Catalytic asymmetric synthesis of glutamate analogues

Article abstract of DOI:10.1021/ol049619m

Utilizing our lateral metalation coupled with Jacobsen's catalytic asymmetric amino nitrile synthesis, we have demonstrated the ability to synthesize isoxazole-containing amino acid glutamate analogues in high yield and high enantiomeric excesses. Chiral

Full text of DOI:10.1021/ol049619m

ORGANIC
LETTERS
2004
Vol. 6, No. 8
1285-1288
Catalytic Asymmetric Synthesis of
Glutamate Analogues
David J. Burkhart, Andrew R. McKenzie, Jared K. Nelson, Katherine I. Myers,
,‡
Xue Zhao, Kathy R. Magnusson,† and Nicholas R. Natale*
301 Renfrew Hall, Department of Chemistry, UniVersity of Idaho, Moscow,
Idaho 83844-2343, and 258 Life Science Building, Department of Biological Sciences,
UniVersity of Idaho, Moscow, Idaho 83844-3051
nrnatale@uidaho.edu
Received March 1, 2004
ABSTRACT
Utilizing our lateral metalation coupled with Jacobsen’s catalytic asymmetric amino nitrile synthesis, we have demonstrated the ability to
synthesize isoxazole-containing amino acid glutamate analogues in high yield and high enantiomeric excesses. Chiral centers r or â at the
C-5 position do not detract from diastereoselectivity of the Jacobsen−Strecker reaction.
The asymmetric Strecker reaction to produce optically pure
amino nitriles is of synthetic and medicinal value since it
provides a direct route to chiral amino acids.1 (S)-Glutamic
acid is the primary excitatory neurotransmitter of the
mammalian CNS.2 Three classes of heterogeneous ionotropic
glutamate receptors (iGluR) named for their respective
agonists, (S)-2-amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)-
propionic acid (AMPA), N-methyl-D-aspartic acid (NMDA),
and kainic acid (KA), make up the bulk of excitatory
neurotransmission in the human CNS.3
These proteins play a significant role in a wide variety of
neurological pathways, including cognition, learning, and
memory formation, and contain small but significant differ-
ences in their subunit composition.4-6 Excitatory imbalance
of these receptors is thought to be a primary culprit in
neurological disorders including epilepsy, cerebral ischemia,
Parkinson’s, Huntington’s, and Alzheimer’s diseases. The
synthesis of subunit selective agonists and antagonists as
potential therapeutics is, therefore, of great importance.7
Structure-activity relationship studies of AMPA analogues
show a distinct enantioselectivity of action. C-5 lipophilic
aryl and alkyl analogues of AMPA bearing the (S)-amino
acid configuration are agonists at AMPA receptors whereas
their enantiomers are antagonists or are inactive.8
The C-5 substituents of analogues bearing the (R)-amino
acid configuration likely disrupt the interdomain interactions
necessary for domain closure and activation of iGluRs
resulting in antagonism. The X-ray structure of (S)-2-amino-
3-(5-tert-butyl-3-hydroxy-4-isoxazolyl)propionic acid ((S)-
ATPA) in complex with the GluR2 subunit reveals this
† Department of Biological Sciences.
(5) Watkins, J. C.; Krogsgaard-Larsen, P.; Honore, T. Trends Pharmacol.
Sci. 1990, 11, 25-33.
‡ Department of Chemistry.
(1) Williams, R. M. Synthesis of Optically ActiVe Amino Acids;
Pergamon: Elmsford, 1989; pp 208-229.
(2) Wheal, H. V.; Thomson, A. M. Excitatory Amino Acids and Synaptic
Transmissionp Academic Press: London, 1995.
(3) Monaghan, D. T.; Wenthold, R. J. The Ionotropic Glutamate
Receptors; Humana Press: Totowa, NJ, 1997.
(6) Watkins, J. C.; Collingridge, G. L. The NMDA Receptor; Oxford
University Press: Oxford, 1989. (b) Krogsgaard-Larsen, P. Pharmacol.
Toxicol. 1992, 70, 95-104.
(7) Brauner-Osborne, H.; Egebjerg, J.; Nielsen, E. O.; Madsen, U.;
Krogsgaard-Larsen, P. J. Med. Chem. 2000, 43, 2609-2645.
(8) Sorensen, U. S.; Falch, E.; Stensbol, T. B.; Jaroszewski, J. W.;
Madsen, U.; Krogsgaard-Larsen, P. Arch. Pharm. Pharm. Med. Chem. 2001,
334, 62-68.
(4) Lodge, D.; Ed. Excitatory Amino Acids in Health and Disease;
Wiley: Chichester, 1988.
10.1021/ol049619m CCC: $27.50 © 2004 American Chemical Society
Published on Web 03/23/2004

unique interaction between the C-5 tert-butyl group of ATPA
and a group of amino acids which make up a partially
lipophillic pocket, giving ATPA a 50-fold selectivity for the
GluR5 subunit. The amino acids which describe the lipo-
phillic pocket vary between subunits which make up AMPA
(GluR1-4) and kainate (GluR5-7 and KA1-2) receptors,
and this makes an attractive target for the design of subunit
selective agonists and antagonists. Based on molecular
modeling using Gouaux’s coordinates,9 we established a
working hypothesis for the receptor’s tolerance of C-5
lipophilic groups (Figure 1).
The Br-BINOL-zirconium catalyst pioneered by Kobayashi,17
and the 5-pivaloyl-substituted Schiff base developed by
Jacobsen18 and co-workers, were the most successful with
both aryl and alkyl aldehydes. Both catalyst systems were
applied to the asymmetric synthesis of ACPA to determine
the best method for production of optically active ACPA
analogues bearing lipophilic and/or chiral hydroxyl groups
at the C-5 position.
For the synthesis of C-5 lipophilic aryl ACPA analogues,
lateral metalation of an isoxazolyl acetal,19 followed by
electrophilic quenching produced 1a-d in good yields
(Scheme 1). To achieve the proper carbon framework of
Scheme 1 a
Figure 1. Molecular surface modeling of the ligand binding domain
of the apo state of the AMPA receptor, bound with putative ligand,
20, using the InsightII program. SGI Surface modeling reveals the
dimensions of the lipophilic pocket in the GluR2 subunit (see the
Supporting Information).
The additional effect of a heteroatom proximal to the
isoxazole ring at the C-5 position is also known to enhance
activity and binding affinity.10 We sought to combine these
effects in a survey of the role of chirality at the C-5 position
of AMPA analogues in search of highly subunit selective
compounds using steric bulk and chirality as selectivity
filters.
To date, the enantioselective syntheses of AMPA ana-
logues have employed chiral pool techniques with limited
application or asymmetric Strecker reactions via chiral
auxiliaries, which resulted in poor optical purity.11,12
Using the highly potent AMPA agonist (S)-2-amino-3-(3-
carboxy-5-methyl-4-isoxazolyl)propionic acid (ACPA)13 as
a lead, we report here the first catalytic asymmetric synthesis
of ACPA and C-5 analogues. Numerous catalysts for the
asymmetric Strecker reaction producing highly enantio-
enriched amino nitriles, have appeared in recent years.14-16
a Key: (i) BF3 etherate, MeOH, lead(IV) acetate, benzene; (ii)
5.0 equiv of BH3 dimethyl sulfide, THF, rt, 30 h; (iii) Dess-Martin
periodinane, CH2Cl2, rt; (iv) 2-amino-m-cresol, HCN, 10 mol %
Zr-Br-BINOL, CH2Cl2, -45 °C, 60 h, then MeI, acetone; (v)
MeOH, HCl, rt, 12 h.
ACPA, these compounds were readily homologated to their
methyl ester via the modified Willgerodt-Kindler conditions
of Ila and Junjappa.20 Borane-THF selectively reduced the
methyl ester of 2a-d to give alcohols 3a-d in good yield,
followed by Dess-Martin21 oxidation to produce aldehydes
4a-d.
(14) Iyer, M. S.; Gigstad, K. M.; Namdev, N. D.; Lipton, M. J. Am.
Chem. Soc. 1996, 118, 4910-4911.
(15) Sigman, M. S.; Jacobsen, E. N. J. Am. Chem. Soc. 1998, 120, 5315-
5316.
(9) Lunn, M.; Hogner, A.; Stensbøl, T.; Gouaux, E.; Egebjerg, J.; Kastrup,
J. S. J. Med. Chem. 2003, 46, 872-875.
(10) Falch, E.; Brehm. L.; Mikkelsen, I.; Johansen, T. N.; Skjaerbaek,
N.; Nielsen, B.; Stensbol, T. B.; Ebert, B.; Krogsgaard-Larsen, P. J. Med.
Chem. 1998, 41, 2513-2523.
(16) Krueger, C. A.; Kuntz, K. W.; Dzierba, C. D.; Wirschun, W. G.;
Gleason, J. D.; Snapper, M. L.; Hoveyda, A. H. J. Am. Chem. Soc. 1999,
121, 4284-4285.
(11) Pajouhesh, H.; Curry, K.; Tetrahedron: Asymmetry 1998, 9, 2757-
(17) Ishitani, H.; Komiyama, S.; Hasegawa, Y.; Kobayashi, S. J. Am.
Chem. Soc. 2000, 122, 762-766.
2760.
(12) Ma, D.; Tang, G.; Tian, H.; Zou, G. Tetrahedron Lett. 1999, 40,
5753-5756.
(13) Johansen, T. N.; Stensbol, T. B.; Nielsen, B.; Vogensen, S. B.;
Frydenvang, K.; Slok, F. A.; J.; Madsen, U.; Krogsgaard-Larsen, P. Chirality
2001, 13, 523-532.
(18) Sigman, M. S.; Vachal, P.; Jacobsen. E. N. Angew. Chem., Int. Ed.
2000, 39, 1279-1281.
(19) Burkhart, D.; Zhou, P.; Blumenfeld, A.; Twamley, B.; Natale, N.;
Tetrahedron 2001, 57, 8039-8046.
(20) Ila, H.; Junjappa, H. Synth. Commun. 1981, 81, 126-127.
1286
Org. Lett., Vol. 6, No. 8, 2004

reaction catalyzed by Jacobsen’s L,R,R-peptide23 provided
allylamino nitriles in excellent yield (>90%) and in good to
excellent optical purity (Scheme 2). A higher catalyst loading
(10 mol %) was required to give consistent results with this
method, but with the ease of the catalyst synthesis it did not
present a problem. Pinner synthesis24 produced amino esters
8a-d, followed by Pd(0)-deallylation25 of the amines to give
9a-d. HCl (2 M) hydrolysis gave the final amino acids in
good yield with no detectable racemization (Scheme 2). The
absolute configuration was established on the basis of the
published rotation of (S)-ACPA.13 Optical purity was deter-
mined by chiral HPLC analysis of racemic mixtures and
asymmetric reaction products for all compounds (see the
Supporting Information).
Scheme 2 a
The aldehydes were converted to the amino acids in
synthetically useful yields with no chromatography. For the
synthesis of ACPA analogues with chiral carbon atoms
bearing heteroatoms R and â to the isoxazole ring, com-
pounds produced by lateral metalation 11a,b were again
employed (Scheme 3).
a Key: (i) allylamine, 10% Jacobsen’s L,R,R peptide catalyst,
TMSCN, MeOH, toluene, -78 °C; (ii) MeOH, HCl, rt, 8 h; (iii)
10 mol % Pd(PPh3)4, N,N′-dimethylbarbituric acid, CH2Cl2, 35 °C;
(iv) 2 M HCl, 74 °C, 22 h.
Since the imines of aldehyde 4a were found to be unstable,
the Kobayashi method offered the advantage of a true three-
component Strecker reaction and was applied first. The
results of the three component asymmetric Strecker reaction
for aldehydes 4a and 4b, using 10 mol % of the Kobayashi
zirconium-Br-BINOL catalyst and 2-amino-m-cresol are
summarized in Scheme 1. A high degree of enantioselectivity
was observed and extended reaction times improved the
yields somewhat, but repeated attempts at deprotection (ceric
ammonium nitrate oxidative cleavage)22 of the R-amino ester
product at various temperatures resulted in decomposition,
rendering the synthesis impractical in our hands.
Over 50 attempts at direct asymmetric induction during
the lateral metalation step in the presence of chiral bis-
oxazolines, binaphthols, spartaiene, and other chiral catalysts
or stoichiometric reagent based adjuvants gave disappointing
results: 6-19% ee. Alternatively, Dess-Martin oxidation
produced the ketones 12a,b in good yield although compound
12b was found to be highly light sensitive.
The asymmetric borane reduction using 10 mol % of
Corey’s 2-methyl-CBS-oxazaborolidine27 produced R- and
â-isoxazolyl alcohols 13a,b in good yield and good optical
purity. The sense of asymmetric induction (R gives S) was
in agreement with that reported by Corey and co-workers.
The tert-butyldiphenylsilyl protecting group was thought to
Using in situ Schiff base formation to circumvent the
instability of aliphatic imines, the asymmetric Strecker
Scheme 3 a
a Key: Dess-Martin periodinane, CH2Cl2, rt; (ii) 10 mol % (R)-2-methyl-CBS-oxazoborolidine, CH2Cl2, 1 equiv of BH3-DMS, -45
°C, 18 h; (iii) 2 equiv of imidazole, 10 mol % DMAP, 1.2 equiv of TPS-Cl, CH2Cl2; (iv) 1 equiv of TsOH, acetone; (v) BF3‚Et2O, MeOH,
lead(IV) acetate, benzene; (vi) 5.0 equiv of borane-DMS, THF, rt 30 h; (vii) Dess-Martin periodinane, CH2Cl2, rt; (viii) allylamine, 10
mol % Jacobsen L,R,R catalyst, TMSCN, MeOH, toluene, -78 °C; (ix) MeOH, HCl, rt, 8 h; (x) 10 mol % Pd(PPh3)4, NMDBA,CH2Cl2
35 °C; (xi) 2 M HCl, 74 °C, 22 h.
Org. Lett., Vol. 6, No. 8, 2004
1287

be robust enough to permit the alcohols to be taken through
the homologation sequence and the silyl ethers were formed
quantitatively.
The p-toluenesulfonic acid hydrolysis gave the methyl
ketones 15a,b in good yield and these compounds were
carried through to the aldehydes 17a,b as before. The
chemical yields for the R TPS-ether were dramatically lower,
presumably due to the steric bulk proximal to the C-4
position. The modified Jacobsen Strecker conditions used
previously were again successful in producing the amino
nitriles 18a,b in excellent yield (>90%) and diastereomeric
excess (>90%).
A significant degree of double asymmetric induction was
observed for the alpha TPS-ethers as the (S,S) isomer was
formed in 99% de versus 90% for the (R,S) diastereomer.
The Pinner reaction produced the amino esters and cleaved
the TPS protecting group in a single step in 53% yield. It is
noteworthy that the 2 M HCl hydrolysis in the final step did
not result in any detectable loss of the C-5 alcohol moiety.
This synthetic strategy enables the production of ACPA
derivatives containing lipophilic aryl groups at the C-5
position in addition to chiral hydroxyl groups R or â to the
isoxazole to take advantage of the proximal heteroatom effect
at AMPA receptors.
This modification of the Jacobsen asymmetric Strecker
reaction permits the facile asymmetric synthesis of optically
active ACPA analogues eliminating isomeric ballast and the
need for tedious resolution.
The overall yield for our catalytic ACPA synthesis is 32%,
compared to the previously reported resolution route of
0.2%.13 We have examined competitive binding of [3H]-
AMPA with (S)-ACPA, 10a, prepared by our independent
route described herein. Figure 2 dramatically illustrates that
Figure 2. ACPA inhibition of [3H]AMPA binding in mouse brain.
(A, B) Pseudocolored computer images of representative auto-
radiograms of [3H]AMPA binding (50 nM in incubation solution)
in the absence (A) and presence (B) of 10 µM (S)-ACPA, 10a, in
horizontal sections of mouse brain. (C) Legend indicates densities
(fmol/mg protein) of binding for different pseudocolor levels.26
(S)-ACPA, 10a, completely displaces [3H]-AMPA from
mouse brain within experimental error. The complete bio-
logical evaluation of our ACPA analogues will be described
in due course. Using this synthetic route, structure based de-
sign will permit further lead optimization in pursuit of subunit
selective therapeutic agents for neurological disorders.
Acknowledgment. Financial support of this research from
the National Institute of Neurological Disorders and Stroke
(Grant NS38444-01) is gratefully acknowledged. We thank
Dr. Eric Jacobsen for a helpful discussion of the peptide
catalyst synthesis. We acknowledge Dr. Shu Kobayashi for
a helpful discussion of his catalyst system. We thank Dr.
Andrzej Paszczynski for access to molecular modeling
facilities (EBI).
(21) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277-
7287.
(22) Kronenthal, D. R.; Han, C. Y.; Taylor, M. K.; J. Org. Chem. 1982,
47, 2765-2768.
(23) Su, J. T.; Vachal, P.; Jacobsen, E. N. AdV. Synth. Catal. 2001, 343,
197-200.
Supporting Information Available: Experimental details
and full characterization data for all final products. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(24) Compagnon, P. L.; Miocque, M. Ann. Chim. (Paris) 1970, 14, 23-
37.
(25) Garro-Helion, F.; Merzouk, A.; Guibe, F. J. Org. Chem. 1993, 58,
6109-6113.
(26) Magnusson, Kathy R. Mech Aging DeV., 1998, 104, 227-248.
(27) Corey, E. J.; Bakshi, R. K.; Shibata, S. J. Am. Chem. Soc. 1987,
109, 5551-5553.
OL049619M
1288
Org. Lett., Vol. 6, No. 8, 2004