Efficient method for the preparation of ( S)-5-hydroxynorvaline

TETRAHEDRON: ASYMMETRY Pergamon Tetrahedron: Asymmetry 11 (2000) 991–994 Efficient method for the preparation of (S)-5-hydroxynorvaline Mónica García,a Anna Serra,a Mario Rubiralta,a Anna Diez,a,∗ Víctor Segarra,b Estrella Lozoya,b Hamish Ryder b and José M. Palacios b a Laboratori de Química Orgànica, Facultat de Farmàcia, Universitat de Barcelona, 08028 Barcelona, Spain b Almirall Prodesfarma, Research Center, Cardener, 68–74, 08024 Barcelona, Spain Received 15 December 1999; accepted 10 January 2000 Abstract (S)-5-Hydroxynorvaline 4 has been prepared from L-glutamic acid 1 by simultaneous protection of the α-amino and α-carboxyl groups, and selective reduction of the resulting boroxazolidone 2. This method is rapid and highly reproducible, and gives very pure (S)-5-hydroxynorvaline after simple anion-exchange purification. It improves existing methods by providing a purer product in higher yields. © 2000 Elsevier Science Ltd. All rights reserved. 1. Introduction Non-proteinogenic hydroxy α-amino acids have been identified as biosynthetic precursors of natural products made by plants1,2 and microorganisms.3,4 In particular, 5-hydroxynorvaline 4 has recently been described as a specific marker of oxidised proteins in the study of age-related diseases.5 This unnatural amino acid has also been used to establish structure–activity relationships of bioactive molecules like cyclosporine,6 and of microbial enzymes.7 In addition, 5-hydroxynorvaline has been used in the attachment of glycosyl derivatives in glycopeptide solid phase synthesis.8 In our case, 5-hydroxynorvaline is essential as a starting material for the synthesis of conformationally restricted pseudopeptides presenting a 3-aminopiperidin-2-one backbone.9–11 2. Results and discussion Our new method for the preparation of 5-hydroxy-2-aminovaleric acid 4 improves the overall yield obtained by classical reduction routes,12 and avoids the problems stemming from the instability of trityl as the amino protecting group.13 ∗ Corresponding author. Laboratori de Química Orgànica, Facultat de Farmàcia, Av. Joan XXIII, s/n, 08028 Barcelona, Spain. Tel: 34 93 402 45 37; fax: 34 93 402 45 39; e-mail: adiez@farmacia.far.ub.es 0957-4166/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved. P I I: S0957-4166(00)00020-3 tetasy 3243 992 M. García et al. / Tetrahedron: Asymmetry 11 (2000) 991–994 Simultaneous protection of the α-amino and the α-carboxyl groups was accomplished by treatment of acid with triethylborane14 in 1,2-dimethoxyethane (DME) (Scheme 1). The presence of two non-equivalent NH protons in the 1 H NMR spectrum allowed unambiguous identification of the product as boroxazolidone 2. The DME soluble complex was directly reduced with BH3 , and final hydrolysis and ion-exchange purification gave pure (S)-5-hydroxynorvaline 4 in 60% overall yield. Compound 4 was identified through the presence of the methylene signal at δ 58.8 in its 13 C NMR spectrum, corresponding to the C-5 reduced position, and of the IR absorption bands at 3400 cm−1 and 2900 cm−1 characteristic of the hydroxyl and amino groups, respectively. L -glutamic Scheme 1. Although the methodology described has been applied to the synthesis of α-amino-γ-butyrolactone,15 in the present case the lactonisation was prevented by the acidity of the medium16 and the conditions of purification. The enantiomeric purity of the synthesised hydroxy amino acid 4 was determined by HPLC, using L -glutamic acid 1 as the reference. For this purpose, both (S)-5-hydroxynorvaline and L -glutamic acid were derivatised with Marfey’s reagent17,18 and analysed by reverse phase HPLC (see the Experimental section). The hydroxy amino acid was detected as a single peak with a retention time of 4.4 min, thus proving that no racemisation occurred during the synthesis. 3. Experimental 3.1. General procedures The melting point was determined in a capillary tube on a Büchi apparatus. Optical rotations were measured with a Perkin–Elmer 241 polarimeter, at 23°C. 1 H and 13 C NMR spectra were recorded on a Varian Gemini-200 instrument (200 MHz) and chemical shifts are expressed in parts per million (δ) relative to Me4 Si. IR spectra were registered on a Nicolet FT-IR spectrophotometer. Mass spectra were determined on a Hewlett–Packard 5988A mass spectrometer by electronic impact (EIMS). The HPLC instrument (Waters) consisted of a pump (Model 515) equipped with a 20 µl sample loop, a C18 column, a UV–vis HPLC detector (Waters 2487) and a model 746 integrator. TLC was performed on SiO2 (silica gel 60 F254, Macherey–Nagel) and developed with n-BuOH:AcOH:H2 O (4:1:1). The spots were located with ninhydrin reagent or KMnO4 . Purification of reagents and solvents was effected according to standard methods. Microanalyses were performed on a Carlo Erba 1106 analyser at the Serveis Científico-Tècnics (Universitat de Barcelona). M. García et al. / Tetrahedron: Asymmetry 11 (2000) 991–994 993 3.2. B,B-Diethylboroxazolidone 2 To a suspension of L-glutamic acid 1 (1 g, 6.8 mmol) in DME (8 ml), triethylborane–THF (8 ml, 8 mmol) was added. The mixture was refluxed under N2 until the solution was clear (2 days). Remains of the insoluble starting material were removed by filtration and the solvent was evaporated. The resulting oil was washed with petroleum ether to obtain boroxazolidone 2 (1.34 g, 92%). TLC indicated the product homogeneity (Rf =0.64). 1 H NMR (d6 DMSO) δ 0.09 (br q, J=33 Hz, 4H, CH2 ), 0.56 (m, 6H, CH3 ), 1.60 (m, 1H, β-H), 1.90 (m, 1H, β′ -H), 2.38 (t, J=18 Hz, 2H, γ-H), 3.60 (m, 1H, α-H), 5.45 and 6.40 (2 br t, J=19 Hz, 1H each, NH); 13 C NMR (d6 DMSO) δ 8.9 (CH3 ), 8.9 (CH3 ), 12.1 (CH2 ), 12.8 (CH2 ), 25.9 (C-3), 30.1 (C-4), 53.6 (Cα), 173.7 (C-5), 173.9 (C-1). 3.3. (S)-5-Hydroxynorvaline 4 To a solution of the amino acid–borane complex 2 (1.34 g, 6.2 mmol) in DME (6.8 ml), cooled to 0°C, BH3 –THF (6.2 ml, 6.2 mmol) was added dropwise. The reaction was stirred under N2 at 0°C for 4 h, and at room temperature for an additional 20 h. HCl (1.5 M, 5 ml) was added and the solvent was removed under reduced pressure. A solution of the residue in HCl (1.5 M, 5 ml) was refluxed for 45 min to allow complete hydrolysis of the complex and filtered to remove the non-reduced glutamic acid. The filtrate was dried in vacuo and washed several times with MeOH to remove boric acid. TLC indicated that 5-hydroxynorvaline 4 (Rf =0.17) was free of glutamic acid 1 (Rf =0.06). Two other less polar spots (Rf =0.27 and 0.33), that could be consistent with lactonisation products, were located by ninhydrin reagent. Purification was achieved by filtration through an anion-exchange resin (OH− ) form.18 The resin was prepared by washing commercial Amberlite® (IR-400, 20–40 mesh, Cl− form) with 1 M aqueous NaOH until a negative chloride test was obtained. The crude 5-hydroxynorvaline was dissolved in aqueous NaOH (pH=10, 100 ml) before elution. The column was eluted with H2 O and a gradient from 0.1 to 2.0 M AcOH. Fractions containing 4 (Rf =0.17) were combined and freeze-dried to give a white solid (565 mg, 70%). Mp. 210°C (H2 O–MeOH); [α]D =+4 (c 1, H2 O); [α]D =+16 (c 1, 0.6 M HCl); IR (KBr) 3400 (OH), 2900 (NH2 ), 1580 (CO) cm−1 ; 1 H NMR (d6 DMSO) δ 1.47 (m, 2H, β-H), 1.74 (m, 2H, γ-H), 3.46 (t, J=12 Hz, 2H, δ-H), 3.59 (t, J=12 Hz, 1H, α-H); 13 C NMR (d6 DMSO) δ 24.9 (C-3), 25.0 (C-4), 52.4 (Cα), 58.8 (C-5), 172.4 (C-1); EIMS m/z (%) 134 (M+ , 1), 102 (2), 88 (34), 71 (100), 56 (34). Anal. calcd for C5 H11 NO3 : C, 45.10; H, 8.33; N, 10.52; found: C, 44.84; H, 8.31; N, 10.27. 3.4. High performance liquid chromatography of the Marfey derivatives To a solution of the amino acid (5 µmol) in aqueous NaHCO3 (1 M, 100 µl), a solution of Marfey’s reagent (1-fluoro-2,4-dinitrophenyl-5-L-alanine amide) in acetone (10 mg/ml, 200 µl) was added, and the reaction was stirred at 40°C for 1 h. The mixture was then acidified with HCl (2 M, 20 µl) and aqueous MeCN (40%, 4 ml) was added prior to analysis by reverse-phase HPLC (Nucleosil column C18, 250×4.6 mm; eluents MeCN:H2 O:TFA (40:60:0.1); UV detection 340 nm; flow rate 1.0 ml/min). A single peak was detected for derivatised L-glutamic acid 1 (used as the reference) with a retention time of 4.8 min. Derivatised (S)-5-hydroxynorvaline 4 also appeared as a single peak at 4.4 min. 994 M. García et al. / Tetrahedron: Asymmetry 11 (2000) 991–994 Acknowledgements Support for this research was provided by the CIRIT (Generalitat de Catalunya) through grants QFN95-4703 and 1997SGR-00075, and by the DGICYT (Ministerio de Educación y Cultura, Spain) through grants PB97-0976 and 2FD97-0293. References 1. 2. 3. 4. 5. 6. 7. 8. Thomson, J. F.; Moris, C. J.; Hunt, G. E. J. Biol. Chem. 1964, 239, 1122–1125. Strack, D.; Schmitt, D.; Reznik, H.; Boland, W.; Grotjahn, L.; Wray, V. Phytochemistry 1987, 26, 2285–2287. Hill, R. E. J. Nat. Prod. 1993, 56, 1246–1254. Ariza, J.; Díaz, M.; Font, J.; Ortuño, R. Tetrahedron 1993, 49, 1315–1326. Ayala, A.; Cutler, R. G. Free Radical Biol. Med. 1996, 121, 65–80. Mikol, V.; Papageorgiou, C.; Borer, X. J. Med. Chem. 1995, 38, 3361–3367. Judice, J. K.; Gamble, T. R.; Murphy, E. C.; de Vos, A. M.; Schultz, P. G. Science 1993, 261, 1578–1581. Broddefalk, J.; Bäcklund, J.; Almqvist, F.; Johansson, M.; Holmdahl, R.; Kihlberg, J. J. J. Am. Chem. Soc. 1998, 120, 7676–7683. 9. Estiarte, M. A.; de Souza, M. V. N.; del Río, X.; Dodd, R. H.; Rubiralta, M.; Diez, A. Tetrahedron 1999, 55, 10 173–10 186. 10. Piró, J.; Rubiralta, M.; Giralt, E.; Diez, A. Tetrahedron Lett. 1999, 40, 4865–4868. 11. Rodríguez, R.; Diez, A.; Rubiralta, M.; Giralt, E. Heterocycles 1996, 43, 513–517. 12. Murphy, W. H.; Gottschalk, A. Biochim. Biophys. Acta 1961, 52, 349. 13. Barlos, K.; Maomos, P.; Papaioannou, D; Patrianakou, E. J. Chem. Soc., Chem. Commun. 1987, 1583–1584. 14. Nefkens, G. H. L.; Zwanenburg, B. Tetrahedron 1983, 39, 2995–2998. 15. Gong, B.; Lynn, D. G. J. Org. Chem. 1990, 55, 4763–4765. 16. Nin, A. P.; Lederkremer, R. M.; Varela, O. Tetrahedron 1996, 52, 12 911–12 918. 17. Ishida, K.; Matsuda, H.; Murakami, M. Tetrahedron 1998, 54, 5545–5556. 18. Coudert, E.; Acher, F.; Azerad, R. Synthesis 1997, 8, 863–865.