Bridgehead enolates and bridgehead alkenes in a welwistatin model series

Article abstract of DOI:10.1039/b820674k

Bridgehead metallation is possible in a ketone having the welwistatin skeleton, and this facilitates installation of the isothiocyanate function present in the natural product, and also enables synthesis of remarkable bridgehead alkenes.

Full text of DOI:10.1039/b820674k

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Bridgehead enolates and bridgehead alkenes in a welwistatin model
seriesw
Valerie Boissel,a Nigel S. Simpkins,z*a Gurdip Bhalay,b Alexander J. Blakea
and William Lewisa
Received (in Cambridge, UK) 19th November 2008, Accepted 17th December 2008
First published as an Advance Article on the web 21st January 2009
DOI: 10.1039/b820674k
Bridgehead metallation is possible in a ketone having the
welwistatin skeleton, and this facilitates installation of the
isothiocyanate function present in the natural product, and also
enables synthesis of remarkable bridgehead alkenes.
The welwitindolinones are a small group of oxindole alkaloids,
first isolated from blue–green algae by Moore and co-workers
in 1994.1,2 Of the seven structures originally reported, one
compound, N-methylwelwitindolinone C isothiocyanate (1),
also later dubbed welwistatin, has provoked special synthetic
interest, due both to its fascinating molecular architecture and
its apparent ability to reverse multi-drug resistance (MDR).3
Despite widespread interest in the development of routes
Scheme 1 Silylation of bridged ketone 2.
have uncovered some remarkable transformations of 2,
involving regiocontrolled bridgehead substitutions, and access
towards 1 no total synthesis has emerged to date.4,5
In a previous report we described a four-step synthesis of a
simple model compound 2, having the characteristic [4.3.1]-
to unusual bridgehead alkenes.
We started the exploration of the enolate reactivity of 2 by
probing the regiochemistry of enolate silylation using lithium
skeleton of welwistatin, starting from commercial 4-bromoindole.6
tetramethylpiperidide (LTMP) as base (LDA effected only
ketone reduction), Scheme 1. Treatment of 2 with around
We also showed that the indole bridge present in ketone 2 could be
one equivalent of LTMP gave silylketone 3 cleanly, with no
converted into an oxindole resembling that in welwistatin 1.
trace of the alternative product silylated at the benzylic bridge-
head position. Increasing the amount of base led to formation
of bis-silylated ketone 4, which could be generated in good
yield by employing 3–5 equivalents of base.
Although the regiochemistry of the bridgehead substitution
leading to 3 was not the desired outcome for welwistatin, we
examined the scope of this substitution process with a range of
alternative electrophiles, leading to ketones 5–10.
The extremely concise access to ketone 2 makes further
controlled embellishment of this system a very attractive
option for the preparation of more advanced model com-
pounds related to 1, and potentially for a synthesis of the
natural product itself. Foremost amongst our aspirations for
ketone 2 in this regard were the installation of the bridgehead
isothiocyanate (possibly implicated as the ‘warhead’ function
in the MDR reversal activity), and introduction of an addi-
tional functional group ‘handle’ into the cyclohexanone ring.
Herein we describe our progress in these directions, which
a School of Chemistry, The University of Nottingham, University
Park, Nottingham, UK NG7 2RD. E-mail: n.simpkins@bham.ac.uk;
Fax: +44 (0)121 4144403; Tel: +44 (0)121 4148905
Thus, the putative bridgehead enolate was successfully alkyl-
ated with reactive alkyl halides, acylated, added to benzaldehyde
or selenylated in good to excellent yields.7
b Global Discovery Chemistry, Novartis Institutes of Biomedical
Research, Horsham Research Centre, Horsham, West Sussex, UK
w Electronic supplementary information (ESI) available: Experimental
procedures and spectroscopic data for new compounds. CCDC 698859
(enone 16) and 698860 (enone 18). For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/b820674k
z Current address: School of Chemistry, University of Birmingham,
Edgbaston, Birmingham, UK B15 2TT.
More interesting in terms of substitution at the required
position for installation of the isothiocyanate function present
in 1, was the finding that silylketone 3 could be further
substituted at the benzylic bridgehead position, e.g. to give
ketoester 11 or selenide 12. We considered that desilylation of
ꢀc
This journal is The Royal Society of Chemistry 2009
1398 | Chem. Commun., 2009, 1398–1400

View Article Online
Scheme 3 Selenoxide eliminations to give bridgehead alkenes.
Scheme 2 Synthesis of isothiocyanate 15. (a) TBAF, THF, 87%; (b) LiI,
pyridine, 88%; (c) DPPA, Et3N, CH2Cl2 reflux then toluene
90 1C, 63%; (d) NaOMe, MeOH, 87%; (e) Me3SiI, CH2Cl2 reflux then
MeOH reflux, 90%; (f) thiocarbonyldi-2(1H)-pyridone, DMAP, CH2Cl2,
76%.
In each case the cyclohexenone adopts a distinctive boat
conformation in which conjugation of the carbonyl function
with the newly introduced CQC bond appears compromised.
In accord with this are the observed stretching frequencies of
the CQO group in 16 and 19 of around 1710 cmꢁ1, and also
the chemical shift of the ‘enone’ b C–H, which in each of the
bridgehead enones is at least 0.5 ppm upfield compared to
cyclohexenone. The most remarkable feature of these
molecules is the geometry of the planarised bridgehead
carbon—C12 in 16, C14 in 18—where there are marked
11 and manipulation of the methyl ester function by means of a
Curtius rearrangement sequence would provide access to simple
isothiocyanate models related to 1. In this way the bridgehead
silicon substituent present in ketone 3 would act as a temporary
blocking group. These expectations were realised, although not
without an initial unexpected detour, as shown in Scheme 2.
Desilylation and ester hydrolysis proceeded routinely to give a
bridgehead carboxylic acid, which was then treated with diphenyl-
phosphoryl azide along the lines described by Rawal et al. in their
synthesis of a related welwistatin model isocyanate.5e To our
surprise the product lacked both of the expected ketone and
isocyanate functional groups, and ultimately proved to be azido
oxazolidinone 13.8
This compound could be further processed by oxazolidinone
ring-opening and carbamate removal to give a bridgehead amine.
Treatment of this compound with thiocarbonyldiimidazole
generated only minor amounts of the desired isothiocyanate
due to incomplete conversion of the imidazolyl thiourea
intermediate. By contrast, the less commonly employed reagent
1,10-thiocarbonyl-2,20-pyridone gave clean and efficient con-
version to the desired isothiocyanate 15.9
Fig. 1 X-Ray structure of enone 16.
We were also interested to see if elimination of the bridgehead
selenides 10 and 12 might provide a means to further elaborate
the cyclohexane ring. To this end we effected selenoxide
syn-elimination reactions by reaction of the selenides 10, 12
and 17 with mCPBA in dichloromethane at ꢁ10 1C, Scheme 3.
These reactions led to the formation of the bridgehead
alkenes 16, 18 and 19 in good yields. Although the sum of the
bridging atoms in these systems (the so-called S-value, here 8) is
large enough that these alkenes are not strictly anti-Bredt in
nature, the accommodation of a cyclohexenone into systems
that already posses a rigid methyleneindole bridge appears
remarkable.5g,10 Indeed, alkenes 18 and 19 appeared highly
improbable, based on simple molecular models. Thankfully,
crystalline samples were obtained for 16 and 18, and X-ray
crystallography secured the proposed structures and provided
further insight into the shape and conformation of these
unusual systems Fig. 1 and 2.11
Fig. 2 X-Ray structure of enone 18.
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 1398–1400 | 1399

View Article Online
2 Further members of this family were isolated subsequently, see:
J. I. Jimenez, U. Huber, R. E. Moore and G. M. L. Patterson,
J. Nat. Prod., 1999, 62, 569.
3 C. D. Smith, J. T. Zilfou, K. Stratmann, G. M. Patterson and
R. E. Moore, Mol. Pharmacol., 1995, 47, 241. For a relevant
review, see: I. D. Kerr and N. S. Simpkins, Lett. Drug
Des. Discovery, 2006, 3, 607.
4 For reviews, see: (a) C. Avendano and J. C. Menendez, Curr. Org.
Synth., 2004, 1, 65; (b) J. C. Menendez, Top. Heterocyl. Chem.,
2007, 11, 63.
Scheme 4 Alkene transposition.
5 (a) J. L. Wood, A. A. Holubec, B. M. Stoltz, M. M. Weiss,
J. A. Dixon, B. D. Doan, M. F. Shamji, J. M. Chen and
T. P. Heffron, J. Am. Chem. Soc., 1999, 121, 6326;
(b) G. I. Elliot, J. P. Konopelski and M. M. Olmstead, Org. Lett.,
1999, 1, 1867; (c) M. E. Jung and F. Slowinski, Tetrahedron Lett.,
distortions from ideal trigonal planar geometry. In 16 the
angles subtended at C12 range from 114.43(11) to 128.59(13)1
and sum to 358.101; C12 lies 0.113 A out of the C11–C13–C17
plane. The geometry at C14 in 18 is even more distorted, with
angles ranging from 111.96(15) to 130.99(17)1 and summing to
357.871; C14 lies 0.120 A out of the C6–C13–C15 plane.
We also found that these enones could not be coerced into
Michael addition chemistry, either using soft heteroatom nucleo-
philes, such as thiolate (PhSH, Et3N, CH2Cl2 at reflux) or
organocuprate reagents. Our disappointment at the reluctance
of these bridgehead alkenes to react with nucleophiles might be
compensated if they could instead be reacted with electrophilic
reagents. However, the co-existence of a much more reactive
nucleophilic indole in these systems dictates that conversion into
oxindoles should be carried out prior to attempting electrophilic
additions to the bridgehead alkenes. Such studies are ongoing.
Finally, we were interested to attempt isomerisation of the
bridgehead enones. Treatment of 16 with mild bases such as
DBU or NaOMe did not result in isomerisation of the bridge-
head a,b-unsaturated ketone into a less strained b,g-unsaturated
isomer. However, treatment of 16 with RhCl3 in EtOH at reflux
for 24 h resulted in quite clean isomerisation to give the
transposed enone 20, Scheme 4.12
2001, 42, 6835; (d) H. Deng and J. P. Konopelski, Org. Lett., 2001,
3, 3001; (e) P. Lo
´
´
pez-Alvarado, S. Garcia-Granda, C. Alvarez-Ru
´
a
and C. Avendano, Eur. J. Org. Chem., 2002, 1703; J. A. MacKay,
R. L. Bishop and V. H. Rawal, Org. Lett., 2005, 7, 3421;
(f) T. J. Greshock and R. L. Funk, Org. Lett., 2006, 8, 2643;
(g) R. Lauchli and K. J. Shea, Org. Lett., 2006, 8, 5287; (h) J. Xia,
L. E. Brown and J. P. Konopelski, J. Org. Chem., 2007, 72,
6885.
6 J. Baudoux, A. J. Blake and N. S. Simpkins, Org. Lett., 2005, 7,
4087.
7 For our previous bridgehead enolate studies, see: G. M. P. Giblin,
D. T. Kirk, L. Mitchell and N. S. Simpkins, Org. Lett., 2003, 5,
1673. For recent use of bridgehead enolates in synthesis, see:
(a) C. Tsukano, D. R. Siegel and S. J. Danishefsky, Angew. Chem.,
Int. Ed., 2007, 46, 8840; (b) N. M. Ahmad, V. Rodeschini,
N. S. Simpkins, S. E. Ward and A. J. Blake, J. Org. Chem.,
2007, 72, 4803.
8 We are not aware of direct precedent for this type of structure. This
structure was subsequently confirmed by X-ray crystallography;
further details will be published elsewhere. The undesired excursion
via the azide 13 could be avoided by removal of excess azide salts
from the intermediate acyl azide prior to thermolysis in toluene,
providing easy access to 15.
9 S. Kim and K. Y. Yi, J. Org. Chem., 1986, 51, 2613.
10 S. M. Sheehan, G. Lalic, J. S. Chen and M. D. Shair, Angew,
Chem., Int. Ed., 2000, 39, 2714.
11 For both structures displacement ellipsoids are drawn at the 50%
probability level. Crystal data for enone 16. C16H15NO, FW =
This isomerisation moves the ring alkene into the position
required for welwistatin, making the sequence of bridgehead
selenylation, elimination and isomerisation a new possibility for
installation of the vinyl chloride found in the natural product.
We expect that the alkene function in 20 could serve as a
very useful handle for preparing more advanced welwistatin
models and synthetic intermediates, and we will report on
further progress in due course.
237.29, monoclinic, space group P21/c,
14.083(2), 10.0083(14) A,
a
b
=
=
9.7651(14),
117.807(2)1,
b
=
c
=
U = 1217.4(3) A3, Z = 4, Dc = 1.295 Mg mꢁ3, m(Mo-Ka) =
0.081 mmꢁ1
,
T
=
150(2) K. 2788 unique reflections
(Rint = 0.106). Final R1 [2263 I 4 2s(I)] = 0.0492, wR2
(all data) = 0.134. Enone 18: C19H23NOSi, FW = 309.47,
orthorhombic, space group Pna21,
a = 23.565(2), b =
10.8079(10), c = 6.5185(6) A, U = 1660.2(4) A3, Z = 4,
Dc = 1.238Mg mꢁ3, m(Mo-Ka) = 0.143 mmꢁ1, T = 150(2) K.
3751 unique reflections (Rint = 0.039). Final R1 [2929 I 4 2s(I)] =
0.0349, wR2 (all data) = 0.0741.
Notes and references
1 K. Stratmann, R. E. Moore, R. Bonjouklian, J. B. Deeter, G. M.
L. Patterson, S. Shaffer, C. D. Smith and T. A. Smitka, J. Am.
Chem. Soc., 1994, 116, 9935.
12 P. A. Grieco, M. Nishizawa and N. Marisovic, J. Am. Chem. Soc.,
1976, 98, 7102.
ꢀc
This journal is The Royal Society of Chemistry 2009
1400 | Chem. Commun., 2009, 1398–1400