Development of an in vitro elisa assay for the quantification of the immunogenic glycoprotein g present in vaccine batches: comparaison of in vivo and in vitro vaccine potency tests

Abstract

Since several years, there is a global tendency towards limiting and even sometimes waiving the use of animal experiments in Research and Production. However for Human and Veterinary rabies vaccine producers and controllers, the in vivo NIH assay still remains the standard routine potency test before batch release. Nevertheless, fundamental studies have been accumulated to correlate the structural presentation of the main rabies antigen, the glycoprotein G, and its immunogenicity. The G protein contains two main antigenic sites: site II requires a folding of the G ectodomain to bring in proximity peptides distant in the primary sequence of the protein; site III which is less dispersed along the ectodomain but also requires a folded conformation of the ectodomain. Both sites are recognized by specific monoclonal antibodies (mAbs) that are in general neutralizing viral infection. Among them mAb D1 (IgG I isotype), directed against site III is specific of the trimeric state of the glycoprotein (it recognizes the native but not mercapto-ethanol and/or SDStreated G) which is presumed to be the most immunogenic form of the antigen. mAb D1 has been extensively used to evaluate the stability of G trimers (Jallet et al., 1999, J. Virol, 73: 225-33; Desmezieres et al., 2003, Virus Res. 91: 181-7; Sissoeff et al, 2005, J. Gen. Virol, 86: 2543-52). It has also been proven suitable in ELISA to monitor the consistency of the lot to lot rabies vaccine production and to evaluate the glycoprotein content (Fournier-Caruana et al, 2003, Biologicals, 31 : 9-16). Since the end of the 90t’s, the French National Regulatory Authority in charge of human rabies vaccine control (ANSM) has decided to use this ELISA test instead of the single radial immuno-diffusion assay (SRD) to monitor the consistency of production of rabies vaccines. The ELISA has been transferred from Pasteur Institute to ANSM, improved, optimised and then validated. Since 2001, the consistency of production has been established on around 1000 batches while comparing the NIH assay and the ELISA test. The results are homogenous between both methods. In the perspective of replacing in vivo by in vitro assays, vaccine samples have been artificially altered by heating and the evaluation of the glycoprotein content was assayed by mAb-D1 ELISA. This assay was shown to be sensitive enough to detect vaccine alterations and to discriminate between low and high-potent potency batches. This type of ELISA assay may have a promising future for waiving in vivo rabies potency test and promote in vitro antigenic/immunogenic quantification/qualification of the G protein for vaccine batches.