3-(1-Aminoalkyl)isoxazole-4-carboxylic acids as peptide bond replacements

Article abstract of DOI:10.1016/S0957-4166(00)00318-9

An orthogonally protected 3-(1-aminoalkyl)isoxazole-4-carboxylic acid has been prepared by 1,3-dipolar cycloaddition of a suitably protected α-aminonitrile oxide with an enaminoester dipolarophile; this protected amino acid has been deprotected and coupled independently at either the C- or N-terminus to produce pseudopeptide segments as peptide mimetics that contain a cis-amide bond replacement. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(00)00318-9

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 11 (2000) 3273–3276
3-(1-Aminoalkyl)isoxazole-4-carboxylic acids as peptide bond
replacements
Raymond C. F. Jones,*,† Stephen J. Hollis and James N. Iley
Department of Chemistry, The Open University, Walton Hall, Milton Keynes MK7 6AA, UK
Received 2 August 2000; accepted 11 August 2000
Abstract
An orthogonally protected 3-(1-aminoalkyl)isoxazole-4-carboxylic acid has been prepared by 1,3-di-
polar cycloaddition of a suitably protected a-aminonitrile oxide with an enaminoester dipolarophile; this
protected amino acid has been deprotected and coupled independently at either the C- or N-terminus to
produce pseudopeptide segments as peptide mimetics that contain a cis-amide bond replacement. © 2000
Elsevier Science Ltd. All rights reserved.
There is an ongoing search for peptide mimetics that incorporate both replacement of
important amide bonds and restriction of conformational freedom relative to the native
peptide.1 Such ‘pseudopeptides’ have potential as agonists or antagonists at peptide receptors, or
as inhibitors of peptidase enzymes. One focus of this search is on molecules that enforce a
reverse turn on a predominantly peptide chain.2 As part of a programme for the incorporation
of heterocycles into peptide mimetics, that has previously reported on the use of cyclic amidines
as amide bond replacements,3 we proposed the 3-(1-aminoalkyl)-4-carboxyisoxazoles 1 as
pseudodipeptides that may be regarded as replacing a cis-amide bond.4 We now report on the
synthesis of suitably orthogonally protected derivatives of isoxazoles 1, and their flexible
incorporation into pseudopeptide segments.
The synthetic strategy was based on 1,3-dipolar cycloaddition of a protected a-aminonitrile
oxide 2 with an enaminoester dipolarophile 3 (Scheme 1).5 Synthesis of the dipole began with
N-tert-butyloxycarbonyl-(S)-alanine methyl ester 4, which was reduced (diisobutylaluminium
hydride, toluene, −78°C) to afford 2-(tert-butyloxycarbonylamino)propanal which was not
generally purified, but converted directly to its oxime 5 (NH2OH·HCl, NaOAc, EtOH–H2O,
60°C) (Scheme 2). Recrystallisation afforded a 2:1 mixture of oxime geometric isomers, syn and
anti geometry not assigned, (63% from ester 4) that proved stable towards racemisation.6
* Corresponding author. E-mail: r.c.f.jones@lboro.ac.uk
†
Current address: Department of Chemistry, Loughborough University, Loughborough, Leics. LE11 3TU, UK.
0957-4166/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(00)00318-9

3274
R. C. F. Jones et al. / Tetrahedron: Asymmetry 11 (2000) 3273–3276
N-Chlorination could be accomplished with N-chlorosuccinimide (CHCl3, reflux; one isomer
converted in 2 h, the other required at least 16 h), but this conversion was performed much
better with tert-butyl hypochlorite (CHCl3, 0°C). When this conversion was complete (45 min as
1
monitored by H NMR spectroscopy), enaminoester 7 (2–3 mol equiv.; previously prepared
from ethyl acetoacetate and pyrrolidine in toluene at reflux, Dean–Stark water removal) was
added to the solution of crude imidoyl chloride 6, and triethylamine was added dropwise
(CHCl3, reflux). Cycloaddition of the nitrile oxide generated in situ from 6, with the enamino-
ester 7 afforded the required isoxazole 9, [h]2D6 −29.8 (c=1.13, CHCl3), as a single regioisomer
(66% from oxime 5), presumably via an intermediate 8 that aromatizes by spontaneous
elimination of pyrrolidine.7
Scheme 1.
Scheme 2. Reagents: (i) i-Bu2AlH, toluene, −78°C; (ii) NH2OH·HCl, NaOAc, EtOH–H2O, 60°C; (iii) t-BuOCl,
CHCl3, 0°C; (iv) Et3N, CHCl3, reflux
Coupling of N-protected amino ester 9 with other amino acid residues could be achieved via
either initial C- or N-terminal deprotection. For example, hydrolysis of ester 9 (LiOH, aq.
EtOH–THF, 20°C, 16 h) afforded the isoxazole-4-carboxylic acid 10 (65%), [h]2D2 −22.1 (c=1.08,
MeOH) (Scheme 3). This was coupled with glycine ethyl ester hydrochloride [1-(3-dimethyl-
aminopropyl)-3-ethylcarbodiimide hydrochloride (EDCI), Et3N, CH2Cl2 or ether] to afford
amide 11a (39%), [h]2D2 −24.0 (c=0.51, MeOH), and with (S)-alanine methyl ester hydrochloride
under these conditions to afford amide 11b (24%), [h]2D5 +14.2 (c=1.37, CHCl3). The latter
1
compound was isolated as a single diastereoisomer (HPLC, 13C and H NMR spectra), verifying
the chiral integrity of isoxazole 9 and carboxylic acid 10. For comparison, when hydrolysis of
ester 9 was effected with KOH in ethanol at reflux, coupling of this sample of acid 10 (97%) with

R. C. F. Jones et al. / Tetrahedron: Asymmetry 11 (2000) 3273–3276
3275
(S)-alanine methyl ester hydrochloride under the same conditions afforded amide 11b but as a
5:1 mixture of two diastereoisomers. Subsequent unmasking of the N-terminus of pseudodi-
peptide 11a (AcCl, EtOH) to give amine 12 as its hydrochloride salt (quantitative) was followed
by coupling of the free amine 12 with N-tert-butyloxycarbonylglycine (EDCI) to give the
pseudotetrapeptide segment 13 in low yield (10%), [h]2D8 +7.2 (c=1.31, CHCl3).
Scheme 3. Reagents: (i) LiOH, aq. EtOH–THF; (ii) H-Gly-OEt·HCl (for 11a) or H-Ala-OMe·HCl (for 11b), EDCI,
Et3N, CH2Cl2; (iii) AcCl, EtOH; (iv) Boc-Gly-OH, EDCI; (v) TFA, CH2Cl2, then 2 M aq. HCl; (vi) Boc-Gly-OH (for
15a) or Boc-Ala-OH (for 15b), EDCI, Et3N; (vii) H-Gly-OEt·HCl, EDCI, Et3N
The orthogonal sequence could also be employed, with initial N-deprotection of isoxazole
ester 9 (TFA–CH2Cl2; then 2 M aq. HCl) to afford the 3-(1-aminoalkyl)isoxazole 14 as its
hydrochloride salt (94%; more easily handled than the trifluoroacetate salt), [h]2D5 −29.9 (c=1.03,
MeOH). The amine salt 14 could be coupled to N-tert-butyloxycarbonylglycine or N-tert-butyl-
oxycarbonyl-(S)-alanine (EDCI, Et3N) to afford the pseudotripeptide amides 15a, [h]2D7 −31.3
(c=1.05, CHCl3), and 15b, [h]2D5 −49.0 (c=1.11, CHCl3), respectively (57, 74%); again, the latter
1
compound was produced as a single diastereoisomer (HPLC, 13C and H NMR spectra),
reinforcing the conclusion that isoxazole ester 9 had been formed without racemisation.
Hydrolysis of the ester 15a as before (LiOH, aq. EtOH–THF; 43%) was followed by coupling
of the crude carboxylic acid 16 with glycine ethyl ester hydrochloride as usual, to again afford
pseudopeptide segment 13 (27%) by this alternative sequence. The sometimes low yields
obtained by coupling to the termini of amino acid residue 1 may be attributed to deactivation
of the C-terminus, which is a vinylogous carbonate, and also to steric factors.
We have thus demonstrated the formation and incorporation into pseudopeptides of the
isoxazole amino acid moiety 1, as a cis-amide bond replacement. Preliminary molecular

3276
R. C. F. Jones et al. / Tetrahedron: Asymmetry 11 (2000) 3273–3276
modelling of these systems8 indicates that they have a number of minimum energy conforma-
tions that place a turn in the peptide backbone.
Acknowledgements
We acknowledge the financial support of the Open University (studentship to S.J.H.), and the
EPSRC National Mass Spectrometry Service Centre (Swansea) for some MS data.
References
1. See, for example: Advances in Amino Acid Mimetics and Peptidomimetics; Abell, A., Ed.; Jai Press Inc: Stamford,
USA, 1999; Vol. 2.
2. See, for example: Symposia-in-Print No. 50, Peptide Secondary Structure Mimetics; Kahn, M., Guest Editor,
Tetrahedron 1993, 49, 3433; Hanessian, S.; McNaughton-Smith, G.; Lombart, H. G.; Lubell, W. D. Tetrahedron
1997, 53, 12789–12854 and references therein.
3. Jones, R. C. F.; Gilbert, I. H.; Rees, D. C.; Crockett, A. K. Tetrahedron 1995, 51, 6315–6336.
4. See: Chung, Y. J.; Ryu, E. J.; Keum, G.; Kim, B. H. Bioorg. Med. Chem. 1996, 4, 209–225, for a related study
with 3-(1-aminoalkyl)-5-carboxy-4,5-dihydroisoxazoles.
5. Jones, R. C. F.; Dawson, C. E.; O’Mahony, M. J. Synlett 1999, 873–876.
6. Cf. Ito, A.; Takahashi, R.; Baba, Y. Chem. Pharm. Bull. 1975, 23, 3081–3087.
7. Stork, G.; McMurry, J. E. J. Am. Chem. Soc. 1967, 89, 5461–5462; McMurry, J. E. Org. Synth. 1973, 53, 59–62;
Zong, K.; Shin, S. I.; Jeon, D. J.; Lee, J. N.; Ryu, E. K. J. Heterocycl. Chem. 2000, 37, 75–78.
8. Performed in SPARTAN 4.1 using the AM1 semi-empirical method on a Silicon Graphics Indy workstation.
.