Preparation and properties of totally synthetic immunogens David C. Jackson*, Catherine J. Fitzmaurice, Lorena E. Brown, Weiguang Zeng Cooperative Research Centre for Vaccine Technology, Department of Microbiology and Immunology, The University of Melbourne, Parkville 3052, Victoria, Australia Abstract Although the induction of antibodies and T cells by synthetic peptides representing de®ned antigenic determinants is a routine laboratory procedure, their use as vaccines has not yet been generally realised. There are a number of reasons for this and paramount is the limitation of valency which aects not only immunogenicity, but also the coverage of the antigenic universe. This paper is a review of our own work in which we have assembled synthetic peptides into multivalent arti®cial proteins and then examined their immunological properties. # 1999 Elsevier Science Ltd. All rights reserved. Keywords: Synthetic peptides; Vaccines; Arti®cial proteins 1. Introduction The prospect of using de®ned synthetic vaccines has encouraged many groups throughout the world to identify important epitopes within their antigen of interest and devise ways of incorporating these into an immunogenic structure which can then be used for vaccination. Despite a very strong movement in this direction and the demand for new and improved vac- cines, there are still few, if any, synthetic peptide-based vaccines available commercially. This is perhaps sur- prising when we consider that 35 years ago it was shown [1] that antibodies raised against fragments of tobacco mosaic virus elicited antibodies that neutral- ised the whole virus. Clearly protein fragments can be eective, but the generic use of synthetic peptides in vaccines has not yet been realised. In the case of vaccines used to elicit antibodies there are a number of contributing reasons for this, among which are the facts that: (i) the peptide sequences used to represent epitopes are generally short and as a consequence contain insucient infor- mation to fold into the correct shape necessary to mimic conformationally-dependent epitopes and (ii) even if the peptide is recognised by a B cell, the B cell must still receive help from a helper T cell recog- nising a sequence within the same immunogen. Traditionally the site of T-cell recognition has been provided by a carrier protein to which peptides are covalently coupled, but this coupling procedure can introduce other problems such as modi®cation of the antigenic determinant during the coupling process and the induction of antibodies against the carrier at the expense of antibodies which are directed toward the peptide [2±4]. Furthermore, the use of irrelevant proteins in the preparation introduces issues of qual- ity control. These problems have of course been recognised and a number of ways of circumventing them have been attempted, ranging from assembly of linear [5,6] or branched [7] constructs comprised of covalently attached T- and B-cell epitopes, the introduction of the multiple antigenic peptide (MAP) approach [8,9], chemical ligation of epitopes [10±12] and controlled polymerisation of synthetic peptides [13±15]. We have tried each of these approaches in this laboratory and review the results of our experiences here. Vaccine 18 (2000) 355±361 0264-410X/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. PII: S0264-410X(99)00205-4 www.elsevier.com/locate/vaccine * Corresponding author. Tel.: +61-3-9342-9819; fax: +61-3-9342- 9782. E-mail address: d.jackson@microbiology.unimelb.edu.au (D.C. Jackson)