Intramolecular [2+2] photocycloadditions as an approach towards the bicyclo[2.1.1]hexane substructure of solanoeclepin A

Article abstract of DOI:10.1039/b003755i

The synthesis of a tricyclic substructure of solanoeclepin A is described. The key step involves an intramolecular [2+2] photocycloaddition between a dioxinone and a tetrasubstituted bicyclic alkene providing the strained bicyclo-[2.1.1]hexane moiety.

Full text of DOI:10.1039/b003755i

Intramolecular [2+2] photocycloadditions as an approach towards the
bicyclo[2.1.1]hexane substructure of solanoeclepin A
Richard H. Blaauw, Jean-François Brière, Remco de Jong, Jorg C. J. Benningshof, Angeline E. van Ginkel,
Floris P. J. T. Rutjes, Jan Fraanje,† Kees Goubitz,† Henk Schenk† and Henk Hiemstra*
Institute of Molecular Chemistry, Laboratories of Organic Chemistry and Crystallography, University of Amsterdam,
Nieuwe Achtergracht 129, 1018 WS Amsterdam. E-mail: henkh@org.chem.uva.nl
Received (in Liverpool, UK) 9th May 2000, Accepted 20th June 2000
Published on the Web 14th July 2000
The synthesis of a tricyclic substructure of solanoeclepin A is
described. The key step involves an intramolecular [2+2]
photocycloaddition between a dioxinone and a tetrasub-
stituted bicyclic alkene providing the strained bicyclo-
[2.1.1]hexane moiety.
known to be readily prepared and to show reliable photo-
chemical behaviour4 and it should eventually provide useful
functionality for our total synthesis endeavour. Aldehyde 2 was
prepared from commercially available tert-butyl acetoacetate
via (1) alkylation with allyl bromide, (2) dioxinone formation5
and (3) oxidative cleavage of the allyl group.6 Subsequent
alkenylation with vinylmagnesium bromide, followed by MOM
protection of the allylic alcohol, afforded cyclisation precursor
3.
Solanoeclepin A (1) is the most active natural hatching agent of
the potato cyst nematode.1 Its heptacyclic structure contains all
ring sizes ranging from three to seven, including a strained
Upon irradiation smooth cyclisation occurred to afford the
expected bicyclo[2.1.1]hexane 4 as a 1+1 mixture of diastereo-
isomers. This product is in accordance with the so-called ‘rule
of five’.7 A close analogue has been earlier prepared by Kaneko
and co-workers.4b Cycloadduct 4 was found to be unstable,
decomposing slowly under the reaction conditions and during
the subsequent work-up. However, exhaustive reduction with
lithium aluminium hydride led to stable diol 5.
Encouraged by this result, we set out to construct a more
appropriately functionalised cyclisation precursor, bearing a
cyclohexenyl side chain, necessary for the construction of 1.
Chromium-mediated coupling of aldehyde 2 and vinyl triflate
68 (Scheme 2), followed by oxidation of the allylic alcohol and
acetalisation,9 afforded cyclisation precursor 7. This oxidised
and protected precursor was chosen to prevent diastereomeric
mixtures after the cycloaddition.
Much to our surprise, cyclisation of 7 resulted in the
exclusive formation of the strained bicyclo[2.2.0]hexane 8. This
cycloadduct exhibited enhanced stability compared to 4, even to
silica gel column chromatography, allowing the complete
characterisation of this molecule. Ultimate proof of the structure
of 8 was obtained by performing a De Mayo fragmentation,
which after esterification with diazomethane gave spiro[3.4]oc-
tane 9.10 The structure of one of the isomers of 9 was
unambiguously secured by X-ray crystallography.
bicyclo[2.1.1]hexanone unit, which to the best of our knowl-
edge is an unprecedented structural feature in natural products.
The structure of 1 to a certain extent resembles that of
glycinoeclepin A,2 the hatching agent of the soybean cyst
nematode. The extreme scarcity of natural material, its
fascinating structure, and its potential role in the search for a
benign way to control potato sickness make 1 a challenging
target for total synthesis. This and the following communica-
tion3 document our first strides towards this goal.
We have investigated intramolecular [2+2] photocycloaddi-
tions4 between a 1,3-dioxin-4-one and variously substituted
alkenes connected at C5 with a two carbon tether, to arrive at
highly substituted bicyclo[2.1.1]hexanes. We wish to report
herein (1) our preliminary results on these cycloadditions,
which exhibit remarkably variable regioselectivities, and (2) our
achievements towards the stereoselective construction of a
tricyclic substructure of solanoeclepin A, containing the
bicyclo[2.1.1]hexane moiety.
We hypothesised that this remarkable regiochemical prefer-
ence could be attributed to stereoelectronic effects. In an
To investigate the viability of a photochemical approach a
simple model system was selected containing the 6-methyl-
1,3-dioxin-4-one moiety (see Scheme 1). This structure is
Scheme 2 Reagents: a, CrCl2, NiCl2 (cat.), DMF, rt; b, (COCl2)2, DMSO,
Et3N, CH2Cl2, 278 °C ? rt; c, (TMSOCH2)2, TMSOTf, CH2Cl2, 0 °C; d,
hn (300 nm), MeCN/acetone (9:1 v/v), rt; e, KOH, dioxane–H2O, rt; f,
CH2N2, MeOH, 0 °C.
Scheme 1 Reagents: a, vinylmagnesium bromide, THF, 278 °C; b,
MOMCl, i-PrNEt2, CH2Cl2, rt; c, hn (300 nm), MeCN–acetone (9:1 v/v), rt;
d, LiAlH4, THF, rt.
DOI: 10.1039/b003755i
Chem. Commun., 2000, 1463–1464
This journal is © The Royal Society of Chemistry 2000
1463

Scheme 4 Reagents: a, LiAlH4, THF, rt; b, TBDMSOTf, 2,6-lutidine,
CH2Cl2, 0 °C; c, KHMDS, phenyl chloroformate, THF, 278 °C; d, CSA,
MeOH, 0 °C; e, NaH, THF, 0 °C.
failure. Therefore, the primary and secondary hydroxy groups
were protected as their TBDMS ethers, followed by functional-
isation of the tertiary alcohol with a phenyl carbonate group,
affording 19. Selective hydrolysis of the primary TBDMS
ethers, yielded diol 20. Upon treatment of this diol with sodium
hydride, cyclic carbonate 21 was formed, leaving one primary
hydroxy group unprotected. This compound contains appro-
priate substitution and stereochemistry for elaboration towards
the right-hand substructure of solanoeclepin A.
These investigations are supported (in part) by the Nether-
lands Research Council for Chemical Sciences (CW) with
financial aid from the Netherlands Technology Foundation
(STW).
Scheme 3 Reagents: a, CrCl2, NiCl2 (cat.), DMF, 50 °C; b, hn (300 nm),
MeCN–acetone (9+1 v/v), rt; c, MOMCl, i-PrNEt2, CH2Cl2, rt.
attempt to direct the cycloaddition to the desired mode of
closure, aldehyde 2 was reacted with triflate 10 bearing an
additional electron-withdrawing ester substituent on the alkene
(Scheme 3). Not unexpectedly,11 the product was not the
hydroxy ester 11, but lactone 12. Gratifyingly, subjection of this
latter precursor to the irradiation conditions smoothly led to
bicyclo[2.1.1]hexane 13 with complete regio- and diastereo-
selectivity in high yield. The cycloadduct 13 appeared unstable
on a silica gel column, but was readily purified by recrystallisa-
tion (mp 177–178 °C), and its structure was confirmed by X-ray
crystallography‡.
To probe the generality of this cyclisation mode we also
investigated the five-membered ring triflate ester 14 as starting
material. Its coupling with aldehyde 2 gave hydroxy ester 15,
which did not lactonise spontaneously nor could it be forced to
do so by heating. Alcohol 15 was therefore protected as the
MOM ether 16. On irradiation of 16 under the usual conditions
a ca. 1:1 mixture of stable stereoisomeric cycloadducts 17 was
obtained containing again the bicyclo[2.2.0]hexane moiety. The
diastereoisomer with the OMOM group trans with respect to the
cyclopentane ring (mp 62–64 °C) was crystalline and allowed
unambiguous structural proof by X-ray diffraction.
Thus, of the four photocyclisation precursors investigated,
two (3 and 12) cyclise in the expected crossed mode obeying the
rule of five, while the other two (7 and 16) cyclise in the
unexpected straight mode. In view of the precedent available, 3
shows normal cyclisation behaviour. However, very little is
known about tri- or tetrasubstituted alkenes in photocycloaddi-
tions with 2-carbon tethered dioxinones. It is tempting to
speculate that the first C–C-bond formation by radical cyclisa-
tion to a 5- or 6-membered ring is reversible,12 depending on the
feasibility of the second, irreversible C–C-bond formation.
Preliminary molecular modeling studies indicate that in the case
of ester 16 initial 5-membered ring formation cannot be readily
followed by a second C–C-coupling due to conformational
constraints. At any rate, photochemistry once again proves to be
a very powerful synthesis technique, producing in one step four
contiguous quaternary carbon centres, exemplified by the
formation of 13 and 17, of which the former has the desired
skeleton for our total synthesis purposes.
Notes and references
† Laboratory of Crystallography.
‡ CCDC 182/1702. See http://www.rsc.org/suppdata/cc/b0/003755i/ for
crystallographic files in .cif format.
1 J. G. Mulder, P. Diepenhorst, P. Plieger, I. E. M. Brüggemann-Rotgans,
PCT Int. Appl. WO 93/02,083; Chem. Abstr., 1993, 118, 185844z; H.
Schenk, R. A. J. Driessen, R. de Gelder, K. Goubitz, H. Nieboer, I. E. M.
Brüggemann-Rotgans, P. Diepenhorst, Croat. Chem. Acta, 1999, 72,
593.
2 Structure elucidation: A. Fukuzawa, A. Furusaki, I. Mitsuhiko and T.
Masamune, J. Chem. Soc., Chem. Commun., 1985, 222. Total synthesis:
A. Murai, N. Tanimoto, N. Sakamoto and T. Masamune, J. Am. Chem.
Soc., 1988, 110, 1985; H. Watanabe and K. Mori, J. Chem Soc., Perkin
Trans. 1, 1991, 2919; E. J. Corey and I. N. Houpis, J. Am. Chem. Soc.,
1990, 112, 8997; E. J. Corey and B. Hong, J. Am. Chem. Soc., 1994, 116,
3149.
3 J. C. J. Benningshof, R. H. Blaauw, A. E. van Ginkel, F. P. J. T. Rutjes,
J. Fraanje, K. Goubitz, H. Schenk and H. Hiemstra, Chem. Commun.,
2000, 1465.
4 (a) J. D. Winkler, C. Mazur Bowen and F. Liotta, Chem. Rev., 1995, 95,
2003; (b) M. Sato, Y. Abe, K. Takayama, K. Sekiguchi and C. Kaneko,
J. Heterocycl. Chem., 1991, 28, 241; (c) M. T. Crimmins and T. L.
Reinhold, Org. React., 1993, 44, 297.
5 M. Sato, H. Ogasawara, K. Oi and T. Kato, Chem. Pharm. Bull., 1983,
31, 1896; S. Huckin and L. Weiler, J. Am. Chem. Soc., 1974, 96,
1082.
6 E. G. Baggiolini, J. A. Iacobelli, B. M. Hennessy, A. D. Batcho, J. F.
Sereno and M. R. Uskokovic, J. Org. Chem., 1986, 51, 3098.
7 R. Srinivasan and K. H. Carlough, J. Am. Chem. Soc., 1967, 89, 4932;
R. S. H. Liu and G. S. Hammond, J. Am. Chem. Soc., 1967, 89, 4936.
8 K. Takai, M. Tagashira, T. Kuroda, K. Oshima, K. Utimoto and H.
Nozaki, J. Am. Chem. Soc., 1986, 108, 6048.
9 T. Tsunoda, M. Suzuki and R. Noyori, Tetrahedron Lett., 1980, 21,
1357.
10 For a review, see: M. Sannigrahi, Tetrahedron, 1999, 55, 9007; M.
Phmakotr, T. Bunlaksananusorn and P. Tuchinda, Tetrahedron Lett.,
2000, 41, 377.
To examine the utility of 13 in model studies towards the
natural product, it was reduced with excess lithium aluminium
hydride to yield the stable tetrahydroxy compound 18 (Scheme
4). Further elaboration of 18 required a differentiation of the two
primary hydroxy groups. Unfortunately, all attempts to se-
lectively mono-protect one of the primary alcohols met with
11 P. Knochel and C. J. Rao, Tetrahedron, 1993, 49, 29.
12 D. Andrew and A. C. Weedon, J. Am. Chem. Soc., 1995, 117, 5647.
1464
Chem. Commun., 2000, 1463–1464