As more recombinant human proteins become available on the market, the incidence of immunogenicity problems is rising. The antibodies formed against a therapeutic protein can result in serious clinical effects, such as loss of efficacy and neutralization of the endogenous protein with essential biological functions. Here we review the literature on the relations between the immunogenicity of the therapeutic proteins and their structural properties. The mechanisms by which protein therapeutics can induce antibodies as well as the models used to study immunogenicity are discussed. Examples of how the chemical structure (including amino acid sequence, glycosylation, and pegylation) can influence the incidence and level of antibody formation are given. Moreover, it is shown that physical degradation (especially aggregation) of the proteins as well as chemical decomposition (e.g., oxidation) may enhance the immune response. To what extent the presence of degradation products in protein formulations influences their immunogenicity still needs further investigation. Immunization of transgenic animals, tolerant for the human protein, with well-defined, artificially prepared degradation products of therapeutic proteins may shed more light on the structure-immunogenicity relationships of recombinant human proteins.

No abstract

Chapter 6 96 AbstractPurpose. To study the influence of protein structure on the immunogenicity in wildtype and immune tolerant mice of well-characterized degradation products of recombinant human interferon alpha2b (rhIFNα2b).Methods. RhIFNα2b was degraded by metal catalyzed oxidation (M), crosslinking with glutaraldehyde (G), oxidation with hydrogen peroxide (H) and incubation in a boiling water bath (B). The products were characterized with UV absorption, circular dichroism and fluorescence spectroscopy, gel permeation chromatography, reversed-phase HPLC, SDS-PAGE, Western blotting and mass spectrometry. The immunogenicity of the products was evaluated in wildtype mice and in transgenic mice immune tolerant for hIFNα2. Serum antibodies were detected by ELISA or surface plasmon resonance.Results. M-rhIFNα2b contained covalently aggregated rhIFNα2b with three methionines partly oxidized to methionine sulfoxides. G-rhIFNα2b contained covalent aggregates and did not show changes in secondary structure. H-rhIFNα2b was only chemically changed with four partly oxidized methionines. B-rhIFNα2b was largely unfolded and heavily aggregated. Native (N) rhIFNα2b was immunogenic in the wildtype mice but not in the transgenic mice, showing that the latter were immune tolerant for rhIFNα2b. The antirhIFNα2b antibody levels in the wildtype mice depended on the degradation product: M-rhIFNα2b > H-rhIFNα2b ~ N-rhIFNα2b >> B-rhIFNα2b; G-rhIFNα2b did not induce anti-rhIFNα2b antibodies. In the transgenic mice, only M-rhIFNα2b could break the immune tolerance.Conclusions. RhIFNα2b immunogenicity is related to its structural integrity. Moreover, the immunogenicity of aggregated rhIFNα2b depends on the structure and orientation of the constituent protein molecules and/or on the aggregate size. Structural characterization and immunogenicity of rhIFNα2b97