 158.1; 157.9;120.7; 120.9; 119.8;
120.5; 116.4; 116.7; 72.9; 70.1; 67.1; 66.8; 52.1; 52.0; 51.3; 50.4;
50.0; 38.2; 32.9; 29.5; 26.7; 26.4; 25.6; 25.3; 22.0; 22.5; 20.4; 20.3;
17.7; 17.8; -3.7; -3.6; -4.2; -4.5; -4.96; -4.94. IR: 2210, 1739.
HRESIMS: calcd for C29H48N2O4Si2 [M + Na]+ 567.3053, found
567.3050.
Tricyclic Ketone 97. Commercial HF-pyridine complex (70%
HF, 0.9 mL) was added to a cold (0 °C), stirred solution of 95 (91
mg, 167 µmol) in MeCN (3 mL). The mixture was stirred for 6 h,
during which time it was allowed to warm to rt, and then quenched
with aq saturated NaHCO3 (2 mL) and concentrated. The residue
was partitioned between EtOAc (15 mL) and more aq saturated
NaHCO3 (20 mL). The organic phase was recovered, and the
aqueous phase was extracted with more EtOAc (3 × 10 mL). The
combined extracts were washed with brine (15 mL), dried (MgSO4),
and concentrated. The crude product was oxidized without further
purification. Thus, a solution of above alcohol and Dess-Martin
periodinane (73 mg, 172 µmol) in CH2Cl2 was stirred at room
temperature for 3 h, then it was concentrated. Chromatography of
the residue (gradient 5% to 10% EtOAc/hexanes) gave 97 (36 mg,
50% over two steps) as a colorless oil. 1H NMR: 3.78 (s, 3H); 3.05
(d, J ) 3.9, 1H); 2.90-2.68 (m, 2H); 2.62 (sept, J ) 7.0, 1H);
2.47 (dd, J ) 13.5, 3.9, 1H); 2.29-2.18 (m, 1H); 2.10 (d, J ) 3.5,
1H); 2.04-1.90 (m, 1H); 1.30 (d, J ) 6.8, 3H); 1.17 (d, J ) 6.8,
3H); 0.96 (s, 9H); 0.26 (s, 3H); 0.23 (s, 3H). 13C NMR: 198.8;
170.9; 156.7; 120.7; 116.3; 115.8; 66.5; 59.6; 57.1; 52.9; 52.7; 36.6;
31.2; 27.2; 25.6; 23.6; 22.1; 20.8; -3.3; -3.9. IR: 2359, 1734,
1646. HRESIMS: calcd for C23H33N2O4Si [M + H]+ 429.2206,
found 429.2210.
Acknowledgment. We thank Bayer CropScience and the
CNRS (BDI fellowship to A.S.), the University of British
Columbia, the Canada Research Chair Program, NSERC, CIHR,
and MerckFrosst Canada for financial support. H.L. is the
recipient of a UBC Graduate Fellowship.
Supporting Information Available: Experimental proce-
dures and characterization data; detailed results of calculations;
NMR spectra of several compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
JO801911S
J. Org. Chem. Vol. 74, No. 4, 2009 1597