Novel isomenthone-derived 1,3-diol ligands identified through parallel synthesis and screening catalyse an asymmetric aldol reaction

Article abstract of DOI:10.1039/b212769p

A library of new (+)-isomenthone-derived 1,3-diol ligands, containing 5 contiguous stereocentres (3 fixed, 2 variable), was prepared in 2 steps by parallel synthesis. Evaluation in parallel identified several different ligands from the library for catalys

Full text of DOI:10.1039/b212769p

Novel isomenthone-derived 1,3-diol ligands identified through parallel
synthesis and screening catalyse an asymmetric aldol reaction†
John M. Gardiner,*a Philip D. Crewe,a Gillian E. Smith‡b and Kenneth T. Vealb
a Department of Chemistry, UMIST, Sackville Street, Manchester, UK M60 1QD.
E-mail: john.gardiner@umist.ac.uk; Fax: 001 510 217 2371; Tel: +44 (0)161 200 4530
b GlaxoSmithKline, New Frontiers Science Park, Harlow, Essex, UK CM19 5AW.
E-mail: gillian_e_smith@gsk.com
Received (in Cambridge, UK) 23rd December 2002, Accepted 24th January 2003
First published as an Advance Article on the web 6th February 2003
A library of new (+)-isomenthone-derived 1,3-diol ligands,
containing 5 contiguous stereocentres (3 fixed, 2 variable),
was prepared in 2 steps by parallel synthesis. Evaluation in
parallel identified several different ligands from the library
for catalysing a Mukaiyama aldol reaction with 87–90%
e.e.
dependent on the structure of R, and also the conditions
employed.8,9 To evaluate the facility for different elaborations
from this entry point, several new types of isomenthyl
derivatives have been prepared both through reactions at the
ring carbonyl (variable diastereoselectivity§ ) and through
elaboration of the side chain hydroxy functionality.9
This offers the possibility of generating diverse ligand
structures from this scaffold, based on:
The application of parallel or combinatorial methods to identify
new chiral catalysts has been the subject of considerable
attention in recent years.1 Solid-supported ligand synthesis has
been one significant area, with amide and imine systems being
popular constructs.2 Solution-phase approaches have included
screening arrays of pre-prepared or commercially available
ligands against a target metal and process,3 screening ligand
arrays with metal and/or conditions variability,4 and the
evaluation of mixed ligand combinations.5 Recent years have
seen particularly important strides in analytical methods for
detecting catalytic activity, such as IR thermography, which
have potentially wide applicability and support the parallel
approach.6
Almost all approaches use divergent ligand assembly but
with common ligating functionality types within ligand struc-
tural diversification, and generate stereochemical diversity
through isomer combinations. We were interested in evaluating
an approach which used a chiral scaffold starting point to
develop solution phase parallel diversification of ligand func-
tionality types to 2 or 3 different ligating atom combinations,
and which concurrently introduced limited stereochemical
randomness (Fig. 1). The protocol was directed towards
allowing generation of ligand subsets which would then be
directly screenable against several different catalytic target
reactions. Our choice of ligand scaffold was based on
(+)-isomenthone derivatives which have been almost com-
pletely unexplored as a source of potential reagents, auxiliaries
or ligands, although terpenoids in general have proven a rich
source of efficient chiral controllers.7 We envisaged that
isomenthone chemistry could be extended to (2)-isopinocam-
phone as a pseudo-enantiomeric analogue.8
(1) variations in R (from aldol)
(2) post-aldol elaborations of the ring carbonyl (herein,
reduction to give 3)
(3) post-aldol elaborations of the side chain hydroxyl
(4) introduction of diastereochemical diversity specifically at
1 or 2 of the 5 contiguous stereocentres of (+)-isomenthone-
derived ligands.
Our aim was to develop a protocol enabling generation of
small ligand libraries from a large virtual library¶ (with wide
variations in R, and different elaborations of both carbonyl and
hydroxyl, with stereochemical flexibility), and which could be
screened directly as subsets. These subsets were envisaged as
being available for screening against multiple processes to
amplify the total screening available without additional synthe-
sis. Herein we report a protocol for parallel aldol synthesis (to
2), followed by reduction (to 3) which then allows for direct
catalyst screening in parallel. This provides lead ligands
catalyzing a model Mukaiyama aldol reaction with up to 90%
e.e.
Synthesis of aldol products was optimized for parallel
chemistry,∑ allowing 12-member aldol library generation in a
reaction carousel. The aldol products were then reduced in
parallel directly with LAH, and a minimum work-up afforded
the diol ligands 3 of sufficient chemical purity for screening.
Several of the ligands were formed as diastereomeric mix-
tures** (at either or both the hydroxyl-bearing centres) but these
were not separated for the initial screening reactions. The
reaction vessels were charged with toluene to provide solutions
of the individual ligands for direct catalysis evaluation in
parallel.
We had previously established that (+)-isomenthone 1
undergoes aldol reactions which are diastereoselective at C6
with universally exclusive R configuration to give 2 (established
through a series of X-ray structure analyses) (Scheme 1). The
side chain centre is obtained with variable diastereoselectivity
These members of the diol ligand library were screened in
parallel against
a Mukaiyama aldol reaction (Scheme
2),10,11using Et2AlCl and ligand at 20 mol% catalyst loading, at
Scheme 1 Reagents and conditions: i, LDA; ii, RCHO; iii, LiAlH4.
Fig. 1
† Electronic supplementary information (ESI) available: HPLC data for
Mukaiyama aldol products. NMR data for a pure reference sample of
Mukaiyama aldol product, independently prepared selected examples of
isomenthone aldol constituents and the diols included in the ligand array.
See http://www.rsc.org/suppdata/cc/b2/b212769p/
Scheme 2 Reagents and conditions: i, Et2AlCl, ligand library, 2 h, PhMe.
This journal is © The Royal Society of Chemistry 2003
618
CHEM. COMMUN., 2003, 618–619

View Article Online
Table 1 Parallel screen using catalyst 3 for synthesis of 6
cinnamaldehyde and benzaldehyde (entries 5 and 6, Table 1) were isolated
and X-ray structures obtained. Accurate diastereoisomeric ratios were not
obtained for library members due to NMR overlap or HPLC co-elution.
†† Chiralpak A5; 90 + 10 hexane + propan-2-ol; 3.3 min; 4.1 min. No other
materials observed.
Entry
R
e.e. (%)a
1
2
3
4
5
6
7
8
9
p-bromophenyl
2,5-dichlorophenyl
isopropylb
2-pyridyl
E-2-phenylvinyl
phenyl
2-thiophenyl
E-crotyl
p-methoxyphenyl
2-naphthyl
3,4-methylenedioxyphenyl
2-furyl
89
63
35
88
84
19
72
90
87
72
77
77
1 S. Dahmen and S. Bräse, Synthesis, 2001, 1431–1449; A. Hagemeyer,
B. Jandeleit, Y. Liu, D. M. Poojary, H. W. Turner, A. F. Volpe and W.
H. Weinberg, Appl. Catal., A: Gen., 2001, 221, 23–43; A. Berkessel and
R. Riedl, J. Comb. Chem., 2000, 2, 215–219; B. Jandeleit, D. J.
Schaefer, T. S. Powers, H. W. Turner and W. H. Weinberg, Angew.
Chem., Int. Ed., 1999, 38, 2494–2532; P. P. Perscarmona, J. V. van der
Waal, I. E. Maxwell and T. Maschmeyer, Catal. Lett., 1999, 63, 1–11;
C. Gennari, H. P. Nestler, U. Piarulli and B. Salom, Liebigs Ann. Chem.,
1997, 637–647.
10
11
12
2 G. Liu and J. A. Ellman, J. Org. Chem., 1995, 60, 7712–7713; M. B.
Francis and E. N. Jacobsen, Angew. Chem., Int. Ed., 1999, 38, 937–941;
M. B. Francis and E. N. Jacobsen, Angew. Chem., Int. Ed., 1999, 38,
937–941; A. H. Hoveyda, Chem. Biol., 1998, 5, R187–191; M. S.
Sigman and E. N. Jacobsen, J. Am. Chem. Soc., 1998, 120,
5315–5316.
a Absolute configuration preferred not determined. The sense of selectivity
is, however, the same with all library members as determined by chiral
HPLC. b Only this system screened as a single diastereomer.
ambient temperature (Table 1). The products were analyzed by
chiral HPLC,†† and as shown in Table 1 four ligand systems
catalyzed the reaction with 87–90% e.e. (entries 1, 4, 8, 9).
Clearly, parallel evaluation could be iteratively employed to
screen other reaction parameters such as solvent, ligand and
catalyst loadings and ratio for further optimization for these
ligands, or for other ligand sets using this same basic
protocol.
Since most of the diol ligand solutions screened contained
diastereomeric mixtures of a given ligand 3, the likelihood is
that enhanced e.e. could be achieved by using purified
diastereomeric ligands. The alternative possibility is that non-
linear diastereomeric ligand cooperativity effects may be
observed, which, although less likely, is not unprecedented.12
Future work will evaluate the diastereoisomeric ligand effects
by diastereomer separation prior to screening.
This work demonstrates the viability of generating new
ligand diversity from the (+)-isomenthone skeleton, and
provides a practicable protocol for generating new ligands
directly for catalysis screening. The library synthesis method
can be extended to other aldol libraries, and to larger arrays with
appropriate equipment. The initial work reported here identifies
new chiral ligand leads with good e.e. In addition to this work,
we have elaborated such aldol products in several other ways
into different bi- and trifunctional ligand types, which will
extend the applicability of screening these ligand systems to bi-
and trifunctional arrays.9
3 A. M. Porte, J. Reibenspies and K. Burgess, J. Am. Chem. Soc., 1998,
120, 9180–9187; C. Gennari, S. Ceccarelli, U. Piarulli, C. A. G. N.
Montalbetti and R. F. W. Jackson, J. Org. Chem., 1998, 63, 5312–5313;
R. T. Buck, D. M. Coe, M. J. Drysdale, C. J. Moody and N. D. Pearson,
Tetrahedron Lett., 1998, 39, 7181–7184; A. M. LaPointe, J. Comb.
Chem., 1999, 1, 101–104; S. R. Gilbertson and C.-W. T. Chang, J. Org.
Chem., 1998, 63, 8424–8431; P. J. Fagan, E. Hauptman, R. Shapiro and
A. Casalnuovo, J. Am. Chem. Soc., 2000, 122, 5043–5051.
4 K. Burgess, H.-J. Lim, A. M. Porte and G. A. Solikowski, Angew.
Chem., Int. Ed. Engl., 1996, 35, 220–222; K. Burgess and A. M. Porte,
Tetrahedron: Asymmetry, 1998, 9, 2465–2469; S. Bromidge, P. C.
Wilson and A. Whiting, Tetrahedron Lett., 1998, 39, 8905–8908.
5 C. Moreau, C. G. Frost and B. Murrer, Tetrahedron Lett., 1999, 40,
5617–5620; K. Ding, A. Ishii and K. Mikami, Angew. Chem., Int. Ed.,
1999, 38, 497–501.
6 T. Bein, Angew. Chem., Int. Ed., 1999, 38, 323–326; A. H. Hoveyda,
Chem. Biol., 1998, 5, 305–308.
7 H.-U. Blaser, Chem. Rev., 1992, 92, 935–952.
8 J. H. Chughtai, J. M. Gardiner, S. G. Harris, S. Parsons, D. W. H. Rankin
and C. H. Schwalbe, Tetrahedron Lett., 1997, 38, 9043–9046; J. M.
Gardiner, J. H. Chughtai and I. H. Sadler, Tetrahedron: Asymmetry,
1998, 9, 599–606; In related work we confirmed that (2)-iso-
pinocamphone also undergoes diastereoselective aldol reactions: J. M.
Gardiner and J. H. Chughtai, unpublished results, 1998.
9 J. M. Gardiner, P. D. Crewe, R. G. Pritchard, J. E. Warren, G. E. Smith
and K. T. Veal, unpublished results, 2001; J. M. Gardiner, P. D. Crewe,
R. G. Pritchard, J. E. Warren, G. E. Smith and K. T. Veal, RSC Annual
Congress, 2001.
10 There have been a number of catalysts developed for asymmetric
Mukaiyama aldol reaction: H. Groger, E. M. Vogl and M. Shibasaki,
Chem. Eur. J., 1998, 4, 1137–1141and references therein; Y. Yama-
shita, H. Ishitani, H. Shimizu and S. Kobayashi, J. Am. Chem. Soc.,
2002, 124, 3292–3302 (Zr catalysts); K. Ishihara, S. Kondo and H.
Yamamoto, J. Org. Chem., 2000, 65, 9125–9128 (oxazaborolidines) D.
A. Evans, M. C. Kozlowski, J. A. Murry, C. S. Burgey, K. R. Campos,
B. T. Connell and R. J. Staples, J. Am. Chem. Soc., 1999, 121, 669–685
(Cu catalysts); K. Mikami, S. Matsukawa, Y. Kayaki and T. Ikariya,
Tetrahedron Lett., 2000, 41, 1931–1934 (Ti catalysts) M. T. Reetz, S.-
H. Kyung, C. Bolm and T. Zierke, Chem. Ind., 1986, 824 (Al catalysts)
and references therein.
11 A ligand library was also evaluated for catalysis of another reaction, the
Diels–Alder reaction of cyclopentadiene and methyl acrylate. In this
case, reactions were incomplete and low levels of induction were
observed. This observation is interesting with respect to reference 12,
and to the high level of induction seen in the Mukaiyama aldol
catalysis.
We thank GlaxoSmithKline and EPSRC for a CASE award
(to P. D. C.). NMR, HPLC and IR characterization at UMIST
used instrumentation funded by EPSRC grants GR/L52246
(NMR), GR/M30135 (IR) and GR/L84391 (HPLC).
Notes and references
‡ GSK, Old Powder Mills, Leigh, Nr. Tonbridge, Kent, UK, TN11 9AN.
Fax: +44 (0)1732 372100, Tel: +44 (0)1732 372467.
§ Addition of hydride or various types of C-nucleophiles show a range of
selectivities, some with complete (S)-preference.
¶ This approach could be extended to prepare and screen larger libraries (by
using alternative equipment).
∑ The isomenthone enolate was generated over 5 mins (in THF), aldehyde
added and the reaction quenched by addition of Amberlyst® 15 resin.
Filtration allowed direct reuse of the product solutions for LAH reduction.
Work-up involved addition of a minimal volume of water and NaOH (aq.),
filtration, and sample evaporation before recharging with new solvent to
provide ligand solutions for catalyst screening.
12 Solvent-dependent non-linear effects were observed in catalysis of a
Diels–Alder reaction using a menthone derived diol ligand isomeric to
entry 6, Table 1 (the worst ligand in our case)—and the only literature
example of any structurally similar diol: G. Naraku, K. Hori, Y. Ito and
T. Katsuki, Tetrahedron Lett., 1997, 38, 8231–8232.
** TLC indicated complete consumption of isomenthone (stage 1) and
complete reaction of aldols (stage 2). Pure aldol diastereoisomers from
CHEM. COMMUN., 2003, 618–619
619