Carbohydrate Chemistry Chemical and Biological Approaches Volume 36

A Review of the Literature Published between January 2009 and December 2009

By Amelia Pilar Rauter, Thisbe K. Lindhorst

The Royal Society of Chemistry

Copyright © 2010 The Royal Society of Chemistry
All rights reserved.
ISBN: 978-1-84755-044-6

Contents

Preface Amelia Pilar Rauter and Thisbe K. Lindhorst, v,
Synthetic vaccines based on N-and O-glycopeptides–molecular tools for immunotherapy and diagnostics Ulrika Westerlind and Horst Kunz, 1,
Mycobacterial lipoarabinomannan fragments as haptens for potential anti-tuberculosis vaccines Pui-Hang Tam and Todd L. Lowary, 38,
α-Galactosylceramides and analogues – important immunomodulators for use as vaccine adjuvants Niamh Murphy, Xiangming Zhu and Richard R. Schmidt, 64,
Synthetic glycoconjugates based on Leishmania lipophosphoglycan structures as potential anti-leishmaniasis vaccines Andrei V. Nikolaev, Nawaf Al-Maharik and Olga V. Sizova, 101,
Solution- and solid-phase synthesis of oligosaccharides Steffen Eller, Markus Weishaupt and Peter H. Seeberger, 127,
Advances in chemoenzymatic synthesis of glycopeptides for cancer research applications Celso A. Reis, 142,
Bioorthogonal chemical reporter methodology for visualization, isolation and analysis of glycoconjugates Geert-Jan Boons, 152,
Azido leaving group in enzymatic synthesis-small and efficient Pavla Bojarová and Vladimír Kren, 168,
Recent advances in the synthesis of functionalized carbohydrate azides Zbigniew J. Witczak, 176,

CHAPTER 1

Synthetic vaccines based on N– and O-glycopeptides–molecular tools for immunotherapy and diagnostics

Ulrika Westerlind and Horst Kunz

DOI: 10.1039/9781849730891-00001

This chapter summarizes available methods for the preparation of synthetic vaccines based on glycopeptides and recent advances in this field. It further includes results of their immunological evaluation. Syntheses of glycopeptides of defined chemical structure and conjugation of these compounds to a carrier protein or an immunostimulant are of interest for the development of new immunotherapeutics and/or antibody-based diagnostics. Since a number of years, the aberrant glycosylation of the tumorassociated mucin MUC1 forming tumor specific epitopes on the epithelial cell surface has been considered an attractive research target for the preparation of such vaccines. Examples of synthetic vaccines directed against the O-glycosylated MUC1 tandem repeats will here be given including synthetic MUC1 glycopeptides conjugated to a T-cell epitope peptide, to a carrier protein, to a lipid immunostimulant or the multimeric presentation of glycopeptides on dendrimers. Other attractive targets for immunotherapy are the viral envelope proteins HIV gp120 and HIV gp41, which are highly glycosylated with highmannose and complex type N-glycans. Examples will be given, which illustrate syntheses of high-mannose HIV gp120 or gp41 glycopeptides with the natural peptide backbone or with a nonnatural cyclic backbone to mimic the high-mannose cluster domain of HIV gp120. In addition the synthesis and immunological evaluation of a vaccine will be described, which contains the highmannose cluster mimotope glycopeptide conjugated to an outer membrane protein complex (OMPC) as the carrier.

1. Introduction

In multi-cellular organisms, many proteins are co- or posttranslationally modified by monoor oligosaccharides. It has become evident that these glycoproteins play important roles in diverse biochemical processes. The saccharides influence the physiochemical properties, which includes conformational effects on the protein, the stability against proteolysis, or the lubrication of cells. In addition, the glycans are involved in cell-cell recognition and cell-external agent interactions. These interactions induce biological events, which are involved in cell-growth and differentiation, cell-proliferation, cell-adhesion, binding of pathogens, fertilization and immune responses. The glycans assist in protein folding and transport, they are involved in pathogenic processes like chronic inflammation, viral and bacterial infections, tumor growth and metastasis11, and autoimmune disorders. The variety of events affected by protein glycosylation is not surprising. Carbohydrates are unique in their complexity of structure. In contrast to oligonucleotides and proteins, they can be connected in more than one form: connected via different configural positions of the glycan residue and also via an anomeric α or β glycosidic linkage. The different modes of connections can lead to long linear or highly branched saccharide structures. By combining different types of glycans like galactose, glucose, N-acetylgalactosamine, N-acetylglucosamine, fucose, mannose, sialic acid and others, an enormous variety of combinations can be formed. Nature is only using a small portion of these possible combinations. The biosyntheses of different oligosaccharides are dependent on glycosyltransferases which are gene-encoded. Each type of connection and glycan residue needs a specific enzyme. Of practical reasons this restricts the number of combinations. Nature also limits the number of final products that can be made by having common core structures. The saccharides can be linked to the protein backbone via an O-glycoside bond or via N-glycosyl amide bond in N-glycoproteins. In the most common type of O-glycoproteins, the mucin type, a GalNAc residue is connected to the protein backbone via an O-glycosidic linkage to serine or threonine. The GalNAc residue can be further extended with additional Gal, GlcNAc or GalNAc residues forming the core structures 1–8. These core structures can then be extended with one or repeated type 1 (Gal β1-3 GlcNAc) or type 2 (Gal β1-4 GlcNAc) glycosylation and terminated with fucose, sialic acid and/or GalNAc residues leading to complex saccharides (Fig. 1a, b).

The majority of N-glycoproteins have a common pentasaccharide core consisting of three mannoses and two GlcNAc residues, one of the GlcNAc residues is linked to the protein core via an amide bond of an aspargine side chain. The pentasaccharide core structure can be additionally extended forming high mannose type, hybrid type or complex type N-glycoproteins (Fig. 1c, d). In spite of the restrictions made by nature when forming complex oligosaccharide structures, a large amount of combinations can be formed, and since the majority of glycans are attached to other molecules like proteins or lipids, an additional dimension of complexity results. To understand the function and relevance of saccharides, it is important to also consider the backbone to which it is connected. In case of glycoproteins, it is often the effect of glycosylation on the protein backbone that is interesting to study in terms of conformational or proteolytic stability, structure recognition, binding or multimeric effects. Since glycans are not direct gene products, they are formed in an environment of competing glycosyltransferases and glycosidases with variations in substrate and donor levels. This results in the microheterogeneity of the glyoproteins.

Therefore, it is problematic to study glycosylation only by gene expression studies on enzyme levels or by investigations of isolated glycoproteins. It is also difficult to isolate pure glycoproteins in sufficient quantities. By synthetic methods, glycopeptides with a defined structure can be produced. These peptides can be employed to study conformational effects of glycosylation, binding events to cell-adhesion molecules or pathogens or for the construction of vaccines or the production of inhibitors mainly in form of glycopeptide mimics. Synthetic glycopeptides can be ligated to proteins forming homogeneously glycosylated proteins, which might be useful for studying glycosylation and for applications in therapy. In this chapter, recent advances in the synthesis of vaccines based on glycopeptides and their immunological evaluation will be described. So far, the aberrant glycosylation of mucin (MUC1) glycoproteins has been the major research target concerning the development of glycopeptide vaccines. N-Glycosylated epitopes of the HIV-1 envelope glycoproteins gp120 and gp41 are other molecular targets, that have been explored during recent years. Syntheses of glycopeptide vaccines employing various conjugation strategies to different immunostimulants will also be discussed. A specific immune response generated from such vaccines could be valuable in immunotherapy through vaccine or humanised monoclonal antibody treatment. In addition, generated monoclonal antibodies from synthetic vaccines with a high binding specificity would be precious tools in immunodiagnosis.

2. Synthesis of tumor-associated glycopeptides and glycopeptide vaccines

In 1984, Springer published that glycoproteins of the outer cell-membrane of epithelial tumor cells show altered glycosylation consisting of the T- and its precursor TN antigen structure. It was also found that monoclonal antibodies generated from these tumor-associated structures showed cross-reactivity to the desialylated form of glycophorin A, i.e. asialoglycophorin A. From these studies it could be concluded, that these T- and TN-glycoproteins on the epithelial tumor cells must be structurally related to asialoglycophorin (Fig. 2a). Glycophorin A is the major sialoglycoprotein on erythrocytes. In the N-terminal domain it contains cryptic Tantigen structures glycosylated with sialic acid at the 3-position of the galactose and at the 6-position of the GalNac residue.

The glycophorin exists in two blood group specificities, M and N, with identical glycoforms, but different in two of the totally 131 amino acids, one of those is the N-terminal residue which is a serine in blood group M, but a leucine in blood group N. These studies initiated syntheses of glycopeptide vaccines targeting the N-terminal region of asialoglycophorin A (Fig. 2b). The antibodies generated from a synthetic glycopeptide vaccine based on the N-terminal region of asialoglycophorin A of blood group M were specific to the T-antigen glycan structure and also to the peptide backbone structure differentiating between blood group M and N. This result suggested that both the glycan and peptide backbone structure are relevant for the antibody binding specificity. The generated antibodies showed affinity to epithelial tumor tissues, but also to normal tissues. It could be concluded that the synthesized T-antigen glycopeptide structure is tumor-associated, but not sufficiently tumor-selective. In this context, more recent investigations showed that mucin MUC1 expressed on epithelial cells is a tumor-associated glycoprotein.

Mucins constitute a class of extensively glycosylated proteins expressed on the surface of epithelial cells or secreted to function in mucus. They normally carry complex and highly branched O-linked carbohydrate structures that shield the protein core. The membrane bound glycoprotein MUC1 is the most intensively studied mucin protein with regard to cancer immunotherapy. The extracellular domain has a structural feature common to all mucins. It contains a region of a variable number of tandem repeats (VNTR). The repeats have identical or very similar amino acid sequences, which are rich in proline, serine and threonine. In MUC1 from different sources, variable numbers (20–125) of the tandem repeat were found. The tandem repeat consists of 20 amino acids of the sequence HGVTSAPDTRPAPGSTAPPA which includes five potential O-glycosylation sites (underlined). On epithelial tumor cells, MUC1 is extensively over expressed and its glycosylation pattern has distinctly been altered. Concomitant down-regulation of glycosyl transferases, in particular the core 2 β-1,6-N-acetylglucosaminyltransferase, and up-regulation of sialyltransferases result in short saccharides often prematurely sialylated. TN, T, Sialyl-TN, (2,3)-Sialyl-T and (2,6)-Sialyl-T structures constitute important examples of such tumor-associated saccharide antigens (Fig. 3). Furthermore, the underglycosylation of the MUC1 extracellular domain also results in the exposure of the peptide backbone. Tumor-associated epitopes consisting of both the saccharide and the peptide structures are formed.

By an efficient induction of antibodies specific to the tumor-associated MUC1 antigens, it should be possible to break the tolerance of the immune system against these structures. In order to produce such a humoral immune response, a naïve B cell needs to be stimulated to proliferate and to differentiate into an antibody-secreting plasma cell. To this end, additional stimulation by activated CD4+ T helper cells is required. The naïve T helper (TH) cells are activated when their T cell receptor (TCR) binds to the peptide epitope presented by the major histocompatibility complex II (MHC II) on the surface of a antigen-presenting cell (APC). The peptide epitope presented by the APCs had to be generated by proteolytic fragmentation of the extracellular antigen that was internalized via recognition of the B-cell epitope. The protolytically cleaved fragments can then either bind to the MHC class I proteins initiating cytotoxic T-cell activation or bind to the MHC class II proteins generating T helper cell activation.

Efforts have been made to synthesize MUC1 tandem repeat glycopeptides that mimic the abberant glycosylated cell surface glycoprotein. These peptides are of interest to develop synthetic vaccines and also to generate monoclonal antibodies for immunotherapy and as diagnostic tools. The tumor-selective MUC1 glycopeptides are only moderately immunogenic, and additional stimulation using a carrier protein, e.g. KLH, or an immunostimulating TH-cell peptide epitope, as for example, a peptide sequence of ovalbumin, is necessary to generate a humoral immune response. During immunization it is also necessary to stimulate the innate immune system, this is done by adding an adjuvant together with the vaccine, for example Freund’s adjuvant, or using an adjuvant built into the vaccine, for example Pam3Cys which is a ligand to the Toll-like receptors. Examples of synthetic vaccines based on the tumor associated MUC1 glycopeptide epitope conjugated with various immunostimmulants will here be given.

2.1 Synthesis of TN, T, sialyl TN, sialyl T glycosyl amino acid building blocks

Chemical synthesis of glycosylated Fmoc-serine and threonine building blocks for solid phase peptide synthesis still is a demanding task. Most such building blocks are back integrated to the TN-antigen intermediate 12. By making use of the high reactivity of the 6- and 3-positions of Gal in combination with proper protecting groups, highly selective glycosylations can be performed. TN-antigen was synthesized from acetobromogalactose 1 via the corresponding galactal 2 and its azidonitration yielding an anomeric mixture of azidonitrate 3, which was subsequently converted to the corresponding α-bromide 4 by treatment with LiBr in acetonitrile. Fmoc-threonine tert-butyl ester, or its serine counterpart (not shown) was then glycosylated with 4 under Koenigs-Knorr conditions providing 5. The azido function was then converted to the corresponding acetamide 6 by thioacetic acid, and the tert-butyl ester was cleaved by TFA in presence of anisole as a scavenger, providing the fully protected TN-antigen building block 7 for further solid phase peptide synthesis (SPS) (Scheme 1).

Synthesis of 2,6-Sialyl TN antigen begins with the preparation of the sialyl donor 11. It has been proven in a number of syntheses that glycosylation reactions with xanthate 11 in the presence of a thiophilic activator are superior to the use of other sialic acid donors. Peracetylation of sialic acid 8 gave the fully acetylated intermediate 9, which was benzylated with benzyl bromide/cesium carbonate to give 10 as an anomeric mixture. Treatment of 10 with acetyl chloride followed by treatment with KS(CS)OEt under thermodynamically controlled conditions afforded xanthate 11 as a single anomer. After careful deacetylation of Fmoc-TN antigen-threonine 6, the resulting product 12 was sialylated selectively in 6-position using xanthate 11 and MeSOTf in an acetonitrile containing solvent at low temperature. At low temperature (–65 1C), only the kinetically favored equatorial product is formed due to assistance of the nitrile solvent. Further transformations include peracetylation of 13 and cleavage of the tert-butyl ester 14 to give the Fmoc-glycosyl amino acid-building block 15 (Scheme 2).

(Continues…)Excerpted from Carbohydrate Chemistry Chemical and Biological Approaches Volume 36 by Amelia Pilar Rauter, Thisbe K. Lindhorst. Copyright © 2010 The Royal Society of Chemistry. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.