|Home | About | Journals | Submit | Contact Us | Français|
The synthesis of an enantiopure isoxazolidine monomer for the incorporation of β3-apartic acid residues into β3-oligopeptides via chemoselective α-ketoacid–hydroxylamine amide formation. This route involves nitrone cycloaddition of 3-thiophenylpropanal and circumvents limitations of other potential starting materials
We have recently reported a highly chemoselective amide-bond forming reaction between α-ketoacids and hydroxylamines. i In addition to its potential for the fragment coupling of unprotected α-peptide fragments, we have explored its application to the iterative, aqueous synthesis of oligo-β3-peptides, ii an exciting class of peptidomimetics of contemporary interest in medicinal and bioorganic chemistry.iii
Our novel approach to the synthesis of these oligopeptides takes advantage of the chemoselective coupling of a growing β-peptide chain with chiral isoxazolidines. The amide formation results in N–O bond cleavage and the formation of an α-ketoester. Following hydrolysis, the chain is poised for iterative elongation (Scheme 1). The key amide forming reaction requires no reagents, produces only carbon dioxide and methanol as byproducts (Scheme 2), and operates in the presence of unprotected functional groups including amines and carboxylic acids. As such, it offers great potential as an alternative route to the preparation of β-oligopeptides that circumvents challenges of the established methods including difficult couplings and deprotections in longer β-peptide chains and improved access to the requisite β-peptide monomers.
The major challenge to the widespread adoption of this method for β3-peptide synthesis is the need to prepare the isoxazolidine monomers as single enantiomers. We have reported the use of Vasella’s mannose-derived nitronesiv as chiral auxiliaries for their asymmetric synthesis via 1,3-dipolar cycloaddition with 2-methoxymethacrylate (Scheme 3). Following purification of the cycloadducts and removal of the auxiliary, the desired isoxazolidine monomers are obtained, in most cases, in enantiomerically pure form. This approach has proven successful for the synthesis of isoxazolidine monomers bearing most of the common amino acid side-chains, including those found in leucine, valine, phenylalanine, lysine, glutamic acid, glycine, and numerous others. Unfortunately, this procedure fails for the most logical synthetic approaches to the momoner containing the side chain found in aspartic acid. This is particularlewy disappointing as some of the most exciting advances in the properties of longer β3-oligopeptides are with sequences rich in this amino acid residue.v
In this communication, we document a versatile workaround to the synthesis of an enantiopure isoxazolidine monomer 3 containing the aspartic acid side chain. This approach is high yielding, readily executed on a preparative scale, and offers an entry into isoxazolidine monomers bearing functionality poised for further elaboration.
Our studies began by considering the most direct methods to access the aspartic acid derived monomer 3. In our previous synthesis of the glutamic acid side chain, which is a one carbon homologue of our target, we successfully employed an omega-tert-butylester aldehyde in the nitrone cycloaddition. In the case of aspartic acid, however, the necessary aldehyde was difficult to prepare and employ due to its propensity to adopt the enol form. (Scheme 4). The obvious alternatives, protected-β-hydroxyaldehydes, could be prepared by underwent elimination or other unproductive pathways during the attempted cycloaddition step. Brief attempts to employ acrolein as a starting aldehyde were not initially successful.
In contrast to the β-oxo-substituted counterparts, we were pleased to find that β-mercaptoaldehydes were easily prepared vi and suitable for use in the chiral auxiliary directed cycloaddition (Scheme 5). Importantly, cycloadduct 6 could be obtained as a single enantiomer following purification by column chromatography and recrystallization. Although the overall yield for the cycloaddition and subsequent removal of the minor diastereomer was modest, the two-step sequence from acrolein could be easily executed on a preparative scale.
In order to convert the sulfide of 6 to the desired carboxylic acid, we effected a Pummerer oxidation by oxidation of the sulfide to the sulfoxide followed by treatment with trifluoroacetic anhydride and workup in the presence of mercury chloride to afford aldehyde 8 in 95% overall yield.vii This aldehyde is, in itself, a valuable scaffold for the preparation of unnatural β-amino acid side chains. It can also be readily oxidized to carboxylic acid 9 under standard conditions and in excellent yield.viii
Deprotection of the sugar can be pursued at this stage of the synthesis, but we found it advantageous in terms of handling and purity of the final product to first protect the carboxylic acid side chain. A number of standard protocols for tert-butyl ester formation failed, but we found success with tert-butyl N,N-diisopropylisourea. ix Removal of the chiral auxiliary was best effected with aqueous hydrazine to afford 11.
Although 11 serves a suitable monomer for β3-oligopeptide synthesis, we have often found it advantageous to employ the side chain unprotected monomers for our ongoing studies on the application of these reagents for aqueous peptide synthesis. The tert-butyl group of 11 can be easily removed by treatment with trifluoroacetic acid, providing TFA salt 3, which can be used directly in the amide forming reaction with α-ketoacids.
1H and 13C NMR analyses of intermediates 6 and 10 showed only a single diastereomer, suggesting that the final isoxazolidine would be obtained as a single enantiomer. To confirm and better quantify the enantiopurity, we first subjected 11 to amide formation with phenypyruvic acid in 1:1 tBuOH/pH 7.4 buffer to afford α-ketoester 14 (Scheme 6). Analysis of this material by HPLC on a chiral column established that 11, and by analogy 10 and 3, were obtained in >99% ee.
In summary, we have disclosed an effective synthetic route to isoxazolidine monomer 3, which allows incorporation of a β3-aspartic acid residue into a growing peptide chain via the ketoacid-hydroxylamine amide ligation. In addition to provding access to an important monomer, this route will allow access to other important monomers including asparagine, methionine, and unnatural side chains derived from 3 or its synthetic intermediates.
This work was supported by the National Institutes of Health (R01GM076320 and F31GM078854) and the Arnold and Mabel Beckman Foundation. We are grateful to Kyowa Hakko for support of H.I. Acrylate 2 was kindly prepared by Bioblocks, Inc. J.W.B is a fellow of the Packard Foundation, Sloan Foundation, and a Research Corporation Cottrell Scholar. We appreciate insightful advice and synthetic protocols from Justin Russak.
Experimental procedures, characterization data, and 1H and 13C NMR spectra for all new compounds.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.