Elsevier

Tetrahedron

Volume 57, Issue 30, 23 July 2001, Pages 6531-6538
Tetrahedron

New design concepts for constraining glycosylated amino acids

https://doi.org/10.1016/S0040-4020(01)00545-2Get rights and content

Abstract

In an effort to probe the mechanism by which glycosyltransferases recognize glycoproteins and assemble the core structures of O-linked oligosaccharides, two constrained glycopeptides based on the α-N-acetylgalactosaminyl serine substructure were chosen for synthesis. These compounds representing one of the two possible gauche conformations of the d-serine and l-serine configurations of the parent substructure were successfully prepared in reasonable overall yield through a convergent strategy.

Introduction

Glycosylation is a ubiquitous posttranslational modification of proteins and is associated with a number of processes both within cells and at cell surfaces. These include protein transport, cell adhesion, and signal transduction.1 Aberrant glycosylation of cellular proteins and glycolipids is associated with various diseases, including cancerous and inflammatory conditions.2 Therefore, an understanding on a molecular level of the structural features of glycoproteins that are recognized by various enzymes and receptors would be valuable in developing inhibitor-based strategies to control carbohydrate-mediated cellular processes. This fundamental understanding could, in turn, lead to new therapeutic strategies for conditions that are characterized by abnormal glycosylation.

The ground state conformations of small glycopeptide fragments have been studied in some detail by NMR spectroscopy;3 however, relatively little is known about the conformations that are recognized by various enzymes and receptors. The preferred torsional angle about the exocyclic C1–O bond (φ) of the carbohydrate moiety in the ground state is somewhat rigid, as predicted by the exoanomeric effect.4 However, other torsional angles within the side chain, including χ2, are somewhat more flexible, as illustrated in the α-N-acetylgalactosaminyl serine substructure 1.

Significant research has gone into designing conformational constraints into various amino acid and peptide motifs to probe bioactive conformations.5 However, conformationally restricted glycopeptides are relatively unexplored.6

In an effort to probe the mechanism by which glycosyltransferases recognize glycoproteins and assemble the core structures of O-linked oligosaccharides, a series of conformationally constrained glycopeptides was chosen for synthesis. These particular targets were identified in an effort to represent the accessible low energy conformations about χ2 of the native structure 1, and will ultimately be incorporated into small peptide fragments for biological evaluation. Among those chosen was compound 2 in which the α-carbon (Cα) of the amino acid component of the structure is incorporated into a six-membered ring, and the χ2 torsional angle is effectively constrained in one of the two possible gauche conformations (Fig. 1). Additionally, the C1–O bond is locked into the low-energy exoanomeric orientation. The unnatural d-serine configuration is represented by the similarly constrained analog 3.

The synthesis strategy for 2 (Scheme 1) was designed with the focus of preferentially establishing the desired l-serine configuration at Cα as well as the necessary anomeric orientation at the spiroketal center. Specifically, the desired anomeric configuration could arise from a thermodynamic spiroketalization event,7 and the stereochemistry at Cα could be established through diastereomeric preferences inherent within this same process. In a general sense, the synthesis would commence with a convergent nucleophilic addition of appropriately protected versions of the carbohydrate (4) and amino acid (5) components of the target structure. Manipulations to the resulting product 6, including reduction of the internal alkyne, would provide the spiroketal precursor 7. Selective spiroketalization of one of the diastereotopic primary hydroxyl groups and subsequent oxidation of the other, followed by protecting group manipulations, should provide the analogs 2 and 3. The unnatural d-serine analog 3 could also prove useful in our studies.

Assuming thermodynamic conditions, the spiroketalization was predicted to afford an excess of the desired l-diastereomer, 2. This prediction was based on the presumed inherent preference of an amine bearing a bulky protecting group, such as a tert-butyl carbamate, over a hydroxymethyl, to reside in the pseudo-equatorial position. This preference is illustrated by the partial half-chair structures 8 and 9 (Fig. 2).

Section snippets

Results and discussion

The synthesis of the nucleophilic amino acid component 14 was accomplished in an efficient, straightforward manner (Scheme 2). The synthesis commenced with protection of the amine group of 108 to give the tert-butyl carbamate 11 in good yield. Swern oxidation of the free primary alcohol in 11 gave the aldehyde 12. A modified three-step alkynylation procedure adopted from a recently reported procedure9 was applied to 12 to give the alkyne 13 in good overall yield. The acetal 14 was then obtained

Conclusion

In conclusion, two constrained glycopeptides were successfully synthesized in reasonable overall yield through a convergent strategy. The compounds 21a and 21b, representing the d-serine and l-serine configurations, respectively, were fully characterized and are poised for incorporation into peptides for subsequent biological evaluation and structural studies. These compounds illustrate a novel concept for constraining torsional angles in glycopeptides. Studies toward establishing a more

General methods

Reagents obtained from commercial suppliers were used without further purification unless otherwise noted. All solvents were purified and dried by standard distillation methods. 1H and 13C NMR spectra are reported in parts per million (δ) relative to CHCl3 (7.24 and 77.23 ppm, respectively) as the internal standard.

Acknowledgements

This research was supported by an NSF Career Award (CHE-9733765). R. L. H. also thanks the Camille and Henry Dreyfus Foundation (Camille Dreyfus Teacher–Scholar Award), Pfizer (Junior Faculty Award), and Novartis (Young Investigator Award) for support. This work was greatly facilitated by a 500 MHz NMR spectrometer that was purchased partly with funds from an NSF Shared Instrumentation Grant (CHE-9523034). Bob Barkley (University of Colorado) is thanked for obtaining mass spectra. Bruce Noll

References (12)

  • W.F DegradoV.J Hruby et al.

    Biochem. J.

    (1990)
    C Toniolo

    Int. J. Peptide Protein Res.

    (1990)
    V.J Hruby et al.D Obrecht et al.

    Adv. Med. Chem.

    (1999)
    M Goodman et al.

    Pure Appl. Chem.

    (1996)
  • D.C Forbes et al.

    Synthesis

    (1998)
  • R Kornfeld et al.

    Annu. Rev. Biochem.

    (1985)
    T.W Rademacher et al.

    Annu. Rev. Biochem.

    (1988)
    A Varki

    Glycobiology

    (1993)
    R.A Dwek

    Chem. Rev.

    (1996)
  • S.-I Hakomori

    Advances Cancer Res.

    (1989)
  • L Kirnarsky et al.

    Biochemistry

    (2000)
    W Wu et al.

    J. Am. Chem. Soc.

    (1999)
    S.E O'Connor et al.

    Chem. Biol.

    (1998)
    R Liang et al.

    J. Am. Chem. Soc.

    (1995)
    B Meyer

    Top. Curr. Chem.

    (1990)
  • E Juarista et al.

    Tetrahedron

    (1992)
    P Deslongchamps

    Stereoelectronic Effects in Organic Chemistry

    (1983)
There are more references available in the full text version of this article.

Cited by (15)

  • Synthesis of novel therapeutic agents for the treatment of multiple sclerosis: A brief overview

    2013, European Journal of Medicinal Chemistry
    Citation Excerpt :

    Synthesis commenced with the protection of diol 33 as its acetonide ether and amine as its Boc derivative followed by Swern oxidation to furnish known aldehyde 34 [109]. Aldehyde 34 was converted to alkyne 35 [110] using one pot protocol developed by Roth [111]. Alkyne intermediate 35 on reaction with 4-octyl iodobenzene [112] under Sonogashira coupling conditions [113] using Pd(PPh3)4 gave compound 36.

View all citing articles on Scopus
View full text