Tandem radical rearrangement/Pd-catalysed translocation of bicyclo[2.2.2]lactones. An efficient access to the oxa-triquinane core structure

Article abstract of DOI:10.1039/b518121f

The skeletal rearrangement of bicyclo[2.2.2] lactones, which involves a mild and chemoselective palladium based transformations, were analyzed. The generalized use of palladium complexes as mediators or catalysts for carbon-carbon and carbon-heteroatom bo

Full text of DOI:10.1039/b518121f

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/obc | Organic & Biomolecular Chemistry
Tandem radical rearrangement/Pd-catalysed translocation of
bicyclo[2.2.2]lactones. An efficient access to the oxa-triquinane core structure†
Jyh-Hsiung Liao, Nuno Maulide, Benoˆıt Augustyns and Istva´n E. Marko´*
Received 21st December 2005, Accepted 26th February 2006
First published as an Advance Article on the web 8th March 2006
DOI: 10.1039/b518121f
The skeletal rearrangement of bicyclo[2.2.2]lactones, involv-
ing a mild and chemoselective palladium-catalysed translo-
cation key-step, provides an efficient and diastereoselective
access to synthetically useful bicyclo[3.3.0]lactones.
Whilst seeking to extend the scope of this methodology, we
became aware of a potential limitation in our two-step protocol
(Scheme 2). Indeed, whereas skeletal rearrangement of lactones
1b and 1c took place smoothly under the “standard” conditions
(TTMSS–AIBN in boiling benzene, followed by stirring with
silica gel in dichloromethane), the translocation of bridged
bicyclo[3.2.1]lactone 2d (formed by radical rearrangement of a-
substituted selenide 1d) to its fused counterpart 3d, required
more vigorous conditions (PTSA, refluxing benzene).2a It appears
that the strain induced by the additional bridgehead substituent
considerably raised the energetic barrier for the acid-catalysed
reorganisation of the allylic lactone moiety in 2d.
The advent of transition metal-mediated transformations con-
siderably enhanced the ability of synthetic chemists to assemble
complex molecular structures. In particular, the generalised use of
palladium complexes as mediators or catalysts for carbon–carbon
and carbon–heteroatom bond forming reactions has clearly revo-
lutionised the field of organic chemistry.1 Most importantly, these
palladium-based transformations often occur with high levels of
selectivity and under exceptionally mild conditions.
We have recently reported a novel tandem radical-initiated
Brønsted-acid catalysed skeletal rearrangement of bicyclo[2.2.2]-
lactones 1a (Scheme 1).2 This rearrangement proceeds with com-
plete transfer of relative and absolute stereochemistry and yields,
depending upon the conditions employed, either the bicyclo[3.2.1]-
lactone 2a or its isomer 3a. The synthetic usefulness of this
sequence was demonstrated by the ready conversion of derivative
4 into Corey’s lactone (Scheme 1).3
Scheme 2 Skeletal rearrangement of substituted bicyclo[2.2.2]lactones.
Recognising that the compatibility of these strongly acidic
conditions with potentially sensitive functional groups, present
in more elaborated substrates, would significantly impair the
scope of our methodology, we became interested in the possibility
of effecting this crucial cationic rearrangement under neutral
conditions. The palladium-catalysed allylic alkylation reaction
(also known as the Tsuji–Trost reaction) has received considerable
attention and ranks amongst the most reliable transition metal-
mediated carbon–carbon bond forming reactions.4 Since allylic
lactones react with Pd(0) complexes,5 we envisioned that the
treatment of 2 with a suitable palladium(0) catalyst could lead
to the p-allyl complex 6, which could be driven to cyclise to the
thermodynamically more stable fused lactone 3, under neutral
conditions (Scheme 3).
Scheme 1 Previously reported skeletal rearrangement and application to
a short synthesis of Corey’s lactone.
Our preliminary experiment involved the addition of 10 mol%
palladium(II) acetate and triphenylphosphine to a THF solution
of racemic, bridged bicyclo[3.2.1]lactone 2a (Table 1, Entry 1).
Pleasingly, smooth isomerisation of 2a ensued and fused lactone 3a
Universite´ catholique de Louvain, De´partement de Chimie, Baˆtiment
Lavoisier, Place Louis Pasteur 1, 1348, Louvain-la-Neuve, Belgium
† Electronic supplementary information (ESI) available: Instrumentation
and procedures. See DOI: 10.1039/b518121f
1464 | Org. Biomol. Chem., 2006, 4, 1464–1467
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Table 1 Pd-catalysed translocation of lactones 2a
Entry
Substrate
Product
Catalyst
Yield (%)
90
1
Pd(OAc)2–PPh3
2
Pd(PPh3)4
90
3
4
Pd(PPh3)4
Pd(PPh3)4
92
91
5
Pd(PPh3)4
90
a All reactions performed in THF from 0 ◦C to rt. For the synthesis of lactones 2, see ref. 6.
synthesis of the angular triquinane core structure, in racemic
form. Triquinanes are a well-known family of biologically relevant,
naturally occurring substances, and numerous approaches towards
their synthesis have been described.7 Angular triquinanes typically
possess an intricate tricyclic structure, featuring two to three all-
carbon quaternary centres, as in the case of isocomene (Scheme 4).
Scheme 3 Proposed palladium-catalysed translocation of bicyclo[3.2.1]-
lactones.
was isolated in high yield. In order to rule out a possible mediation
by traces of acetic acid formed in situ, this experiment was
repeated using palladium tetrakis(triphenylphospine). A similar
result was obtained, clearly establishing the viability of the allylic
rearrangement under non-acidic conditions (Table 1, Entry 2).
These conditions were then applied to the rearrangement of
a selection of other bridged bicyclic lactones (Table 1). Thus,
bicyclo[3.2.1]lactones 2b and 2c6 smoothly reacted in the presence
of the palladium catalyst affording the desired, fused lactones
3b and 3c in 92 and 91% yields, respectively. The crucial test
came with the more resilient bicyclo[3.2.1]lactone 2d6 (Table 1,
Entry 5). In the event, this bridged lactone also underwent clean
rearrangement into its fused isomer 3d.
Having established that these neutral conditions were broadly
applicable, we next sought to apply this novel radical-mediated/
palladium-catalysed skeletal rearrangement sequence to a short
Scheme 4 Antithetic analysis of angular oxa-triquinane core structure 7.
Org. Biomol. Chem., 2006, 4, 1464–1467 | 1465
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Our proposed retrosynthetic analysis of the angular oxa-
triquinane core 7 is depicted in Scheme 4. Intramolecular alky-
lation of a-carbomethoxylactone derivative 8 (X = leaving group)
should lead to the desired tricyclic system 7 by generating ring C.
Application of the Pd-catalysed retron to 8 reveals lactone 9 as
the bridged intermediate. Retro-radical rearrangement ultimately
produces bicyclic lactone 10. Finally, straightforward IEDDA
(Inverse Electron-Demand Diels–Alder) disconnection of bicy-
clo[2.2.2]lactone 10 reveals 3-carbomethoxy-2-pyrone (3-CMP) 11
and the a-substituted vinyl selenide 12, bearing on the terminal
carbon, either a protected alcohol or a suitable leaving group.
The synthesis of 7 begins with the preparation of suitably
functionalised dienophiles 12a–d (Scheme 5). Lithiation of readily
available vinylphenyl selenide 13 by LDA,8 followed by reaction
of the in situ generated alkenyllithium species 13a with TBS-
protected 3-bromopropanol 14, afforded selenide 12a in 70% yield.
Analogously, alkylation of 13a with 3-chloro-1-bromo-propane
15 provided the corresponding dienophile 12b in similar yields.
Fluoride-induced deprotection of the TBS ether in 12a, followed
by acylation of the alcohol function under standard conditions,
then provided the Ac- or Boc-protected dienophiles 12c and d in
high overall yields.
protected derivatives 10c and 10d. The generation of a mixture of
epimers is, however, inconsequential as this stereogenic centre will
be destroyed upon the subsequent radical-initiated rearrangement.
These bicyclo[2.2.2]lactones were then subjected to the action of
TTMSS and AIBN, in boiling benzene (Scheme 6). Gratifyingly,
lactones 10c and 10d underwent clean skeletal reorganisation,
affording adducts 9c and 9d in quantitative yields, following
purification. Much to our surprise, the TBS-protected derivative
10a failed to provide the desired, bridged lactone 9a. Although
complete consumption of the starting material 10a was observed
by TLC, only a complex mixture of products was obtained.10
The stage was now set for the application of our palladium-
catalysed allylic translocation on these functionalised substrates
(Scheme 7). In the event, the conversion of lactones 9c and 9d to
their fused bicyclo[3.3.0] counterparts 8c and 8d smoothly took
place under our newly-developed conditions, providing the desired
adducts in essentially quantitative yields. It thus transpires that
this neutral palladium-catalysed rearrangement protocol is a mild,
highly chemoselective and reliable alternative to our previously
reported acid–based procedure.
Scheme 5 Synthesis of dienophiles 12a–d.
Scheme 7 Palladium-catalysed allylic rearrangement and completion of
the synthesis of 7.
These four diversely decorated dienophiles were then submitted
to the IEDDA reaction with 3-CMP 11, under ou◦r typical high-
pressure conditions (CH2Cl2 as solvent, 16 kbar, 40 C, 3 d).3,9 The
results are depicted in Scheme 6.
To complete our preparation of 7, all that remained was a depro-
tection step, followed by the conversion of the hydroxyl function
into a suitable leaving group and intramolecular cyclisation to
deliver ring C.
This sequence of events could be attained in only two synthetic
operations (Scheme 7). Thus, chemoselective deprotection of 8c
and 8d, using potassium carbonate in refluxing methanol, led
to the same alcohol 17. Mesylation of this compound, followed
by addition of NaH to the reaction mixture, finally afforded the
desired tricyclic compound 7, possessing the oxatriquinane core
structure and featuring two contiguous quaternary centres built
with complete diastereocontrol.
In summary, a novel tandem sequence, featuring a rad-
ical rearrangement–palladium-catalysed translocation of bicy-
clo[2.2.2]lactones, provides an efficient and versatile access to a
variety of fused, bicyclo[3.3.0]lactones. The scope of this procedure
has been evaluated and its synthetic utility demonstrated by the
expeditious assembly of the tricyclic lactone 7, bearing the angular
triquinane core structure, in a small number of steps and good
overall yields. Current efforts are now directed towards expanding
the scope of this unique methodology and applying it to the
synthesis of triquinanes, as well as other relevant natural products.
The results of these investigations will be reported in due course.
Scheme 6 IEDDA reaction and radical-initiated rearrangement.
Whilst the x-chloro selenide 12b afforded only a complex
mixture of products, from which no lactone 10b could be iso-
lated, all the hydroxy-protected dienophiles underwent smoothly
the IEDDA cycloaddition with 3-CMP 11, providing the bicy-
clo[2.2.2]lactones 10a, 10c and 10d in good yields. Interestingly,
adduct 10a was obtained as a 10 : 1 endo–exo (relative to the
selenide substituent) mixture of bicyclic lactones, as opposed to
the exclusive endo-selectivity in the case of the acetate- and Boc-
1466 | Org. Biomol. Chem., 2006, 4, 1464–1467
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
4361. For recent reviews on the asymmetric allylic alkylation reaction,
see: (c) B. M. Trost and M. L. Crawley, Chem. Rev., 2003, 103, 2921;
(d) B. M. Trost, J. Org. Chem., 2004, 69, 5813. See also ref. 1e.
5 For examples, see: (a) H. F. Olivo and J. Yu, Tetrahedron: Asymmetry,
1997, 8, 3785; (b) S. Maki, K. Toyoda, T. Mori, S. Kosemura and S.
Yamamura, Tetrahedron Lett., 1994, 35, 4817; (c) E. J. Bergner and G.
Helmchen, J. Org. Chem., 2000, 65, 5072.
6 Prepared in racemic form from the Inverse Electron-Demand Diels–
Alder (IEDDA) adducts 1a–d by radical rearrangement, as shown
below. Products 1a–d can be obtained in enantioenriched form by using
chiral Lewis acid catalysts. See ref 2a.
Acknowledgements
Financial support for this work by the Universite´ catholique
de Louvain, the National Science Council of Taiwan (post-
doctoral fellowship to J.-H.L.), the Fonds pour la Recherche
dans l’Industrie et l’Agriculture (F.R.I.A., studentship to N.M.),
Merck, Sharp and Dohme (Merck Academic Development Award
to I.E.M.) and SHIMADZU Benelux (financial support for
the acquisition of a FTIR-8400S spectrometer) is gratefully
acknowledged. N.M. is grateful to the Fond National de la
Recherche Scientifique (F.N.R.S.) for receiving an “Aspirant
F.N.R.S.” fellowship.
References
1 For in-depth reviews and treatises, see: (a) Metal-Catalyzed Cross-
Coupling Reactions, 2nd edn, ed. A. de Meijere and F. Diederich, Wiley-
VCH, Weinheim, 2004; (b) L. S. Hegedus, Transition Metals in the
Synthesis of Complex Organic Molecules, 2nd edn, University Science
Books, Sausalito, 1999; (c) Handbook of Organopalladium Chemistry for
Organic Synthesis, ed. E. Negishi, Wiley Interscience, New York, 2002;
(d) J. Tsuji, Palladium Reagents and Catalysts: New perspectives for the
21st Century, Wiley-VCH, Weinheim, 2004. For a recent overview on
applications of palladium-catalysed couplings in total synthesis, see:
(e) K. C. Nicolaou, P. G. Bulger and D. Sarlah, Angew. Chem., Int. Ed.,
2005, 44, 4442.
2 (a) I. E. Marko´, S. L. Warriner and B. Augustyns, Org. Lett., 2000, 2,
3123. For a conceptually similar strategy employed in the construction
of azabicyclic compounds, see: (b) D. M. Hodgson, S. Hachisu and
M. D. Andrews, Org. Lett., 2005, 7, 815 and references therein.
3 B. Augustyns, N. Maulide and I. E. Marko´, Tetrahedron Lett., 2005,
46, 3895.
7 For a review on the synthesis of poliquinane natural products, see:
(a) G. Mehta and A. Srikrishna, Chem. Rev., 1997, 97, 671. For a
recent example of the synthesis of dioxatriquinanes, see: (b) K. P.
Kaliappan and R. S. Nandurdikar, Chem. Commun., 2004, 21,
2506.
8 H. J. Reich, W. W. Willis, Jr. and P. D. Clark, J. Org. Chem., 1981, 46,
2775.
9 For previous examples of IEDDA reactions of 3-CMP, see: (a) I. E.
Marko´ and G. R. Evans, Tetrahedron Lett., 1993, 34, 7309; (b) I. E.
Marko´ and G. R. Evans, Synlett, 1994, 6, 431; (c) I. E. Marko´,
G. R. Evans, P. Seres, I. Chelle´-Regnault, B. Leroy and S. L. Warriner,
Tetrahedron Lett., 1997, 38, 4269; (d) I. E. Marko´, G. R. Evans and J.-P.
Declercq, Tetrahedron, 1994, 50, 4557. For other leading references in
this field, see, e.g.: (e) G. H. Posner and N. Johnson, J. Org. Chem., 1994,
59, 7855. For a review on Diels–Alder cycloadditions of 2-pyrones,
see: (f) K. Afarinkia, V. Vinader, T. D. Nelson and G. H. Posner,
Tetrahedron, 1992, 48, 9111.
4 For early reviews on the Tsuji–Trost reaction, see: (a) B. M. Trost,
Acc. Chem. Res., 1980, 13, 385; (b) J. Tsuji, Tetrahedron, 1986, 42,
10 At this stage it is not clear whether the silyl substituent acts as a radical
quencher by favoring intramolecular radical translocation.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1464–1467 | 1467
©