Synthesis of the cyclobutanone core of solanoeclepin A intramolecular allene butenolide photocycloaddition.

Article abstract of DOI:10.1039/b311415e

The compact tricyclic substructure of solanoeclepin A containing the cyclobutanone ring was prepared by using as the key step a highly regioselective intramolecular [2 + 2]-photocycloaddition reaction between one of the [small pi]-bonds of an allene and t

Full text of DOI:10.1039/b311415e

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis of the cyclobutanone core of solanoeclepin A
via intramolecular allene butenolide photocycloaddition†
Bui T. B. Hue, Jan Dijkink, Sanne Kuiper, Kimberly K. Larson,‡ Frank S. Guziec, Jr.,‡
Kees Goubitz, Jan Fraanje, Henk Schenk, Jan H. van Maarseveen and Henk Hiemstra*
Institute of Molecular Chemistry, University of Amsterdam, Nieuwe Achtergracht 129,
1018 WS Amsterdam, The Netherlands. E-mail: hiemstra@science.uva.nl
Received 24th September 2003, Accepted 23rd October 2003
First published as an Advance Article on the web 29th October 2003
The compact tricyclic substructure of solanoeclepin A con-
taining the cyclobutanone ring was prepared by using as the
key step a highly regioselective intramolecular [2 ؉ 2]-
photocycloaddition reaction between one of the ꢀ-bonds
of an allene and the CC double bond of a butenolide.
Solanoeclepin A (1) is a potent natural hatching agent of
potato cyst nematodes (PCN) showing activity at nanomolar
concentration.1 Its structure was finally elucidated in the early
nineties by X-ray analysis after a long period of intense studies
through the joint efforts of a number of Dutch research organ-
izations.2 The resemblance of the structure of 1 and that of
glycinoeclepin A (2), the hatching agent of the soybean cyst
nematode is remarkable (Fig. 1).3 The great interest for the
structure of the PCN hatching agent resulted from the need to
develop an environmentally benign way to combat PCN, which
cause serious losses in potato production. While the complexity
of the natural product has somewhat reduced the agricultural
significance, the chemical interest has considerably increased as
the fascinating structure poses a formidable challenge for total
synthesis. Moreover, the virtual unavailability of the compound
from natural sources renders a chemical synthesis an obvious
method to collect information on the properties and about
structure–hatching activity relationships.
Scheme 1 Retrosynthesis of solanoeclepin A (arbitrary protective
groups indicated as R).
angular methyl and the hydroxyl function in the correct stereo-
chemical arrangement. However, the dioxenone moiety caused
chemoselectivity problems in later phases of the synthesis. In
this communication we show the virtue of 7 as the photo-
chemistry precursor, preserving the advantage of the lactone
function, but with an allene moiety allowing a much more
efficient synthetic approach to solanoeclepin A (1).
We commenced our studies with the synthesis of allene 11 as a
model system (Scheme 2). The commercially available cyclic
anhydride 8 was reduced with NaBH4 in THF7 to give the known
butenolide 9,8 which was then converted into the triisopropylsilyl
dienolate 10 by using TIPSOTf and Hünig’s base.9 Without puri-
fication this enol ether was mixed with 1-bromo-2,3-butadiene10
in dichloromethane and then treated at Ϫ78 ЊC with silver tri-
fluoroacetate according to the procedure reported by Jefford
et al.11 In this way the photochemistry precursor 11 was obtained
in 64% overall yield from furanone 9.
Fig. 1 Natural hatching agents.
Several years ago we embarked on a synthetic approach
toward solanoeclepin A (1) based on the eventual coupling of
the aldehyde 3 and enol triflate 4 and closure of the seven-
membered ring (Scheme 1).4 The synthesis of 3 in enantiopure
form and its coupling with a model triflate was found to pro-
ceed satisfactorily, so that the next goal was the preparation of
the tetracyclic substructure 4. An obvious strategy to access the
cyclobutane containing tricyclic core structure would be the
intramolecular [2 ϩ 2]-photocycloaddition.5 In earlier studies
we discovered that acetone sensitized irradiation of dioxenone 6
produced a cyclobutane of type 5 in very high yield.6 In this
process the lactone connection appeared essential to obtain the
desired crossed regioselectivity in the photocycloaddition.
Reductive opening of the lactone should eventually lead to the
† Electronic supplementary information (ESI) available: experimental
details. See http://www.rsc.org/suppdata/ob/b3/b311415e/
‡ Current address: Department of Chemistry, Southwestern University,
Georgetown, Texas 78627, USA.
Scheme 2 Preparation of the photochemistry precursor and the
[2 ϩ 2]-cycloaddition.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 3 6 4 – 4 3 6 6
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
4364

View Article Online
The key photoreaction was carried out under argon in an air-
cooled quartz vessel containing a 50 mM solution of 11 in a
9 : 1 v/v acetonitrile/acetone mixture. Rayonet RPR-3000 Å
lamps served as the light source in a Rayonet photoreactor. The
reaction was easily followed by TLC which showed complete
conversion of starting material after ca. 5 h. NMR of the crude
material showed the formation of a single product, which was
proximity of the secondary alcohol causing formation of a
cyclic ether.6c Therefore, a protection–deprotection sequence
had to be followed giving benzyl ether 14 in 76% overall yield
from 13. Tosylation of 14 was successful and its reduction with
sodium triethylborohydride15 gave 15 in good yield.
Oxidative cleavage of the double bond in 15 with ozone was
not successful, leading to destruction of the skeleton instead.
Osmium-mediated dihydroxylation was only productive with
stoichiometric amounts of osmium tetroxide in a pyridine/
water mixture at 65 ЊC.16 In this way a single diol 16 was
obtained. From 1H NMR NOE measurements it was clear that
the dihydroxylation had taken place from the more open endo
face of the alkene. Removal of the benzyl protective group by
hydrogenolysis followed by oxidative cleavage of the diol with
sodium periodate furnished the desired cyclobutanone 17 in
acceptable overall yield as a crystalline solid (mp 103–106 ЊC;
IR ν 1798 and 1766 cmϪ1). The structure was proven by X-ray
analysis (see Fig. 2 and ESI). Interestingly, this rather strained
β-hydroxyketone appeared to be quite stable, although its
stability in aqueous medium at different pH was not studied.17
We also investigated the oxidation of the exocyclic methylene
at an earlier stage. Interestingly, the dihydroxylation of the
alkene in lactone 12 occurred much more readily than the
corresponding reaction of benzyl ether 15. Thus, treatment of
12 with catalytic osmium tetroxide and NMO gave a single
diol 18 in 68% yield (Scheme 4). Like in the case of 15 the
dihydroxylation had taken place from the more open endo
face of the double bond. Unfortunately, diol cleavage of 18
appeared to be a difficult process giving only a low yield of the
cyclobutanone 19. Apparently, subtle steric and strain effects
play a crucial role in the chemistry of the compact skeleton of
the carbotricyclic structures.
1
isolated as a crystalline solid in 70% yield. Its H NMR spec-
trum clearly showed that it was the result of cycloaddition of
the internal allene double bond in view of the presence of two
singlets of an exocyclic methylene group at 4.61 and 4.76 ppm.
The structure of 12 was proven by X-ray analysis§ (see Fig. 2
and ESI). The exclusive formation of 12 emphasizes the great
preference for 5-membered ring formation in intramolecular
[2 ϩ 2]-photocycloaddition (the so-called rule of five).12 Remark-
ably, the homologous allene with a penta-3,4-dienyl side chain
was reported by Coates13 to give poor regioselectivity.14 This
difference may be explained by the fact that 5-membered ring
formation in the case of 11 leads to an allylic radical intermedi-
ate, whereas 5-membered ring formation from Coates’ substrate
gives a much less stable vinyl radical, so that other cyclization
modes can compete.
Fig. 2 Spartan representation of the X-ray crystal structures of
lactone 12 (left) and cyclobutanone 17.
The tricyclic carbon skeleton of 12 contains the key struc-
tural features of the core of solanoeclepin A with the exocyclic
methylene group as a potential protective group of a ketone. In
order to investigate whether a cyclobutanone can be indeed
obtained from 12 its lactone was first reduced to the diol 13
(Scheme 3). We planned to generate the angular methyl group
by reductive removal of a sulfonate. However, attempted direct
synthesis of the monotosylate from 13 by using tosyl chloride in
pyridine followed by reduction failed, probably because of the
Scheme 4 Oxidative cleavage of the exocyclic methylene starting from
lactone 12.
The natural product 1 requires the final hydrogen of the
cyclobutanone ring in 17 to be replaced by a cyclopropane-
carboxylic acid function. To probe whether the photocycloaddi-
tion proceeds well for this purpose the more highly substituted
allenes 20a and 20b were used as starting materials. Both a silyl
substituent (a dimethylphenylsilyl group is a potential synthetic
equivalent for a hydroxyl function) and an ester group were
deemed useful functionalities to be transformed into the cyclo-
propane at a later stage. Allene 20a was prepared according to a
literature procedure.18
The ester substituted allene 20b was prepared in three steps
as shown in Scheme 5. By using carefully controlled conditions
the orthoester Claisen rearrangement starting from 2-butyne-
1,4-diol could be influenced to occur only once to a consider-
able extent so that the hydroxyester 21 could be isolated pure in
28% yield.19 From 21 bromide 20b was readily obtained in
excellent yield as a stable compound.
Scheme 3 Preparation of the tricyclic core structure of solanoeclepin
A containing the cyclobutanone ring.
Scheme 5 Expedient three step synthesis of the allene 20b containing
an ester function.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 3 6 4 – 4 3 6 6
4365

View Article Online
CCDC reference numbers 222276 and 222277. See http://
www.rsc.org/suppdata/ob/b3/b311415e/ for crystallographic data in
CIF or other electronic format.
The silver-mediated coupling of the allenic bromides with
silyl dienol ether 10 proceeded in acceptable yields (unopti-
mized) in view of the sensitivity of the allenic starting materials
(Scheme 6). The photocycloadditions cleanly gave single regio-
isomers at a somewhat higher rate than for allene 11. The ease
of the cycloaddition leading to three quaternary centers in a
highly compact setting is noteworthy.
1 J. G. Mulder, P. Diepenhorst, P. Plieger and I. E. M. Brüggemann-
Rotgans, PCT Int. Appl. WO 93/02,083 (Chem. Abstr., 1993, 118,
185844z).
2 H. Schenk, R. A. J. Driessen, R. de Gelder, K. Goubitz, H. Nieboer,
I. E. M. Brüggemann-Rotgans and P. Diepenhorst, Croat. Chem.
Acta, 1999, 72, 593–606.
3 A. Fukuzawa, A. Furusaki, M. Ikura and T. Masamune, J. Chem.
Soc., Chem. Commun., 1985, 222–224; Correction: A. Fukuzawa,
A. Furusaki, M. Ikura and T. Masamune, J. Chem. Soc., Chem.
Commun., 1985, 748.
4 (a) J. C. J. Benningshof, R. H. Blaauw, A. E. van Ginkel, J. H. van
Maarseveen, F. P. J. T. Rutjes and H. Hiemstra, J. Chem. Soc., Perkin
Trans. 1, 2002, 1693–1700; (b) J. C. J. Benningshof, M. IJsselstijn,
S. R. Wallner, A. L. Koster, R. H. Blaauw, A. E. van Ginkel,
J.-F. Brière, J. H. van Maarseveen, F. P. J. T. Rutjes and H. Hiemstra,
J. Chem. Soc., Perkin Trans. 1, 2002, 1701–1713.
5 The most direct approach to the bicyclo[2.1.1]hexanone would be an
intramolecular ketene olefin cycloaddition, but this process was
expected to be unproductive on the basis of literature precedent:
B. B. Snider and R. A. H. F. Hui, J. Org. Chem., 1985, 50, 5167–
5176.
6 (a) R. H. Blaauw, J.-F. Brière, R. de Jong, J. C. J. Benningshof, A. E.
van Ginkel, F. P. J. T. Rutjes, J. Fraanje, K. Goubitz, H. Schenk and
H. Hiemstra, J. Org. Chem., 2001, 66, 233–242; (b) J.-F. Brière, R. H.
Blaauw, J. C. J. Benningshof, A. E. van Ginkel, J. H. van Maarseveen
and H. Hiemstra, Eur. J. Org. Chem., 2001, 2371–2377; (c) R. H.
Blaauw, J. C. J. Benningshof, A. E. van Ginkel, J. H. van Maarseveen
and H. Hiemstra, J. Chem. Soc., Perkin Trans. 1, 2001, 2250–2256.
7 D. M. Bailey and R. E. Johnson, J. Org. Chem., 1970, 35, 3574–3576.
8 D. Butina and F. Sondheimer, Synthesis, 1980, 543–545.
9 (a) S. F. Martin, K. J. Barr, D. W. Smith and S. K. Bur, J. Am. Chem.
Soc., 1999, 121, 6990–6997; (b) J. Boukouvalas and N. Lachance,
Synlett., 1998, 31–32.
Scheme 6 Photocycloaddition with substituted allenes.
In conclusion, the most intricate tricyclic substructure of
solanoeclepin A containing the bicyclo[2.1.1]cyclohexanone
moiety with the correct substitution pattern was prepared by
using the [2 ϩ 2]-photocycloaddition reaction between an
allene and a butenolide as the key step. This major break-
through paves the way towards the total synthesis of the natural
hatching agent which is under active investigation in our
laboratories.
10 J. Pornet, B. Randrianoelina and L. Miginiac, J. Organomet. Chem.,
1979, 174, 1–13.
11 (a) C. W. Jefford, A. W. Sledeski and J. Boukouvalas, Helv. Chim.
Acta, 1989, 72, 1362–1370; (b) C. W. Jefford, A. W. Sledeski, J.-C.
Rossier and J. Boukouvalas, Tetrahedron Lett., 1990, 31, 5741–5744.
12 See e.g. M. T. Crimmins and T. L. Reinhold, Org. React., 1993, 44,
297–588.
Acknowledgements
We thank the NUFFIC for its financial support to B. T. B. H.
through a development cooperation project between the
Universities of Amsterdam and Can Tho (Vietnam). The stay
of K. L. L. in Amsterdam was financially supported by a
Mundy Fellowship and the Dishman Endowment. We acknow-
ledge the Netherlands Science Foundation (NWO) for a
Visitor’s Grant to F. S. G. We also thank Professor Paul Wender
(Stanford University) for a useful discussion.
13 R. M. Coates, P. D. Senter and W. R. Baker, J. Org. Chem., 1982, 47,
3597–3607.
14 (a) For a review on the photochemistry of butenolides, see A. I.
Hashem, A. Senning and A.-S. Hamad, Org. Prep. Proc. Int., 1998,
30, 403–425; (b) For intramolecular δ-lactone photocycloadditions
of enantiopure allenes, see M. S. Shepard and E. M. Carreira, Tetra-
hedron, 1997, 53, 16253–16276.
15 S. Krishnamurthy and H. C. Brown, J. Org. Chem., 1976, 41, 3064–
3066.
16 T. Honda, H. Takada, S. Miki and M. Tsubuki, Tetrahedron Lett.,
1993, 34, 8275–8278.
17 The reported instability of the natural product in basic medium may
be associated with the β-hydroxyketone moiety.
18 (a) P. H. Lee, K. Bang, H. Ahn and K. Lee, Bull. Korean Chem. Soc.,
2001, 22, 1385–1389; (b) B. M. Trost and H. Urabe, J. Am. Chem.
Soc., 1990, 112, 4982–4983.
19 (a) Y. Ishino, I. Nishiguchi, M. Kim and T. Hirahima, Synthesis,
1982, 740–742; (b) A. Srikrishna, S. Nagaraju and P. Kondaiah,
Tetrahedron, 1995, 51, 1809–1816.
Notes and references
§ Crystal data for 12: C12H14O2, M = 190.24, triclinic, a = 6.6969(4),
b = 7.0489(6), c = 10.9328(7) Å, α = 77.964(6), β = 75.679(9),
3
¯
γ = 80.977(8)Њ, V = 486.04(6) Å , T = 250 K, space group P1, Z = 2,
µ(Cu-Kα) = 0.70 mmϪ1, 1806 observed unique reflections.
For 17: C11H16O2,
M = 180.24, monoclinic, a = 7.0166(3),
b = 10.5703(6), c = 13.7222(13) Å, β = 102.880(5)Њ, V = 992.13(12) Å3,
T = 295 K, space group P21/n, Z = 4, µ (Cu-Kα) = 0.65 mmϪ1, 1743
observed unique reflections.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 3 6 4 – 4 3 6 6
4366