The functional properties of 125I-labeled antibodies and antigens adsorbed on polystyrene and silicone were compared to their counterparts immobilized by non-adsorptive methods. Less than 20% of polyclonal (pAb) and 1-2% of monoclonal (mAb) capture antibody equivalents remained functional after adsorption as a monolayer. Survivability circa doubled or was totally rescued, when the same antibodies were immobilized via a streptavidin bridge or by using a first stage polyclonal antiglobulin capture antibody, respectively. Similarly, the antigenicity of bovine IgGs for pAb and mAb anti-IgGs was highest when the IgGs were immobilized via a streptavidin bridge or when secondarily adsorbed to an albumin monolayer. IgGs in these configurations were significantly more antigenic than when directly adsorbed on polystyrene or a silicone elastomer. Similar activity was seen after adsorption on polystyrene or silicone. Interestingly, these IgGs were equally antigenic when denatured and subsequently adsorbed in 6M guanidine-HCl versus adsorption in PBS without prior denaturation. Although many of the above finding on antibodies and antigens could be explained by the greater accessibility of antigenic epitopes or antibody binding sites when molecules are immobilized by some type of underlying molecular layer, we also show that certain mAb and pAbs preferentially recognized allotopes on IgG2a when IgG2a was adsorbed. Furthermore, such antigenicity was highest when IgG2a was adsorbed at low, sub-monolayer concentrations. Finally, we show that differences in antigenicity need not be related to the method of immobilization, but can also result from differences in the microenvironment of the epitope. This was demonstrated using a filamentous phage clone specific for fluorescein (FLU). This clone recognizes the fluorescein hapten differently depending on the carrier protein used and the method of conjugation. Data presented in this report indicate that antibodies and antigens adsorbed on hydrophobic polymers undergo changes in their functional properties. Data suggest that both changes in conformation and the accessibility of antigen epitopes or antibody binding sites, most likely occur. Especially in the case of the latter, the functional concentration may be 1-2 orders of magnitude lower than the antibody protein concentration. These observations have implications for immunodiagnostics and emphasize the need to determine the specificity of an antibody in the assay in which it is employed and to make no assumptions about the behavior of solid-phase antigens and antibodies from their behavior in solution. Our studies are also relevant to the use of silicone medical prostheses. The antigenicity of IgGs adsorbed on silicone as a multilayer (secondary layer) is much higher than when directly adsorbed. Since such surfaces would be exposed to very high protein concentrations in vivo, multilayers not a monolayer, would be expected. Thus it would seem from these studies that host protein adsorbed on silicone would be expressed to the immune system at the surface of multilayers. This being the case, it seems unlikely that the adsorption of host protein in vivo would generate new epitopes against which the host's immune system could respond and subsequently initiate an autoimmune syndrome.