Identification of antigenically active tryptic fragments of apical membrane antigen-1 (AMA1) of Plasmodium chabaudi malaria: strategies for assembly of immunologically active peptides

Abstract

Apical membrane antigen-1 (AMA1) is a prime vaccine candidate for inclusion in a vaccine against malaria. It is known that the disulphide bond stabilised conformation of this antigen is important for eliciting a protective antibody response, however little is known about the epitopes within this molecule that are targeted by the immune response. We have used a peptide approach for the identification and characterisation of such regions. In this study, the in vitro refolded, recombinant ectodomain of AMA1 from the D strain of Plasmodium chabaudi adami, was digested with trypsin and individual peptide fragments examined for antigenic activity. We found that a tryptic fragment, which was derived from a loop-like structure within the putative domain I of the intact AMA1 molecule, was highly reactive with antibodies from the sera of hyperimmune mice. Two different synthetic peptide constructs incorporating this antigenically active fragment were assembled. The first consisted of two separate peptide chains which were linked through a disulphide bond formed using chemo-selective chemistry. A larger 45-mer loop peptide, generated by the oxidation of two cysteine residues close to the N- and C-termini of the 45-mer, represented the complete loop structure and incorporated the tryptic fragment. Each peptide construct was also able to elicit production of high titres of antibodies in mice and furthermore, the 45-residue loop peptide elicited antibodies capable of binding to AMA1 with titres comparable to those present in a mouse which had recovered from multiple exposures to P. chabaudi adami parasites. Passive immunisation with anti-loop antibodies did not suppress the development of parasitaemia in mice challenged with P. chabaudi adami suggesting that although highly immunogenic, the peptides represented inadequate or inappropriate epitopes for vaccination purposes.

Introduction

Malaria continues to be one of the major health problems in many tropical countries with potentially one-third of the world’s population at risk from infection. In some countries, the incidence of disease and mortality is increasing; important contributing factors being the appearance of insecticide-resistant mosquito vectors and drug-resistant parasites. Because of this the development of an effective vaccine for malaria is a priority.

Apical membrane antigen-1 (AMA1) is one of several merozoite surface antigens rated highly as potential vaccine candidates against malaria. This protein can be detected in blood-stage parasites just prior to schizont rupture. A mature form of AMA1 can be detected in the rhoptries of merozoites and about the time of schizogony a smaller N-terminally processed form spreads circumferentially around the merozoite surface [1]. Although AMA1 is expressed only for a narrow time interval around the time of schizogony it is a target of the protective host immune response. Furthermore, both native and recombinant forms of the antigen have been able to induce various levels of protection against challenge from Plasmodium parasites in various animal models [2], [3], [4], [5], [6]. The vaccine potential of the Plasmodium falciparum AMA1 ectodomain is currently in the early stages of clinical assessment.

The primary structure of AMA1, found in various Plasmodium species, contains 16 conserved cysteine residues [7], [8], [9] which form eight intra-molecular disulphide bonds [10]. The protective immune response against AMA1 is largely directed toward disulphide bond-dependent conformational epitopes as the reduced and alkylated antigen gives poor protection against parasite challenge in vaccine trials [5] and is poorly recognised by antibodies from the serum of hyperimmune individuals who live in malaria endemic regions [11]. Using antibodies from hyperimmune mice we were interested in identifying and then characterising the vaccine potential of disulphide-linked peptide fragments from trypsin-digested recombinant P. chabaudi adami AMA1.

An antigenically active fragment was identified and found to represent part of a loop region, closed by a disulphide bond, of a putative sub-domain of AMA1. Two synthetic peptide-based constructs representing this antigenically active tryptic fragment were assembled. One of the constructs was made using two individually synthesised peptide chains linked through cysteine residues to yield a peptide heterodimer. The second construct was synthesised as a linear 45 amino acid residue peptide which was then folded by in vitro oxidation of the two cysteine residues to yield a closed loop. In this way, both constructs were assembled to incorporate a disulphide bond of the same assignment as that present within the native sequence. The two peptide constructs were examined for immunogenic activity and vaccine potential by inoculation into mice.