Characterization of immunoreactive octapeptides of human‐cytomegalovirus gp58

Abstract

We have mapped continuous epitopes, for positions 591 -673 of the human cytomegalovirus 58-kDa glycoprotein using overlapping synthetic peptides and human sera. This region contains a fragment previously described as including the dominant site for induction of human-cytomegalovirus antibodies. Since the selected sequence is highly conserved among herpes viruses, we have considered the possible presence of antigenic cross-reactivity, particularly with the Epstein-Barr virus. Several peptides in the studied region were antigenic and two main continuous epitopes have been identified. Serological cross-reactions observed with Epstein-Barr virus are discussed, focussing on the possible implications of structural features and sequence similarity between humancytomegalovirus and Epstein-Barr-virus glycoproteins.Human-cytomegalovirus (HCMV) infections play an important role as a source of human disease; although primary infection is usually mild or asymptomatic the virus is a major pathogen in immunocompromised individuals and a common cause of congenital defects [l]. A more detailed understanding of the antigenic features involved in the human immune response against HCMV would allow for better control and detection of this infection in susceptible hosts.The HCMV genome codes for 35 -45 structural proteins, most of which are immunogenic [2-41. Outer-envelope glycoproteins, in particular, seem to be major targets for viral neutralization. In several reports, monoclonal antibodies have been described that react with HCMV glycoproteins [5-71, and individually purified or expressed glycoproteins have also been shown to elicit neutralizing responses in experimental animals [8-111. HCMV glycoproteins have been recently grouped into three distinct families, designated gCI-gCIII [ 121. gCI is the major envelope component, consisting of two disulphidelinked protein species of estimated molecular mass 55-58 kDa and 100-116 kDa [13]; these mature forms are generated by proteolytic cleavage of a high-molecular-mass precursor, which is both co-translationally and post-translationally glycosylated [ 14-161.The gene encoding the 58-kDa glycoprotein (gp58) has been mapped and sequenced [ 10, 171 ; the translational product shows sequence similarity with other herpes-virus gly- coproteins, namely the Epstein-Barr-virus (EBV) and herpessimplex-virus (HSV) glycoprotein B (gB), and varicella-herpes-zoster-virus (VHZV) glycoprotein 11, whose role in eliciting a host-immune response has been clearly shown [lo, 17-191. Since cross-reactions of antisera directed against these proteins have been described [20], this feature must be considered for a detailed analysis of HCMV gp58 antigenic properties. These studies are significant, since it has been recently shown that the majority of neutralizing activity in convalescent-human serum is directed against gCI [21], and the isolated complex is also able to induce both humoral and cellular-immune responses in human volunteers [22].In previous studies a HCMV gp58 linear antigenic site, recogniz...

Antigenic domain 1 of human cytomegalovirus glycoprotein B induces a multitude of different antibodies which, when combined, results in incomplete virus neutralization

“…In order to quantify neutralizing and non-neutralizing AD-1-specific antibodies in human sera, reagents had to be produced which allowed separate analysis of the respective antibody type. Fine specificity of AD-1-binding antibodies, however, cannot be analysed by use of short (10–40 aa) synthetic peptides since the entire primary amino acid sequence between residues 552 and 635 of gB is necessary for antibody binding (Wagner et al , 1992 ; Bonci et al , 1993 ; Ohlin et al , 1993 ; Schoppel et al , 1996 ). Therefore, to investigate the binding requirements of different types of antibodies for AD-1 in more detail, a number of mutant plasmids were constructed resulting in point mutations within the primary amino acid sequence.…”

“…Antibody binding requires the presence of the entire AD-1 sequence. Attempts to define conventional linear epitopes within AD-1 using synthetic peptides of various length and monoclonal antibodies (MAbs) as well as human sera have been uniformly unsuccessful (Wagner et al , 1992 ; Bonci et al , 1993 ; Ohlin et al , 1993 ). Point mutations within AD-1, most notably those of the two internal cysteines or prolines, eliminate the antibody-binding properties of the entire polypeptide, indicating that the domain is recognized as a single antigenic determinant.…”