Repertoire of antibodies against type 1 poliovirus in human sera

Abstract

A blocking-ELISA procedure was used to quantify antibodies in sera of humans immunized with poliovirus vaccines. Titers determined by this method demonstrated an excellent correlation with the results of neutralization test. Testing of serum potency with a panel of type 1 poliovirus strains altered antigenically was used to evaluate the composition of polyclonal sera with respect to the epitope specificity of constituent antibodies. Paratope profiles of various polyclonal sera determined by this new method differed, depending on the type of vaccine used for immunization. Antibodies induced in response to inactivated poliovirus vaccine (IPV) contained antibodies directed primarily against antigenic site 1, while sera from recipients of the oral poliovirus vaccine (OPV) contained antibodies to site 3. Antibodies to antigenic sites 2 and 4 were minor constituents in both types of sera. Pre-immunization sera had paratope profiles similar to OPV-induced antisera, allowing the discrimination between antibodies induced by IPV and maternal antibodies. The new method may be useful for analyzing results of clinical trials and to compare immunity induced by different poliovirus vaccines.

Introduction

Poliomyelitis is caused by polioviruses belonging to one of three serotypes, each defined by failure of neutralization by antisera to either of the other two types. Both live attenuated poliovirus and inactivated trivalent poliovirus vaccines induce strong antibody responses to all three serotypes and are highly effective in protecting vaccines against poliomyelitis. Determination of the three-dimensional structure of poliovirus virions (Hogle et al., 1985) and the introduction of monoclonal antibody (mAb) technology (Brioen et al., 1982, Emini et al., 1982, Emini et al., 1983, Evans et al., 1983, Ferguson et al., 1984) were major advances in the understanding of poliovirus structure and antigenicity. Growth of poliovirus in the presence of a mAb induced one or more mutations that rendered the virus resistant to neutralization by the same antibody. Sequencing of antibody-resistant (escape) mutants revealed location of sites within the poliovirus capsid (epitopes) that are critical for binding neutralizing antibodies. Such epitopes are clustered in several areas on the virion surface called antigenic sites. Most neutralizing epitopes of poliovirus are conformational, meaning that disruption of the conformation of native virion destroys both their binding to antibodies that react with native virus and their ability to induce a neutralizing immune response (Blondel et al., 1986, Diamond et al., 1985, Minor, 1990, Minor et al., 1983, Minor et al., 1986a, Minor et al., 1986b). These findings are consistent with the fact that many epitopes are located at the interfaces between different polypeptide chains on the virion surface. For instance, a single mAb can induce specific mutations in different polypeptide chains, because in most cases the affected amino acids are located close to each other in the native 3d-structure (Page et al., 1988).

The role of individual epitopes and antigenic sites in antigenicity and protective properties of poliovirus vaccines in humans remains unknown. The ease with which antibody escape mutants can be generated in vitro suggests that to offer reliable protection, polyclonal sera should contain a spectrum of antibodies of different specificities, both to maximize their ability to neutralize the virus and to prevent antigenic drift. Therefore, vaccines should contain multiple poliovirus epitopes that induce different protective antibodies. Traditionally, the potency of an inactivated poliovirus vaccine (IPV) is determined by ELISA with a well-characterized polyclonal antiserum (Moynihan et al., 1980, Souvras et al., 1980, Wood et al., 1995). This procedure ignores possible variations in epitope composition of the product and depends on the quality of polyclonal antisera used in the potency assays. Some ELISA protocols for potency testing use mAbs (Sawyer et al., 1993, Singer et al., 1989), but these do not allow direct evaluation of the relative contributions of each antigenic site. Recently a method based on a blocking-ELISA protocol using mAbs was developed in this laboratory. That method allowed quantitation of the contribution of individual epitopes into the overall antigenicity of both IPV and strains of live poliovirus (Rezapkin et al., 2005b). It showed that different virus strains have different profiles of reactivity with mAbs and that the process of virus inactivation with formalin destroys selectively some epitopes. As a result, conventional IPV produced from wild strains of poliovirus and Sabin IPV made from attenuated strains had different epitope profiles. The practical significance of these epitope differences remains unclear, but epitope analysis might prove useful for monitoring the consistency of vaccine manufacture and to guide development of effective alternative vaccines.

One approach to revealing potential roles of individual epitopes in protective properties of polio vaccines would be to analyze the antibody composition of polyclonal antisera with respect to their epitope specificities. If the spectra of antibodies, or “paratope profiles” of polyclonal antisera are characterized quantitatively, then their role in protection might be clarified. This information could also be used to compare different vaccines. All existing information on poliovirus epitopes has been generated using mouse mAbs. Some circumstantial evidence derived from analysis of antigenically modified vaccine-derived polioviruses (VDPV) suggests that some antigenic sites identified using a mouse system may not be active in humans (Yakovenko et al., 2006). Therefore, it is also important to determine whether the human immune system recognizes the same sites in poliovirus capsid as in the case in mice, and whether live and inactivated polio vaccines induce the same spectrum of antibodies.

This paper describes attempts to develop a method for characterizing the antibody repertoire of human polyclonal sera. Methods based on competition between mAbs and polyclonal sera have failed to discriminate the contribution of individual paratopes. Therefore, a method using a panel of mutant poliovirus strains with epitopes altered selectively was created to characterize the contributions of respective antibodies to ELISA reactivity and virus neutralization. The results demonstrate that antibody profiles induced in humans by IPV and by live attenuated oral polio vaccine (OPV) are significantly different. Those differences permitted easy discrimination between maternally derived OPV-induced antibodies and antibodies induced in infants by vaccination with IPV.